1 may include a drum 40 provided (or disposed) inside the tub 30. The drum 40 may have a substantially cylindrical shape with one side open. In the front surface of the drum 40, a drum opening may be provided at a position corresponding to the opening of the housing 10 and the tub opening 31 of the tub 30. The drum 40 may receive laundry. The drum 40 may receive rotational power from a driver 60 and rotate inside the tub 30 with respect to the tub 30. The drum 40 may perform washing, rinsing, and/or spinning while rotating inside the tub 30.

According to one or more embodiments, the drum 40 may include a lifter 41 and/or a plurality of through holes 42. For example, the lifter 41 may lift the laundry while the drum 40 rotates so that the laundry repeatedly rises and falls, thereby evenly washing laundry on several surfaces thereof. The through hole 42 may be, e.g., a passage formed so that the washing water received in the tub 30 flows into the drum 40 or the washing water inside the drum 40 is discharged to the outside. In an example, the lifter 41 or the through hole 42 may be omitted.

According to one or more embodiments, the washer 1 may include a control panel 50 that supports interaction between the user and the washer 1. In an example, the control panel 50 may be disposed at an upper end of the front surface of the housing 10 as illustrated in FIG. 1A, but the disclosure is not limited thereto. In an example, the control panel 50 may include an input unit 51 and a display unit 52.

According to one or more embodiments, the input unit 51 may include, e.g., any type of user input means for obtaining a user input for controlling the washer 1. The user may input power on/off, washing setting information (e.g., operation start/stop, course selection, time selection, etc.) of the washer 1 through the input unit 51. For example, the input unit 51 may be a tact switch, a push switch, a slide switch, a toggle switch, a micro switch, or a touch switch, but the disclosure is not limited thereto. For example, the input unit 51 may be in the form of a jog shuttle that the user may grip and rotate. In an example, the input unit 51 may include an infrared sensor. The user may remotely input the setting information through the remote control, and the input setting information may be received by the input unit 51 as an infrared signal. In an example, the input unit 51 may include a microphone. Setting information by the user's voice may be obtained through a microphone.

According to one or more embodiments, The display unit 52 may display various washing setting information input from the user and/or operation state information about the washer 1. The display unit 52 may include various types of display panels such as an LCD, an LED, an OLED, a QLED, and a micro LED. For example, the display unit 52 may be implemented as a touch screen with a touch pad provided on the front surface thereof, but the disclosure is not limited to a specific type of display means. In an example, the display unit 52 may include any type of audio output means including a speaker, and may output each of the above-described information as an auditory signal through the audio output means. In an example, the display unit 52 may operate to audibly provide the user with information for guiding the user's input and/or information related to the ongoing process.

According to one or more embodiments, the washer 1 may include a driver 60 for rotating the drum 40. The driver 60 may include a motor 61 and a driving shaft 62 for transferring the driving force generated by the motor 61 to the drum 40. The motor 61 may include a fixed stator 611 and a rotor 612 that rotates by electromagnetically interacting with the stator 611 to convert an electric force into a mechanical rotational force. The rotational force generated by the motor 61 may be transferred to the drum 40 through the driving shaft 62. The driving shaft 62 may be press-fitted into the rotor 612 of the motor 61 to rotate together with the rotor 612. The driving shaft 62 may, e.g., partially penetrate the rear wall of the tub 30 to connect the drum 40 and the motor 61. The driver 60 may rotate the drum 40 forward or backward to perform washing, rinsing, and/or spinning operations.

According to one or more embodiments, the washer 1 may include a water supply device 70 for supplying washing water to the drum 40 and/or the tub 30. The water supply device 70 may include at least one water supply pipe 71 and at least one water supply valve 72. The at least one water supply pipe 71 may be provided to supply washing water into the tub 30 using an external water supply source. One of the at least one water supply pipe 71 may be connected to a detergent supply device 13 provided in the housing 10. Here, the detergent supply device 13 may be divided into a plurality of spaces, and each space may be provided with a detergent, a rinsing agent, or the like. The washing water passing through the detergent supply device 13 may be supplied to the tub 30 together with the detergent (or rinsing agent) through the detergent supply pipe 131. Another one of the at least one water supply pipe 71 may be directly connected to the tub 30. For example, the washing water supplied through the water supply pipe 71 directly connected to the tub 30 may be directly supplied to the tub 30 without going through an intermediate component such as the detergent supply device 13.

According to one or more embodiments, the washer 1 may include a drain device 80 for draining the washing water received in the drum 40 and/or the tub 30. The drain device 80 may include a drain valve 81, a first drain pipe 82, a second drain pipe 83, or a pump chamber 84. The drain device 80 may be disposed, e.g., under the tub 30 to discharge the washing water discharged from the tub 30 to the outside of the washer 1.

According to one or more embodiments, the drain valve 81 may be provided to open and close the drain port 32. When the drain valve 81 is opened, the washing water received in the tub 30 may flow through the drain port 32 to the drain device 80.

According to one or more embodiments, the first drain pipe 82 and the second drain pipe 83 may form a flow path that guides washing water to be discharged to the outside. For convenience of description, the upper stream of the pump chamber 84 is referred to as the first drain pipe 82 and the lower stream is referred to as the second drain pipe 83. The first drain pipe 82 and the second drain pipe 83 may be integrally formed. The first drain pipe 82 may have, e.g., one end connected to the drain port 32 and the other end connected to the pump chamber 84. The washing water may move into the pump chamber 84 along the first drain pipe 82. The second drain pipe 83 may have, e.g., one end connected to the pump chamber 84 and the other end connected to the outside of the washer 1. Accordingly, the washing water passing through the pump chamber 84 may be discharged to the outside of the washer 1 along the second drain pipe 83.

According to one or more embodiments, the pump chamber 84 may be provided under the tub 30 to store washing water drained from the tub 30. Inside the pump chamber 84, e.g., a drain pump 841 for discharging the stored washing water to the outside may be provided. The washing water pumped by the drain pump 841 may be guided to the outside of the housing 10 through the second drain pipe 83.

FIG. 1C is a block diagram illustrating an example of a configuration of a washer according to one or more embodiments of the disclosure.

According to one or more embodiments, the washer 1 may include an input/output device 130 (e.g., the input unit 51 of FIG. 1A). The input/output device 130 may include any type of user input means for obtaining setting information from the user for controlling the operation of the washer 1. Various user inputs obtained through the input/output device 130 may be transferred to the processor 110 to be described below. In an example, various user inputs obtained through the input/output device 130 may be transmitted to the outside through the communication device 160 to be described below, but the disclosure is not limited thereto.

According to one or more embodiments, the washer 1 may include a communication device 160 that supports signal transmission/reception to/from the outside. In an example, the communication device 160 may include a communication circuit and may receive and/or transmit a wired/wireless signal to/from an external wired/wireless communication system, an external server, and/or other devices according to a predetermined wired/wireless communication protocol. In an example, the communication device 160 may include one or more modules to connect the washer 1 to one or more networks. In an example, the communication device 160 may include at least one of a mobile communication module, a wired/wireless Internet module, a short-range communication module, and/or a location information module.

According to one or more embodiments, the mobile communication module may transmit/receive wireless signals with at least one of an external base station, an external UE, and an external server through the mobile communication network according to any communication protocol among various communication protocols for mobile communication. The wireless signals may include various types of data signals. In an example, the wireless signals may include voice call signals, video call signals, and text/multimedia message signals, but the disclosure is not limited thereto.

According to one or more embodiments, the wired/wireless Internet module may support wireless LAN (WLAN), wireless-fidelity (Wi-Fi), Wi-Fi direct, digital living network alliance (DLNA), wireless broadband (WiBro), world interoperability for microwave access (WiMAX), high speed downlink packet access (HSDPA), high speed uplink packet access (HSUPA), long term evolution (LTE), or long term evolution-advanced (LTE-A), but is not limited thereto. In an example, the wired/wireless Internet module of the communication device 160 may transmit/receive data according to at least one wired/wireless Internet technology among Internet technologies not listed above.

According to one or more embodiments, the short-range communication module may be intended for, e.g., short-range communication and may support short-range communication using at least one of Bluetooth, radio frequency identification (RFID), infrared data association (IrDA), ultra-wideband (UWB), ZigBee, near-field communication (NFC), Wi-Fi, Wi-Fi Direct, or wireless universal serial bus (USB) technology. The short-range communication module may support, e.g., wireless communication between the washer 1 and a wireless communication system, between the washer 1 and another device, or between the washer 1 and a network in which the other device is positioned through a short-range wireless communication network.

According to one or more embodiments, the location information module may be, e.g., a global positioning system (GPS) module or a Wi-Fi module as a module for obtaining the location of the washer 1. When the washer 1 utilizes the GPS module, the washer 1 may receive information about the location of the washer 1 using the signal transmitted from the GPS satellite. When the washer 1 utilizes the Wi-Fi module, the washer 1 may receive information about the location of the washer 1 based on information about a wireless access point (AP) that transmits and receives a wireless signal to and from the Wi-Fi module.

According to one or more embodiments, the communication device 160 may receive the configuration (e.g., setting) data signal input by the user on the mobile terminal of the user in the form of a wireless signal according to a predetermined wireless communication protocol. In an example, the communication device 160 may receive information and/or a command for controlling the operation of the washer 1 from an external server in the form of a signal according to a predetermined wired/wireless communication protocol. The communication device 160 may transfer various received signals to the processor 110 to be described below. In an example, the communication device 160 may transmit various data generated or obtained on the washer 1 in the form of a wired/wireless signal according to a predetermined wired/wireless communication protocol, e.g., to a mobile terminal of the user or an external server.

According to one or more embodiments, the washer 1 may include a sensor device 150 for detecting an operating state and/or an internal environment of the washer 1. In an example, the sensor device 150 may include a plurality of vibration sensors 151 and 152, a weight sensor, a spin sensor, a current sensor, a door sensor, a speed sensor, and/or a temperature sensor, but this is illustrative and the disclosure is not limited thereto.

According to one or more embodiments, the plurality of vibration sensors 151 and 152 may include a first vibration sensor 151 and/or a second vibration sensor 152.

According to one or more embodiments, the first vibration sensor 151 may be disposed adjacent to the housing 10 of the washer 1. For example, the first vibration sensor 151 may be attached to one surface (e.g., an inner surface or an outer surface) or one side of the housing 10 (or an outer wall) of the washer 1. The first vibration sensor 151 may be used, e.g., to identify the vibration of the washer 1. For example, the first vibration sensor 151 may be used to identify a resonance (or vibration) associated with the floor (e.g., the state of the floor) when performing the spin cycle. The first vibration sensor 151 may be, e.g., a micro-electro-mechanical systems (MEMS) sensor. The MEMS sensor may be, e.g., an accelerometer or a gyroscope, but is not limited thereto. The data obtained by the first vibration sensor 151 (or MEMS sensor) may include acceleration data on the x-axis, y-axis, and z-axis, or angular velocity data on the x-axis, y-axis, and z-axis. The data obtained by the first vibration sensor 151 (or MEMS sensor) may be a sequence of values (e.g., acceleration values on the x/y/z axis or angular velocity values on the x/y/z axis) sampled according to the sampling frequency.

According to one or more embodiments, the second vibration sensor 152 may be disposed adjacent to the tub 30 of the washer 1. For example, the second vibration sensor 152 may be attached to one surface (e.g., an inner surface or an outer surface) or one side of the tub 30 of the washer 1. The second vibration sensor 152 may be used, e.g., to identify the vibration of the tub 30 caused by laundry. For example, the second vibration sensor 152 may be used to identify a resonance (or vibration) associated with laundry when performing the spin cycle. The resonance associated with the laundry may be, e.g., resonance due to eccentricity of the laundry, distribution of the laundry, an imbalance of the laundry, an imbalanced load of the laundry, or an uneven distribution of the laundry. The second vibration sensor 152 may be, e.g., a MEMS sensor. The MEMS sensor may be, e.g., an accelerometer or a gyroscope, but is not limited thereto. The data obtained by the second vibration sensor 152 may include acceleration data on the x-axis, y-axis, and z-axis, or angular velocity data on the x-axis and y-axis. The data obtained by the second vibration sensor 152 (or MEMS sensor) may be a sequence of values (e.g., acceleration values on the x/y/z axis or angular velocity values on the x/y/z axis) sampled according to the sampling frequency.

According to one or more embodiments, the weight sensor may be used to detect the weight of the laundry. The weight sensor may be a load cell disposed at the lower end of the drum 40, a pressure sensor disposed at the floor of the washer 1 or the drum 40, or a piezo sensor, but is not limited thereto.

According to one or more embodiments, the spin sensor is a sensor provided to detect the water level in the tub 30. In an example, the spin sensor may be provided to identify a spin progress or the like when performing the spin cycle. The spin sensor may transfer an electrical signal regarding the water level in the tub 30 to the processor 110.

According to one or more embodiments, the current sensor may be provided to detect a current flowing through the motor 61 of the driver 60. The electrical signal regarding the current value of the motor 61 generated by the current sensor may be transferred to the processor 110.

According to one or more embodiments, the door sensor may be provided to determine whether the door 20 is closed before the processor 110 performs the washing operation. An electrical signal regarding whether the door 20 is opened or closed generated by the door sensor may be transferred to the processor 110.

According to one or more embodiments, the speed sensor may be provided to detect the rotational speed, the number of revolutions (e.g., revolutions per minute (RPM)), the rotational angle or the rotational direction of the driver 60 (e.g., the motor 61) or the drum 40. In an example, the speed sensor may use, e.g., a method of detecting an on/off signal of the hall sensor adjacent to the position of the rotor while the motor 61 rotates. In an example, the speed sensor may use a method of measuring the magnitude of the current applied to the motor 61 while the drum 40 rotates. The electrical signal regarding the rotation speed, the number of revolutions, the rotation angle, or the rotation direction of the drum 40 generated by the speed sensor may be transferred to the processor 110.

According to one or more embodiments, the temperature sensor may be provided to detect the ambient environment temperature of the washer 1 or the temperature of the internal components, or to detect the temperature of the washing water in the tub 30. The temperature sensor may be implemented as, e.g., a thermistor, which is a type of resistor using the property which the resistance of a material changes according to temperature. The electrical signal regarding the temperature generated by the temperature sensor may be transferred to the processor 110.

According to one or more embodiments, the washer 1 may include a memory 120 for storing or recording a program and/or data for controlling each component of the washer 1, and a processor 110 for generating a control signal for controlling each component of the washer 1 according to the program and/or data stored in the memory 120 and information obtained from each of the other components.

According to one or more embodiments, the processor 110 may execute commands (or instructions) included in a program (or application) stored in the memory 120. The processor 110 may include, e.g., a central processing unit (CPU), a graphic processing unit (GPU), a neural processing unit (NPU), a sensor processing unit (TPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), and/or a programmable logic device, but may include a program (or an instruction or Instructions) are not limited as long as they may be executed.

According to one or more embodiments, the memory 120 may include a volatile memory and/or a non-volatile memory, and may include, e.g., a hard disk storage device, RAM, ROM, and/or flash memory, but this is not limited thereto.

According to one or more embodiments, the memory 120 may include one or more storage media and store various data that may be used to control the operation of each component of the washer 1. The memory 120 may store, e.g., a plurality of application programs used in the washer 1, data for controlling the operation of the washer 1, and instructions. At least some of the application programs stored in the memory 120 may be downloaded from an external server through wireless communication. At least some of the application programs stored in the memory 120 may be stored in the memory 120 from the time of shipment for the basic functions of the washer 1.

According to one or more embodiments, the processor 110 may receive various input/setting information, e.g., power on/off of the washer 1, washer operation setting information (e.g., operation start/stop, course selection, time selection, etc.), or other various control information from the input/output device 130 and/or the communication device 160 described above. The processor 110 may obtain various detection information from the sensor device 150, e.g., information about the vibration of the washer 1 detected by the vibration sensor 151, information about the vibration of the tub 30 detected by the second vibration sensor 152, information about the weight of laundry detected by the weight sensor, information about the water level in the tub 30 detected by the spin sensor, information about the current flowing through the motor 61 detected by the current sensor, information indicating whether the door is opened or closed, detected by the door sensor, and/or information about, e.g., the RPM of the motor 61 or the drum 40 detected by the speed sensor.

According to one or more embodiments, the processor 110 may generate an operation control command for each component of the washer 1, based on various information received from the input/output device 130, the communication device 160, and/or the sensor device 150. In an example, the processor 110 may control each related component to perform at least one of the washing cycle, the rinsing cycle, the spinning cycle, or the drying cycle. The processor 110 may control the operation of, e.g., the driver 60, the water supply device 70, and/or the drain device 80 to control the execution of at least one of the washing operation, the rinsing operation, the spinning operation, or the drying operation. In an example, the processor 110 may rotate the drum 40 by controlling the rotation of the motor 61 of the driver 60. For example, the processor 110 may control the opening and closing of the water supply valve 72 of the water supply device 70 to adjust the washing water supplied to the drum 40 and/or the tub 30. For example, the processor 110 may control the drain valve 81 and/or the drain pump 841 of the drain device 80 to drain the washing water in the drum 40 and/or the tub 30. In an example, the processor 110 may continuously obtain information from the input/output device 130, the communication device 160, and/or the sensor device 150 while performing at least one of the washing cycle, the rinsing cycle, the spinning cycle, or the drying cycle, and may continuously update and control the operation of each component based on the obtained information.

According to one or more embodiments, the processor 121 may generate a command for controlling whether and how to display information through the display 140, based on various types of information received from the input/output device 130, the communication device 160, and/or the sensor device 150.

In this drawings, it is disclosed that the processor 110 is a comprehensive component for controlling all the components included in the washer 1, but the disclosure is not limited thereto. In an example, the washer 1 may be configured to include a plurality of processor 110 components that individually control some of the components of the washer 1.

FIG. 2 is a view illustrating an AI system including a washer according to one or more embodiments of the disclosure.

According to one or more embodiments, the AI system may include a washer 1 (e.g., the washer 1 of FIG. 1A, 1B, or 1C), an AI server 210, an electronic device 220, and/or an external server 230.

According to one or more embodiments, the washer 1 may provide at least one function (e.g., a function of training at least one AI model and/or a function of performing an inference using the trained at least one AI model). The description of FIGS. 1A to 1C may be referenced for a description of the configuration of the washer 1. No duplicate description thereof is presented below.

According to one or more embodiments, the AI server 210 may provide at least one function (e.g., a function of training at least one AI model and/or a function of performing an inference using the trained at least one AI model). The AI server 210 may include a processor 221, a memory 212, and/or a communication device 213. Meanwhile, it is merely exemplary that the AI server 210 includes the processor 221, the memory 212, and/or the communication device 213, and at least some of the operations of the AI server 210 may be implemented by a cloud server. It will be understood by one of ordinary skill in the art that the AI server 210 may be implemented as a distributed server, and the implementation form of the server is not limited.

According to one or more embodiments, the processor 211 may execute commands (or instructions) included in a program (or application) stored in the memory 212. The processor 211 may include, e.g., a CPU, a GPU, an NPU, a TPU, a DSP, an ASIC, and/or a programmable logic device, but is not limited as long as it is a means capable of executing a program (or instructions or commands). The processor 211 may execute a program for providing an AI service. The program for providing an AI service may be stored in the memory 212. According to one or more embodiments, the memory 212 may include a volatile memory and/or a non-volatile memory, and may include, e.g., a hard disk storage device, RAM, ROM, and/or flash memory, but this is not limited thereto. The program for providing an AI service is a program for a server, and may cause, e.g., generation of data for providing an AI service, provision of the generated data, identification of a user input, and/or generation and provision of data for providing an updated AI service based on the identified user input, and may include commands (or instructions) corresponding to at least some of operations performed by the server 210 of the disclosure. The communication device 213 may support establishing a communication channel between the AI server 210, the washer 1, and the electronic device 220 through the network 240 and performing communication through the established communication channel. The communication device 213 may be a device capable of providing a wide area network (e.g., the Internet), but is not limited thereto. The operation performed by the AI server 210 may be performed by, e.g., the processor 211, or may be performed by other hardware under the control of the processor 211. A command (or instruction) causing the AI server 210 to perform an operation may be stored in the memory 212. The processor 211, the memory 212, and/or the communication device 213 may transmit/receive data via the bus (or a communication interface or a network) of the AI server 210.

According to one or more embodiments, the electronic device 220 may perform at least one operation for providing an AI service using data for an AI service. The electronic device 220 may include at least one of a processor 221, a memory 222, an input/output device 223, a display 224, a sensor device 225, a camera 226, or a communication device 227. The processor 221 may include, e.g., a CPU, a GPU, an NPU, a TPU, a DSP, an ASIC, an FPGA, and/or a programmable logic device, but is not limited as long as it is a means capable of executing a program (or instructions or commands). For example, the processor 221 may execute a program for an AI service. The program for an AI service is a program for client and may cause, e.g., reception of data from an AI service from the AI server 210, provision of at least one operation for providing an AI service based on the received data, identification of a user input and/or transmission of a user input (or a command corresponding to the user input) to the AI server 210, and may include commands (or instructions) corresponding to at least some of the operations performed by the electronic device 220 of the disclosure. According to one or more embodiments, the memory 222 may include a volatile memory and/or a non-volatile memory, and may include, e.g., a hard disk storage device, RAM, ROM, and/or flash memory, but this is not limited thereto. According to one or more embodiments, the input/output device 223 may include a touch pad, a button, a mouse, a digital pen, and/or a microphone, but is not limited as long as it is a device for receiving (or sensing) a user input. For example, the touch screen panel, which is one example of the input/output device 223, may be implemented integrally with the display 224. The input/output device 223 may include a speaker, a haptic module, and/or a light emitting module, but is not limited as long as it is a device for outputting content related to an AI service. According to one or more embodiments, the sensor device 225 may include a gesture sensor, a gyro sensor, an atmospheric pressure sensor, a magnetic sensor, an acceleration sensor, a grip sensor, a proximity sensor, a color sensor, an infrared (IR) sensor, a biometric sensor, a temperature sensor, a humidity sensor, and/or an illuminance sensor. According to one or more embodiments, the camera 226 may include one or more lenses, image sensors, image signal processors, or flashes. According to one or more embodiments, the communication device 227 may support establishing a communication channel between the AI server 210, the washer 1, and the electronic device 220 through the network 240 and performing communication through the established communication channel. The communication device 227 may be a device capable of providing a wide area network (e.g., the Internet), but is not limited thereto. The communication device 227 may support wired communication and/or wireless communication. For example, the communication device 227 may support short-range communication (e.g., short-range communication such as Bluetooth, wireless fidelity (Wi-Fi) direct, or infrared data association (IrDA)). The communication device 227 may transmit and receive data to and from the washer 1 based on short-range communication. For example, when the electronic device 220 is implemented as a standalone type, the communication device 227 may support a function of wirelessly accessing the network 240. The communication device 227 may support cellular communication such as LTE, 5G, or 6G, and/or IEEE 802 series-based communication (e.g., may be referred to as Wi-Fi). The communication device 227 may be implemented to support wired communication, and the implementation method thereof is not limited. When the electronic device 220 is implemented as a non-standalone type, the electronic device 220 may communicate with the washer 1 and the AI server 210 through a relay device connectable to the network 240. In this case, the communication device 227 may support short-range communication such as Bluetooth, wireless fidelity (Wi-Fi) direct, or infrared data association (IrDA), and may perform communication with the AI server 210 and washer 1 through the relay device using the short-range communication. The operation performed by the electronic device 220 may be performed by, e.g., the processor 221, or may be performed by other hardware devices under the control of the processor 221. An command (or instruction) causing the electronic device 220 to perform an operation may be stored in the memory 222. The processor 221, the memory 222, the input/output device 223, the display 224, the sensor device 225, the camera 226, and/or the communication device 227 may transmit/receive data through the bus (or a communication interface or a network) of the electronic device 220. Meanwhile, it is merely exemplary that the washer 1, the AI server 210 and the electronic device 220 transmit and receive data based on an application for an AI service, and those skilled in the art will understand that the server 210 and the electronic device 220 may transmit and receive at least some data based on the web.

Hereinafter, e.g., a method in which a washer (e.g., the washer 1 of FIG. 1A, 1B, 1C, or 2) detects and avoids resonance (or vibration) associated with the floor where the washer is installed during a cycle (e.g., a spin cycle) is described. In the disclosure, the resonance associated with the floor may include, e.g., a resonance associated with the state of the floor (floor state) (resonance by the floor state), but is not limited thereto, and the resonance due to various causes not associated with the laundry may be included as the resonance associated with the floor. The resonance associated with the laundry may include, e.g., resonance due to eccentricity of the laundry or distribution of the laundry, an imbalance of the laundry, an imbalanced load of the laundry, or an uneven distribution of the laundry. In the disclosure, the resonance associated with the floor (or the resonance not associated with the laundry) may be referred to as a first type resonance, and the resonance associated with the laundry may be referred to as a second type resonance. Hereinafter, for convenience of description, the resonance associated with the state of the floor is described as an example of the resonance associated with the floor, and the resonance caused by the eccentricity of the laundry is described as an example of the resonance associated with the laundry, but the disclosure is not limited thereto.

The washer according to various embodiments of the disclosure may be an example of a clothing treatment device. According to one or more embodiments, the washer may include a top-loading washer in which an inlet for inserting or withdrawing laundry is provided to face upward, or a front-loading washer in which an inlet for inserting or withdrawing laundry is provided to face forward. According to one or more embodiments, the washer may include a washing/drying machine that provides both a washing function and a drying function.

FIG. 3 is a view illustrating an operation of initially setting and updating a driving profile of a washer according to one or more embodiments of the disclosure.

In the following embodiments, each operation may be performed sequentially, but is not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In the following embodiment, some of the operations may be omitted (e.g., operation 320 of FIG. 3 is omitted) or an additional operation may be further performed.

According to one or more embodiments, the driving profile of the washer (e.g., the washer 1 of FIG. 1A, 1B, 1C, or 2) may be a profile including at least one piece of information for driving the washer. For example, the driving profile may include information associated with driving configuration information used to perform at least one operation of the washer. The information associated with the driving configuration information may include, e.g., driving configuration information itself used to perform at least one operation of the washer, or information used to obtain the corresponding driving configuration information. The at least one cycle may include, e.g., a washing cycle, a rinsing cycle, a spin cycle, and/or a drying cycle, but is not limited thereto. In the disclosure, the driving profile may be abbreviated as a profile.

According to one or more embodiments, the driving configuration information may include, e.g., configuration information for each of at least one cycle. The configuration information for each cycle may include, but is not limited to, information related to the execution period of the corresponding cycle and/or information related to the revolutions per minute (RPM) of a drum (e.g., the drum 40 of FIG. 1B) or a driver (e.g., the driver 60 or the motor 61 of FIG. 1B) for performing the corresponding cycle. For example, the configuration information for each cycle may further include configuration information for other parameter(s) (e.g., a power consumption-related parameter) used to perform the corresponding cycle. The information related to each execution period may include, e.g., information indicating a setting value of a period for performing the corresponding cycle. For example, the information related to the execution period for the spin cycle may include information indicating a setting value of the period for performing the spin cycle. Each RPM-related information may include, e.g., information indicating a setting value of RPM of a drum or a driver (or a motor) used to perform the corresponding cycle, and/or a setting value of a duration of performing the corresponding cycle at the corresponding RPM. For example, the RPM-related information about the spin cycle may include information indicating a setting value of the spin RPM (e.g., final spin RPM) of the drum or driver (or motor) used to perform the spin cycle and/or information indicating a setting value of the duration of performing the spin cycle at the spin RPM (e.g., final spin RPM). The final spin RPM may be, e.g., the final RPM of the high-speed spin period of the spin cycle, and may be used for a duration of the final spin RPM to finish spin of the laundry.

According to one or more embodiments, a plurality of driving profiles may be set. For example, the plurality of driving profiles may be set in advance through statistical analysis of a residential environment. Statistical analysis of the residential environment may include, e.g., statistical analysis of differences in the residential environment. The difference in the residential environment may include, e.g., a difference in the floor state (e.g., a soft floor, a hard floor, etc.) of the residential area, a difference in the weight of the fabric according to the seasons and weather of the residential area (e.g., weather with four distinct seasons, weather with mild temperatures throughout four seasons, weather with clear distinction between winter and summer, etc.), and/or a difference in the urban form (e.g., an apartment buildings-dense area, a residential area, etc.), but is not limited thereto.

According to one or more embodiments, the state of the floor may be divided into a soft floor and a hard floor according to a designated criterion. The designated criterion may be, e.g., a designated strength (e.g., a compressive strength (e.g., a pound per square inch (PSI)) or a designated hardness (e.g., a shore hardness)). For example, the soft floor may represent, e.g., a floor having a strength (e.g., compressive strength (e.g., PSI)) or hardness (e.g., shore hardness) equal to or less than a designated strength (or hardness). For example, the soft floor may include wood floor and/or a wood floor having a designated strength (or hardness) or less, but is not limited thereto. The hard floor may represent, e.g., a floor in which the strength (e.g., compressive strength (e.g., PSI)) or hardness (e.g., shore hardness) of the floor exceeds a designated strength (or hardness). For example, the hard floor may include, but is not limited to, a concrete floor, a cement floor, or a wood floor exceeding a designated strength (or hardness). The soft floor may be generally more vulnerable to resonance by the floor than the hard floor. Accordingly, in the case of a soft floor, a driving profile that prioritizes vibration noise reduction (a vibration noise reduction profile) may be set as an initial driving profile, and in the case of a hard floor, a driving profile that prioritizes performance (a high performance profile) may be set as an initial driving profile. However, considering the use state, the initial driving profile may be adjusted in a designated manner. According to one or more embodiments, the spin RPM (e.g., the final spin RPM) corresponding to the high performance profile may be set to be higher than the spin RPM (e.g., the final spin RPM) corresponding to the vibration noise reduction profile.

According to one or more embodiments, the plurality of driving profiles may be preset (or defined) by a server (e.g., the AI server 210 or the external server 230 of FIG. 2) having a big data analysis function. The defined plurality of driving profiles may be transferred from the server to the washer (e.g., the washer 1 of FIG. 2).

Table 1 illustrates an example of the plurality of driving profiles. The driving profiles of Table 1 are merely an example, various types of driving profiles may be used in embodiments of the disclosure.

TABLE 1

Residential environment

Driving profile
(conditions)
Characteristics

Driving profile 1
Apartment-buildings dense area
Cement floor-based vibration

in Korea
noise reduction profile

Driving profile 2
Single homes area in Korea
Cement floor-based high

performance profile

Driving profile 3
Apartment buildings-dense area
Wood floor-based vibration

in the U.S.
noise reduction profile

Driving profile 4
Single homes area in the U.S.
Wood floor-based vibration

noise reduction-performance

trade-off profile

. . .
. . .
. . .

Driving profile N
Tropical area
Wood floor, non-seasonal

profile

Referring to Table 1, the driving profile 1 is, e.g., a driving profile for a residential environment in an apartment buildings-dense area in Korea having a cement floor (e.g., a hard floor), and may have characteristics of a cement floor-based vibration noise reduction profile that prioritizes vibration noise reduction. The driving profile 2 is, e.g., a driving profile for a residential environment in a single-homes area in Korea having a cement floor, and may have characteristics of a vibration noise reduction profile based on a cement floor that prioritizes performance. The driving profile 3 is, e.g., a driving profile for a residential environment of an apartment in the United States having a wood floor (e.g., a soft floor), and may have characteristics of a vibration noise reduction profile based on a wood floor that prioritizes vibration noise reduction. The driving profile 4 is, e.g., a driving profile for a residential environment in a single-homes area of the United States having a wood floor, and may have characteristics of a wood floor-based profile that compromises between vibration noise reduction and performance. The driving profile N is a driving profile for a residential environment in a tropical area, and may have characteristics of a wood floor and a non-seasonal profile. As such, the plurality of driving profiles may be set considering residential environment characteristics.

According to one or more embodiments, some or all of the information related to the driving configuration information included in each of the plurality of driving profiles may be different from each other. For example, some or all of the information related to the driving configuration information (e.g., the setting value of the spin RPM (e.g., the final spin RPM)) for the spin cycle included in each driving profile may be different. Accordingly, all or some of the driving configuration information obtained based on each driving profile may be different. For example, the setting value of each spin RPM (e.g., the final spin RPM) for the spin cycle obtained based on each driving profile may be different. According to one or more embodiments, the setting value of the spin RPM (e.g., the final spin RPM) corresponding to the first driving profile may be different from the setting value of the spin RPM (e.g., the final spin RPM) corresponding to the second driving profile.

Referring to FIG. 3, in operation 310, the washer (e.g., the washer 1 of FIGS. 1A to 2) may set an initial driving profile. The initial driving profile may be set, e.g., through residential analysis (e.g., residential address, residential environment analysis). The setting of the initial driving profile may be performed, e.g., before a trial operation of the washer is performed, but is not limited thereto.

According to one or more embodiments, the washer may set an initial driving profile based on information received from a server (e.g., the AI server 210 or the external server 230 of FIG. 2). According to one or more embodiments, the server may obtain information related to the user's residential environment, select one of the plurality of driving profiles (e.g., the driving profiles in Table 1) as an initial driving profile through residential analysis based on the information related to the residential environment, and transmit information about the selected initial driving profile to the washer. The washer may set the initial driving profile based on the information about the selected initial driving profile received from the server. The information related to the residential environment may include, e.g., information about the address of the user to install the washer and/or information about the state of the floor where the washer is to be installed, but is not limited thereto. The information related to the residential environment may be transmitted to the server through an electronic device (e.g., the electronic device 220 of FIG. 2) or through the washer.

According to one or more embodiments, the washer may set the initial driving profile based on a user input. For example, the washer may set the initial driving file based on a user input of selecting one of the plurality of driving profiles (e.g., the driving profiles of Table 1) as the initial driving profile. For example, the user input may be received through the electronic device (e.g., the electronic device 220 of FIG. 2) or may be directly input through an input/output device (e.g., the input/output device 130 of FIG. 1C) of the washer.

According to one or more embodiments, the washer may obtain driving configuration information (e.g., initial driving configuration information) based on the set initial driving profile. The washer may perform a trial operation or a normal operation based on the initial driving configuration information corresponding to the obtained initial driving profile.

In operation 320, the washer may adjust and apply the initial driving profile through the trial operation. Adjustment and application of the initial driving profile may include, e.g., adjustment and application of initial driving configuration information corresponding to the initial driving file. The trial operation may be, e.g., an operation performed to identify whether the washer is normally installed. The trial operation may be performed, e.g., by an installer of the washer, may be performed without laundry, and may be performed for a shorter period than the normal operation, but is not limited thereto. For example, the trial operation and the normal operation may be performed using the same method (e.g., a method using an AI model) and/or an algorithm. The normal operation may be, e.g., an actual operation of performing at least one cycle (e.g., washing, rinsing, spinning, and/or drying cycle) of the washer in a state in which laundry is received in the washer after the washer is normally installed through the trial operation.

According to one or more embodiments, the washer may perform the trial operation using the initial driving configuration information obtained based on the initial driving profile obtained through operation 310, and may adjust the initial driving profile based on the information obtained through the trial operation. For example, the washer may maintain the initial driving configuration information (e.g., final spin RPM=950) or may increase or decrease it by a designated value (e.g., 50 RPM value) within an allowable range (e.g., an allowable RPM range (e.g., 200 RPM to 1300 RPM)) based on information obtained by performing the trial operation based on the initial driving configuration information corresponding to the initial driving profile. An example of the operation of performing the trial operation based on initial driving configuration information and an example of information obtained by performing the trial operation are described below with reference to FIGS. 4 to 7.

According to one or more embodiments, the washer may adjust the initial driving profile based on a user input after performing the trial operation using the initial driving configuration information obtained based on the initial driving profile obtained through operation 310. For example, the washer may maintain the initial driving configuration information (e.g., final spin RPM=950) or may increase or decrease it by a designated value (e.g., 50 RPM value) within an allowable range (e.g., an allowable RPM range (e.g., 200 RPM to 1300 RPM)) based on a user input (e.g., a user input of the installer of the washer) for adjusting the initial driving configuration information obtained after performing the trial operation based on the initial driving configuration information corresponding to the initial driving profile. As described above, the user input may be received through the electronic device or may be directly input through the input/output device of the washer. Through the user input, the installer of the washer may directly adjust the initial driving profile (or driving configuration information) considering the result of identifying whether there is an error related to noise/vibration as a result of the trial operation and/or the result of identifying the state of the floor.

According to one or more embodiments, the washer may use the adjusted initial driving profile (or the adjusted initial driving configuration information) as a driving profile (or driving configuration information) for performing a normal operation.

In operation 330, the washer may adjust and apply the driving profile through normal operation. Adjustment and application of the driving profile may include, e.g., adjustment and application of driving configuration information corresponding to the driving profile.

According to one or more embodiments, the washer may perform the normal operation using the driving configuration information (e.g., the adjusted initial driving configuration information) obtained based on the driving profile (e.g., the adjusted initial driving profile) obtained through operation 320, and may adjust the driving profile based on the information obtained through the normal operation. For example, the washer may maintain the driving configuration information (e.g., final spin RPM=1000) or increase or decrease it by a designated value (e.g., 50 RPM value), based on the information obtained by performing the normal operation based on the driving configuration information corresponding to the driving profile. An example of the operation of performing the normal operation based on driving configuration information and an example of information obtained by performing the normal operation are described below with reference to FIGS. 4 to 7.

According to one or more embodiments, the washer may repeatedly perform operation 330 using driving configuration information (or a driving profile) adjusted by performing the previous normal operation as driving configuration information (or a driving profile) for performing the next normal operation. Through the repeated operation, driving configuration information (or driving profile) may be optimized. For example, the final spin RPM for the spin cycle may be optimized.

In the embodiments of FIGS. 4 to 7, for convenience of description, the update of the driving configuration information for the spin cycle is described as an example of the update of the driving configuration information. However, embodiments are not limited thereto, and the description of FIGS. 4 to 7 may be applied to the update of driving configuration information for another operation (e.g., a washing cycle or a drying cycle).

FIG. 4 is a flowchart illustrating a method for updating driving configuration information using an AI model by a washer according to one or more embodiments of the disclosure.

In the following embodiments, each operation may be performed sequentially, but is not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In the following embodiments, some of the operations may be omitted, or an additional operation may be further performed.

According to one or more embodiments, the method of FIG. 4 may be performed by a washer (e.g., the washer 1 of FIG. 1A, 1B, 1C, or 2) using at least one AI model stored in the washer. The at least one AI model may be, e.g., an AI model trained by an external electronic device (e.g., the AI server 210 of FIG. 2) and distributed to the washer, or an AI model trained and stored by the washer itself.

Referring to FIG. 4, in operation 410, the washer may obtain driving configuration information based on the driving profile of the washer. For example, the washer may obtain initial driving configuration information based on an initial driving file (e.g., one driving profile among the driving profiles of Table 1) of the washer. For example, the washer may obtain driving configuration information about the current cycle (e.g., the current spin cycle) based on the driving profile (or driving configuration information) adjusted in the previous cycle (e.g., the previous spin cycle). In the embodiment of FIG. 4, the current cycle may be referred to as a first cycle, and the current spin cycle may be referred to as a first spin cycle.

According to one or more embodiments, the driving configuration information may include all or some of the information included in the driving configuration information described with reference to FIG. 3. For example, the driving configuration information may include RPM-related information about the driver (e.g., the driver 60 or the motor 61 of FIG. 1B) or the drum (e.g., the drum 40 of FIG. 1B) associated with the spin cycle of the washer. The RPM-related information may include, e.g., information indicating a setting value of the final spin RPM to be used in the first spin cycle (hereinafter, referred to as final spin RPM information) and/or information indicating a setting value of the duration of the final spin RPM (hereinafter, referred to as final spin RPM duration information).

According to one or more embodiments, the final spin RPM information and/or the final spin RPM duration information may be essential RPM-related information for performing the spin cycle. This is because the section in which the substantially greatest spin effect occurs on the laundry during the spin cycle corresponds to the high-speed spin section performed based on the final spin RPM information and the final spin RPM duration information.

In operation 420, the washer may obtain RPM data and vibration data while the first spin cycle is performed based on the driving configuration information.

According to one or more embodiments, the RPM data may include RPM values of the driver (or the motor) or the drum obtained while the first spin cycle is performed. According to one or more embodiments, the vibration data may include values related to vibration of the washer obtained through a first sensor (e.g., the first vibration sensor 151 of FIG. 1B) while the first spin cycle is performed.

According to one or more embodiments, the washer may obtain vibration data using the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed on one surface of a housing (e.g., the housing 10 of FIGS. 1A and 1B) of the washer. For example, the washer may obtain vibration data using the first sensor attached to an inner surface or an outer surface of the housing (e.g., the housing 10 of FIGS. 1A and 1B) of the washer. The first sensor may be, e.g., a MEMS sensor. The data obtained by the first sensor (or MEMS sensor) may include acceleration data on the x-axis and y-axis, or angular velocity data on the x-axis and y-axis. The data obtained by the first sensor (or MEMS sensor) may be a sequence of values (e.g., acceleration values on the x-axis/y-axis or angular velocity values on the x-axis/y-axis) sampled according to the sampling frequency (e.g., 10 Hz).

In operation 430, the washer may identify whether resonance associated with the floor (e.g., the floor state) is generated using at least one AI model, based on the RPM data and the vibration data. Resonance may be generated while the spin cycle is performed, and the resonance may be associated with the floor (e.g., the floor state) or with laundry (e.g., eccentricity of laundry). Among them, the resonance associated with the floor may cause unnecessary noise and vibration during the spin cycle, causing inconvenience to the user. Therefore, required is a process to identify (or detect) the resonance associated with the floor (e.g., the floor state) among the resonances generated during the spin cycle, and to prevent such resonance from being generated.

According to one or more embodiments, the washer may input first input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model and identify whether the resonance associated with the floor (e.g., the state of the floor) is generated using a designated algorithm for filtering resonance by laundry (e.g., the eccentricity of the laundry) received in the drum, based on washer data including the RPM data and the vibration data and the information about the state of the floor. According to one or more embodiments, the information about the state of the floor may be set to one of a first value indicating that the floor is a soft floor or a second value indicating that the floor is a hard floor. An example of a method for identifying whether resonance associated with a floor (e.g., a floor state) is generated using a first AI model is described below with reference to FIG. 6A.

According to one or more embodiments, the washer may input first input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model, and input second input data generated based on the information about the state of the floor and washer data including the RPM data and the vibration data to a second AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the second AI model. According to one or more embodiments, the information indicating whether the resonance associated with the state of the floor is generated may be set to one of a first value indicating that the resonance associated with the state of the floor is generated or a second value indicating that the resonance associated with the state of the floor is not generated. An example of a method for identifying whether resonance associated with a floor (e.g., a floor state) is generated using a first AI model and a second AI model is described below with reference to FIG. 6B.

According to one or more embodiments, the washer may input third input data generated based on washer data including the RPM data and the vibration data to a third AI model and obtain information indicating whether the resonance associated with the state (e.g., the state of the floor) of the floor is generated as output data of the third AI model. An example of a method of identifying whether resonance associated with a floor (e.g., a floor state) is generated using a third AI model is described below with reference to FIG. 6C.

In operation 440, the washer may adjust RPM-related information (e.g., the value of the final spin RPM) using the RPM value associated with the resonance, based on identifying that the resonance associated with the floor (e.g., the state of the floor) is generated. According to one or more embodiments, the adjusted RPM-related information may be used to perform a second spin cycle performed after the first spin cycle. For example, the driving configuration information (or the driving profile) including the adjusted RPM-related information may be set as the second driving configuration information (or the second driving profile) for performing the second spin cycle. An example of operation 440 is described below with reference to FIG. 7.

In operation 450, the washer may maintain RPM-related information (e.g., the value of the final spin RPM) based on identifying that the resonance associated with the floor (e.g., the state of the floor) is not generated. According to one or more embodiments, the maintained RPM-related information may be used to perform a second spin cycle performed after the first spin cycle. For example, the driving configuration information (or the driving profile) including the maintained RPM-related information may be set as the second driving configuration information (or the second driving profile) for performing the second spin cycle.

According to one or more embodiments, the washer may perform operations 410 to 450 on the second spin cycle performed after the first spin cycle, based on the second driving configuration information (or the second driving profile). For example, the washer may obtain the second driving configuration information based on the second driving profile of the washer, may obtain RPM data and vibration data while the second spin cycle is performed based on the second driving configuration information, may identify whether a resonance associated with the floor (e.g., the floor state) is generated using at least one AI model based on the RPM data and the vibration data, adjust the RPM-related information (e.g., the value of the final spin RPM) using the RPM value associated with the resonance based on identifying that the resonance associated with the floor (e.g., the state of the floor) is generated, and maintain the RPM-related information (e.g., the value of the final spin RPM) based on identifying that the resonance associated with the floor (e.g., the state of the floor) is not generated. The driving configuration information (or driving profile) including the adjusted RPM-related information or the maintained RPM-related information may be set as the third driving configuration information (or the third driving profile) for performing the third spin cycle performed after the second spin cycle. Thereafter, the washer may perform operations 410 to 450 on the third spin cycle performed after the second spin cycle based on the third driving configuration information (or the third driving profile). Through the repeated operation, driving configuration information including RPM-related information may be continuously adjusted (or updated) for each spin cycle to be optimized, thereby preventing or reducing generation of a resonance associated with the floor (e.g., the floor state).

According to one or more embodiments, the washer may perform operations 410 to 450 in a trial operation process or a normal operation process.

According to one or more embodiments, in the trial operation, the washer may perform the above-described operations 410 to 450 using the initial driving profile (e.g., the initial driving profile obtained through operation 310 of FIG. 3).

According to one or more embodiments, in the normal operation process, the above-described operations 410 to 450 may be performed using the initial driving profile (e.g., the initial driving profile obtained through operation 310 of FIG. 3) or the driving profile obtained through the previous operation (e.g., the driving profile obtained through the operation of the trial operation process or the driving profile obtained through the previous operation of the normal operation process).

FIG. 5A is a flowchart illustrating a method for updating driving configuration information using an AI model by a server according to one or more embodiments of the disclosure.

In the following embodiments, each operation may be performed sequentially, but is not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In the following embodiments, some of the operations may be omitted, or an additional operation may be further performed.

According to one or more embodiments, the method of FIG. 5A may be performed by a server (e.g., the AI server 210 of FIG. 2) using at least one AI model stored in the server. The at least one AI model may be, e.g., an AI model trained by an external electronic device (e.g., the washer 1 of FIG. 2) and distributed to the server, or an AI model trained and stored by the server itself.

Referring to FIG. 5A, in operation 510a, the server may receive, from the washer, RPM data and vibration data obtained while the first spin cycle is performed based on the driving configuration information.

According to one or more embodiments, as illustrated in operation 410 of FIG. 4, the washer may obtain driving configuration information based on the driving profile of the washer. For example, the washer may obtain initial driving configuration information based on an initial driving file (e.g., one driving profile among the driving profiles of Table 1) of the washer. For example, the washer may obtain driving configuration information about the current cycle (e.g., the current spin cycle) based on the driving profile (or driving configuration information) adjusted in the previous cycle (e.g., the previous spin cycle). In the embodiment of FIGS. 5A and 5B, the current cycle may be referred to as a first cycle, and the current spin cycle may be referred to as a first spin cycle.

According to one or more embodiments, as illustrated in operation 420 of FIG. 4, the washer may obtain RPM data and vibration data while the first spin cycle is performed based on the driving configuration information.

According to one or more embodiments, the washer may obtain vibration data using the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed inside or outside a housing (e.g., the housing 10 of FIGS. 1A and 1B) of the washer. For example, the washer may obtain vibration data using a sensor (e.g., a vibration sensor) attached to an inner surface (e.g., an inner wall) or an outer surface (e.g., an outer wall) of a housing (e.g., the housing 10 of FIGS. 1A and 1B) of the washer. The first sensor may be, e.g., a MEMS sensor. The data obtained by the first sensor (or MEMS sensor) may include acceleration data on the x-axis and y-axis, or angular velocity data on the x-axis and y-axis. The data obtained by the first sensor (or MEMS sensor) may be a sequence of values (e.g., acceleration values on the x-axis/y-axis or angular velocity values on the x-axis/y-axis) sampled according to the sampling frequency (e.g., 10 Hz).

In operation 520a, the server may identify whether resonance associated with the floor (e.g., the floor state) is generated using at least one AI model, based on the RPM data and the vibration data. Resonance may be generated while the spin cycle is performed, and the resonance may be associated with the floor (e.g., the floor state) or with laundry (e.g., eccentricity of laundry). Among them, the resonance associated with the floor may cause unnecessary noise and vibration during the spin cycle, causing inconvenience to the user. Therefore, required is a process to identify (or detect) the resonance associated with the floor (e.g., the floor state) among the resonances generated during the spin cycle, and to prevent such resonance from being generated.

According to one or more embodiments, the server may input first input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model and identify whether the resonance associated with the floor (e.g., the state of the floor) is generated using a designated algorithm for filtering resonance by laundry (e.g., the eccentricity of the laundry) received in the drum, based on washer data including the RPM data and the vibration data and the information about the state of the floor. According to one or more embodiments, the information about the state of the floor may be set to one of a first value indicating that the floor is a soft floor or a second value indicating that the floor is a hard floor. An example of a method for identifying whether resonance associated with a floor (e.g., a floor state) is generated using a first AI model is described below with reference to FIG. 6A.

According to one or more embodiments, the server may input first input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model, and input second input data generated based on the information about the state of the floor and washer data including the RPM data and the vibration data to a second AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the second AI model. According to one or more embodiments, the information indicating whether the resonance associated with the state of the floor is generated may be set to one of a first value indicating that the resonance associated with the state of the floor is generated or a second value indicating that the resonance associated with the state of the floor is not generated. An example of a method for identifying whether resonance associated with a floor (e.g., a floor state) is generated using a first AI model and a second AI model is described below with reference to FIG. 6B.

According to one or more embodiments, the server may input third input data generated based on washer data including the RPM data and the vibration data to a third AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the third AI model. An example of a method of identifying whether resonance associated with a floor (e.g., a floor state) is generated using a third AI model is described below with reference to FIG. 6C.

In operation 530a, the server may adjust RPM-related information (e.g., the value of the final spin RPM) using the RPM value associated with the resonance, based on identifying that the resonance associated with the floor (e.g., the state of the floor) is generated, and transmit the adjusted RPM-related information to the washer. According to one or more embodiments, the adjusted RPM-related information may be used by the washer to perform a second spin cycle performed after the first spin cycle. For example, the driving configuration information (or the driving profile) including the adjusted RPM-related information may be set as the second driving configuration information (or the second driving profile) for performing the second spin cycle. An example of operation 530a is described below with reference to FIG. 7.

In operation 540a, the server may maintain RPM-related information (e.g., the value of the final spin RPM) based on identifying that the resonance associated with the floor (e.g., the state of the floor) is not generated. According to one or more embodiments, the server may explicitly or implicitly notify the washer that the RPM-related information is maintained without being adjusted. For example, the server may explicitly notify the washer by transmitting information indicating that the RPM-related information is maintained without being adjusted to the washer. For example, when information indicating that RPM-related information is adjusted is not received from the server for a designated period after transmitting RPM data and vibration data to the server, the washer may implicitly identify that RPM-related information is maintained. According to one or more embodiments, the maintained RPM-related information may be used by the washer to perform a second spin cycle performed after the first spin cycle. For example, the driving configuration information (or the driving profile) including the maintained RPM-related information may be set as the second driving configuration information (or the second driving profile) for performing the second spin cycle.

According to one or more embodiments, the server may perform operations 510a to 540a for the second spin cycle performed after the first spin cycle by receiving RPM data and vibration data obtained while the second spin cycle is performed from the washer based on the second driving configuration information (or the second driving profile). For example, the server may receive, from the washer, RPM data and vibration data obtained while the second spin cycle is performed based on the second driving configuration information, may identify whether a resonance associated with the floor (e.g., the floor state) is generated using at least one AI model based on the RPM data and the vibration data, adjust the RPM-related information (e.g., the value of the final spin RPM) using the RPM value associated with the resonance based on identifying that the resonance associated with the floor (e.g., the state of the floor) is generated and transmit the adjusted RPM-related information to the washer, and maintain the RPM-related information (e.g., the value of the final spin RPM) based on identifying that the resonance associated with the floor (e.g., the state of the floor) is not generated. The driving configuration information (or driving profile) including the adjusted RPM-related information or the maintained RPM-related information may be set by the washer as the third driving configuration information (or the third driving profile) for performing the third spin cycle performed after the second spin cycle. Thereafter, the server may perform operations 510a to 540a for the third spin cycle performed after the second spin cycle by receiving RPM data and vibration data obtained while the third spin cycle is performed from the washer based on the third driving configuration information (or the third driving profile). Through the repeated operation, driving configuration information including RPM-related information may be continuously adjusted (or updated) for each spin cycle to be optimized, thereby preventing or reducing generation of the resonance associated with the floor (e.g., the floor state).

According to one or more embodiments, the server may perform operations 510a to 540a in a trial operation process or a normal operation process.

According to one or more embodiments, in the trial operation process, the server may perform the above-described operations 510a to 540a using RPM data and vibration data obtained by the washer and transmitted to the server during the spin cycle using the initial driving configuration information obtained based on the initial driving profile (e.g., the initial driving profile obtained through operation 310 of FIG. 3).

According to one or more embodiments, in the normal operation process, the above-described operations 510a to 540a may be performed using RPM data and vibration data obtained by the washer and transmitted to the server during the spin cycle using the driving configuration information obtained based on the initial driving profile (e.g., the initial driving profile obtained through operation 310 of FIG. 3) or the driving profile obtained through the previous process (e.g., the driving profile obtained through the process of the trial operation or the driving profile obtained through the previous process of the normal operation).

FIG. 5B is a signal flowchart illustrating a procedure performed by a washer and a server to update driving configuration information according to one or more embodiments of the disclosure.

In the following embodiments, each operation may be performed sequentially, but is not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In the following embodiments, some of the operations may be omitted, or an additional operation may be further performed.

According to one or more embodiments, the method of FIG. 5B may be performed by a server (e.g., the AI server 210 of FIG. 2) using at least one AI model stored in the server. The at least one AI model may be, e.g., an AI model trained by an external electronic device (e.g., the washer 1 of FIG. 2) and distributed to the server, or an AI model trained and stored by the server itself.

Referring to FIG. 5B, in operation 501b, the washer 1 may obtain driving configuration information based on the driving profile of the washer. For the description of operation 501b, the description of operation 410 of FIG. 4 may be referred to. Therefore, no duplicate description is given.

In operation 502b, the washer may obtain RPM data and vibration data while the spin cycle is performed based on the driving configuration information. For the description of operation 502b, the description of operation 420 of FIG. 4 may be referred to. Therefore, no duplicate description is given.

In operation 510b, the washer may transmit the obtained RPM data and vibration data to the server 210. For the description of operation 510b, the description of operation 510a of FIG. 5A may be referred to. Therefore, no duplicate description is given.

In operation 520b, the server may identify whether resonance associated with the floor (e.g., the floor state) is generated using at least one AI model, based on the RPM data and the vibration data. For the description of operation 520b, the description of operation 520a of FIG. 5A may be referred to. Therefore, no duplicate description is given.

In operation 530b, the server may adjust RPM-related information (e.g., the value of the final spin RPM) using the RPM value associated with the resonance, based on identifying that the resonance associated with the floor (e.g., the state of the floor) is generated, and transmit the adjusted RPM-related information to the washer. For the description of operation 530b, the description of operation 530a of FIG. 5A may be referred to. Therefore, no duplicate description is given.

In operation 540b, the server may maintain RPM-related information (e.g., the value of the final spin RPM) based on identifying that the resonance associated with the floor (e.g., the state of the floor) is not generated. For the description of operation 540b, the description of operation 540a of FIG. 5A may be referred to. Therefore, no duplicate description is given.

FIG. 6A is a flowchart illustrating a method for determining whether resonance associated with a state of a floor is generated using a first AI model according to one or more embodiments of the disclosure.

According to one or more embodiments, the method of FIG. 6A may be, e.g., an example of operation 430 of FIG. 4, operation 520a of FIG. 5A, or operation 520b of FIG. 5B.

According to one or more embodiments, the method of FIG. 6A may be performed by a washer (e.g., the washer 1 of FIG. 2) or a server (e.g., the AI server 210 of FIG. 2).

Referring to FIG. 6A, in operation 610a, the washer (or server) may input first input data generated based on RPM data and vibration data obtained while the spin cycle (e.g., the first spin cycle) is performed to the first AI model, and may obtain information about the state of the floor as output data of the first AI model.

According to one or more embodiments, the information about the state of the floor may be set to one of a first value indicating that the floor is a soft floor (e.g., a wood floor) or a second value indicating that the floor is a hard floor (e.g., a cement/concrete floor).

In one or more embodiments, the washer (or server) may pre-process the RPM data and vibration data to generate the input data. An example of a method for pre-processing RPM data and vibration data is described below with reference to FIGS. 8A and 8B.

According to one or more embodiments, in operation 620a, the washer (or server) may identify whether the resonance associated with the floor (e.g., the state of the floor) is generated using a designated algorithm for filtering the resonance caused by the laundry (e.g., the eccentricity of the laundry) received in the drum, based on the information about the state of the floor and the washer data including RPM data and vibration data. In the disclosure, the designated algorithm for filtering resonance by laundry (e.g., eccentricity of laundry) may be referred to as a filtering-out algorithm.

According to one or more embodiments, the washer data may further include data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed. The data related to the laundry may include, but is not limited to, information about the weight of the laundry received in the drum and/or information about the type of the laundry while the spin cycle (e.g., the first spin cycle) is performed.

According to one or more embodiments, the filtering-out algorithm may include a first algorithm configured to filter the resonance by the laundry (e.g., the eccentricity of the laundry) using information about a pattern associated with the resonance by the laundry (e.g., the eccentricity of the laundry), a second algorithm configured to filter the resonance by the laundry (e.g., the eccentricity of the laundry) using the vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed and additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed adjacent to the tub (e.g., the tub 30 of FIG. 1B) while the first spin cycle is performed, and/or a third algorithm configured to filter the resonance by the laundry (e.g., the eccentricity of the laundry) using information about a frequency of resonance generation. The additional vibration data obtained through the second sensor (or MEMS sensor) may include acceleration data on the x-axis and y-axis, or angular velocity data on the x-axis and y-axis. The additional vibration data obtained through the second sensor (or MEMS sensor) may be a sequence of values (e.g., acceleration values on the x-axis/y-axis or angular velocity values on the x-axis/y-axis) sampled according to the sampling frequency (e.g., 10 Hz).

According to one or more embodiments, when the first algorithm is used, the washer (or server) may store information about a pattern associated with resonance by the laundry (e.g., eccentricity of the laundry). For example, the washer (or server) may store information about the pattern associated with resonance by the eccentricity of the laundry for each state of the floor (e.g., soft floor or hard floor) and/or load condition (e.g., weight, type, etc. of the laundry). The load condition of the laundry may be identified based on data related to the laundry obtained through the above-described third sensor (e.g., the weight sensor of FIG. 1C).

According to one or more embodiments, when the first algorithm is used, the washer (or server) may compare the pattern (stored pattern) associated with the resonance by the eccentricity of the laundry corresponding to the floor state identified based on the information about the floor state obtained through the first AI model and/or the load condition identified based on data related to the laundry obtained through the third sensor with the pattern (e.g., the first pattern 1101d, the second pattern 1102d, or the third pattern 1103d of FIG. 11D) obtained based on the washer data, thereby determining whether the resonance is generated by the floor state or the eccentricity of the laundry. For example, when the two patterns are substantially the same, the washer (or server) may determine that the resonance is the resonance associated with eccentricity. For example, when the two patterns are not substantially the same, the washer (or server) may determine that the resonance is the resonance associated with the floor state.

According to one or more embodiments, when the second algorithm is used, the washer (or server) may identify an increase in the magnitude (e.g., vibration magnitude) of vibration data obtained through the first sensor and an increase in the magnitude (e.g., vibration magnitude) of additional vibration data obtained through the second sensor. For example, when there is no increase in the magnitude of the additional vibration data of the second sensor but there is an increase in the magnitude of the vibration data of the first sensor, the washer (or server) may determine that the corresponding resonance is the resonance associated with the floor state. For example, when the magnitude of the additional vibration data of the second sensor increases and the magnitude of the vibration data of the first sensor increases, the washer (or server) may determine that the corresponding resonance is the resonance associated with eccentricity.

According to one or more embodiments, when the third algorithm is used, the washer (or server) may determine whether the frequency of generation of resonance is larger than or equal to a designated frequency (e.g., three times during 10 spin cycles). For example, when the frequency of generation of the resonance is lower than the designated frequency (e.g., when the resonance is generated once during the 10 spin cycles), the washer (or server) may determine that the resonance is the resonance associated with eccentricity. For example, when the frequency of generation of the resonance is larger than or equal to the designated frequency (e.g., when the resonance is generated four times during the 10 spin cycles), the washer (or server) may determine that the resonance is the resonance associated with the floor state.

According to one or more embodiments, when the third algorithm is used, the washer (or server) may further consider a continuity condition (or a persistence condition) (e.g., three or more times consecutively) together with whether the frequency of generation of resonance is a designated frequency (e.g., two times during 10 spin cycles) or more. For example, when the frequency of generation of the resonance is lower than the designated frequency (e.g., when the resonance is generated once during the 10 spin cycles), the washer (or server) may determine that the resonance is the resonance associated with eccentricity. For example, when the frequency of generation of the resonance is larger than or equal to the designated frequency but does not meet the continuity condition (e.g., when resonance is generated five times during the 10 spin cycles, but is not generated three or more times consecutively), the washer (or server) may determine that the resonance is the resonance associated with eccentricity. For example, when the frequency of generation of the resonance is larger than or equal to the designated frequency and the continuity condition is met (e.g., when resonance is generated five times during the 10 spin cycles, and generated consecutively three or more times), the washer (or server) may determine that the resonance is the resonance associated with the floor state.

According to one or more embodiments, the washer (or server) may identify whether resonance associated with the floor (e.g., the floor state) is generated, using all of the first algorithm, the second algorithm, and the third algorithm. In this case, when it is identified that the resonance is the resonance associated with the laundry (e.g., the eccentricity of the laundry) by at least one of a first determination by the first algorithm, a second determination by the second algorithm, or a third determination by the third algorithm, the resonance may be filtered out, and the washer (or server) may identify that the resonance associated with the floor (e.g., the floor state) is not generated.

According to one or more embodiments, the washer (or server) may identify whether resonance associated with the floor (e.g., the floor state) is generated, using one or two of the first algorithm, the second algorithm, or the third algorithm. In this case, when it is identified that the resonance is the resonance associated with the laundry (e.g., the eccentricity of the laundry) by at least one of the determinations by the at least one algorithm used, the resonance may be filtered out, and the washer (or server) may identify that the resonance associated with the floor (e.g., the floor state) is not generated.

FIG. 6B is a flowchart illustrating a method for determining whether resonance associated with a state of a floor is generated using a first AI model and a second AI model according to one or more embodiments of the disclosure.

According to one or more embodiments, the method of FIG. 6B may be, e.g., an example of operation 430 of FIG. 4, operation 520a of FIG. 5A, or operation 520b of FIG. 5B.

According to one or more embodiments, the method of FIG. 6B may be performed by a washer (e.g., the washer 1 of FIG. 2) or a server (e.g., the AI server 210 of FIG. 2).

Referring to FIG. 6B, in operation 610b, the washer (or server) may input first input data generated based on RPM data and vibration data obtained while the spin cycle (e.g., the first spin cycle) is performed to the first AI model, and may obtain information about the state of the floor as output data of the first AI model.

In one or more embodiments, the washer (or server) may pre-process the RPM data and vibration data to generate the input data of the first AI model. An example of a method for pre-processing RPM data and vibration data is described below with reference to FIGS. 8A and 8B.

In operation 620b, the washer (or server) may input second input data generated based on information about the state of the floor and washer data including the RPM data and the vibration data to a second AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the second AI model.

According to one or more embodiments, the information indicating whether the resonance associated with the state of the floor is generated may be set to one of a first value indicating that the resonance associated with the state of the floor is generated or a second value indicating that the resonance associated with the state of the floor is not generated.

According to one or more embodiments, the washer data may further include data related to the laundry obtained through a third sensor (e.g., a weight sensor and/or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed adjacent to the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed. The data related to the laundry may include, but is not limited to, information about the weight of the laundry received in the drum and/or information about the type of the laundry while the spin cycle (e.g., the first spin cycle) is performed.

In one or more embodiments, the washer (or server) may pre-process the information about the state of the floor and the washer data to generate the input data of the second AI model. An example of a method for pre-processing the information about the state of the floor and the washer data is described below with reference to FIGS. 8A and 8B.

In the embodiment of FIG. 6B, unlike the embodiment of FIG. 6A using the filtering-out algorithm, an operation of determining whether resonance associated with the floor (e.g., the floor state) is generated using the information about the floor state and the washer data may be performed using an AI model (e.g., the second AI model).

FIG. 6C is a flowchart illustrating a method for determining whether resonance associated with a state of a floor is generated using a third AI model according to one or more embodiments of the disclosure.

According to one or more embodiments, the method of FIG. 6C may be, e.g., an example of operation 430 of FIG. 4, operation 520a of FIG. 5A, or operation 520b of FIG. 5B.

According to one or more embodiments, the method of FIG. 6C may be performed by a washer (e.g., the washer 1 of FIG. 2) or a server (e.g., the AI server 210 of FIG. 2).

Referring to FIG. 6C, in operation 610c, the washer (or server) may input third input data generated based on washer data including the RPM data and the vibration data to a third AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the third AI model.

In one or more embodiments, the washer (or server) may pre-process the washer data to generate the input data of the first AI model. An example of a method for pre-processing the washer data is described below with reference to FIGS. 8A and 8B.

In the embodiment of FIG. 6C, unlike the embodiments of FIGS. 6A and 6B, the operation of determining whether resonance associated with the floor (e.g., the floor state) is generated using the washer data may be performed using the third AI model without using the first AI model.

FIG. 7 is a flowchart illustrating a method for adjusting driving configuration information according to one or more embodiments of the disclosure.

In the embodiment of FIG. 7, an example is described in which the adjusted driving configuration information is RPM-related information (e.g., final spin RPM), but embodiments are not limited thereto. For example, various types of information included in the driving configuration information illustrated in FIG. 3 may be adjusted.

According to one or more embodiments, the method of FIG. 7 may be, e.g., an example of operation 440 of FIG. 4, operation 530a of FIG. 5A, or operation 530b of FIG. 5B.

According to one or more embodiments, the method of FIG. 7 may be performed by a washer (e.g., the washer 1 of FIG. 2) or a server (e.g., the AI server 210 of FIG. 2).

Referring to FIG. 7, in operation 710, the washer (or server) may identify the basis spin RPM causing the resonance in response to identifying that the resonance associated with the floor (e.g., the state of the floor) is generated. The washer (or server) may identify, e.g., 636 RPM of the first pattern 1101d, 617 RPM of the second pattern 1102d, or 584 RPM of the third pattern 1103d of FIG. 11D as the basis spin RPM.

In operation 720, the washer (or server) may obtain a harmonic resonance RPM based on the basis spin RPM. For example, when 636 RPM of the first pattern 1101d, 617 RPM of the second pattern 1102d, or 584 RPM of the third pattern 1103d of FIG. 11D is identified as the basis spin RPM, the washer (or server) may identify 1272 RPM, 1234 RPM, or 1168 RPM, which is a multiple of the basis spin RPM, as the harmonic resonance RPM.

In operation 730, the washer (or server) may adjust the final spin RPM to a value that is a designated first size higher, or a designated second size lower, than the harmonic resonance RPM within an allowable RPM range based on the harmonic resonance RPM. The designated first size and the designated second size may be, e.g., the same size or different sizes. According to one or more embodiments, the washer (or server) may adjust the final spin RPM based on the harmonic resonance RPM using the avoidance maneuver. For example, when 1272 RPM, 1234 RPM, or 1168 RPM is identified as the harmonic RPM, the washer (or server) may set the final spin RPM to a value (e.g., 1300 RPM) higher than the harmonic RPM within the allowable RPM range using the avoidance maneuver. According to one or more embodiments, the washer (or server) may adjust the final spin RPM based on the harmonic resonance RPM using a cut-off method. For example, when 1272 RPM, 1234 RPM, or 1168 RPM is identified as the harmonic RPM, the washer (or server) may set the final spin RPM to a value (e.g., 950 RPM) lower than the harmonic RPM within the allowable RPM range (e.g., 200 RPM to 1300 RPM) using the cut-off method.

FIG. 8A is a view illustrating a training processing module for training an AI model according to one or more embodiments of the disclosure.

FIG. 8B is a view illustrating an inference processing module for performing an inference using an AI model according to one or more embodiments of the disclosure.

The AI model of the embodiment of FIGS. 8A and 8B may be one of the AI models used in operation 430 of FIG. 4, operation 520a of FIG. 5A, operation 520b of FIG. 5B, operation 610a of FIG. 6A, operations 610b and 620b of FIG. 6B, and operation 610c of FIG. 6C.

Referring to FIG. 8A, a training processing module 800a for training an AI model may include a data collection module 810a, a pre-processing module 820a, and/or a training module 830a. According to one or more embodiments, the training processing module 800a may be included in a server (e.g., the AI server 210 of FIG. 2) and/or a washer (e.g., the washer 1 of FIG. 2).

According to one or more embodiments, the training processing module 800a may be set for each AI model to be trained.

According to one or more embodiments, the data collection module 810a may collect a training data set for training the corresponding AI model. The training data set of the data collection module 810a may be, e.g., a data set obtained in a laboratory.

For example, when the AI model is, e.g., the first AI model of FIGS. 6A and 6B, the training data set for the first AI model may include a plurality of pieces of first training data, and each piece of first training data may include first washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle) and first label data associated with the first washer data. The first label data may be set to a value indicating the state of the floor associated with the first washer data (e.g., a first value indicating a soft floor or a second value indicating a hard floor). The first washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed adjacent to the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

For example, when the AI model is, e.g., the second AI model of FIG. 6B, the training data set for the second AI model may include a plurality of pieces of second training data, and each piece of second training data may include information about the floor state obtained through the first AI model, second washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle) and second label data associated with the second washer data. The second label data may be set to a value indicating whether resonance associated with the floor (e.g., the floor state) associated with the second washer data is generated (e.g., a first value indicating that resonance associated with the floor state is generated or a second value indicating that resonance associated with the floor state is not generated). The second washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed on the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

For example, when the AI model is, e.g., the third AI model of FIGS. 6C, the training data set for the third AI model may include a plurality of pieces of third training data, and each piece of third training data may include third washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle) and third label data associated with the third washer data. The third label data may be set to a value indicating whether resonance associated with the floor (e.g., the floor state) associated with the third washer data is generated (e.g., a first value indicating that resonance associated with the floor state is generated or a second value indicating that resonance associated with the floor state is not generated). The third washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed on the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

According to one or more embodiments, the pre-processing module 820a may pre-process each piece of training data included in the corresponding training data set to generate pre-processed data. For example, the pre-processing module 820a may input pre-processed data obtained by performing pre-processing for each piece of training data to the training module 830a. An example of the training data may be as shown in FIG. 11A. For example, the training data may include RPM data having the pattern 1110a of FIG. 11A and first vibration data (e.g., MEMS data) having the first pattern 1101a of FIG. 11A.

According to one or more embodiments, the pre-processing module 820a may include a filtering module 821a and/or a dimension reduction/feature conversion module 822a. The operation of the filtering module 821a may be performed, e.g., before the operation of the dimension reduction/feature conversion module 822a is performed.

According to one or more embodiments, the pre-processing module 820a may further perform an optional down-sampling operation before performing the operation of the filtering module 821a. For example, when the training data is data obtained using a first sampling rate (e.g., a sampling rate of 10 Hz), the down-sampling operation may include down-sampling at a sampling rate (e.g., a sampling rate of 5 Hz) lower than the first sampling rate. Through this down-sampling process, the amount of data to be processed by the training processing module 800a is reduced by about half, but performance within an acceptable error range may be provided.

According to one or more embodiments, the filtering module 821a may filter the input training data (e.g., raw data or down-sampled data) to obtain noise-removed filtering data. For example, the filtering module 821a may perform filtering using a butter worth filter, but is not limited thereto. The butter worth filter may have, e.g., a filter order of 5th-order, a cut-off frequency of 0.1 Hz, and a sampling frequency of 5 Hz, but is not limited thereto. An example of the filtering data may be the same as shown in FIG. 11B. For example, the filtering data may include filtering data of RPM data having the pattern 1110a of FIG. 11B and filtering data of first vibration data (e.g., MEMS data) having the first pattern 1101b of FIG. 11B.

According to one or more embodiments, the dimension reduction/feature conversion module 822a may obtain the dimension-reduced and/or feature-converted data by applying the dimension reduction function and/or the feature conversion function to the input filtering data. For example, the dimensional reduction/feature conversion module 822a may obtain feature-converted data by applying the feature conversion function to the input filtering data.

According to one or more embodiments, the dimension reduction/feature conversion module 822a may obtain the feature-converted data, based on filtering data of RPM data and filtering data of vibration data. For example, the dimension reduction/feature conversion module 822a may set data obtained by multiplying filtering data of RPM data and filtering data of vibration data corresponding to the same time (e.g., the same sampling time) to the feature-converted data. For example, the dimension reduction/feature conversion module 822a may set data obtained by multiplying filtering data of RPM data and filtering data of vibration data corresponding to the same time (e.g., the same sampling time) and divided by a designated value (e.g., 5252) as the feature-converted data.

According to one or more embodiments, an example of the feature-converted data may be the same as those illustrated in FIGS. 11C and 11D. For example, the feature-converted data may include data having the first pattern 1101c of FIG. 11C or data having the first pattern 1101d of FIG. 11D. The feature-converted data may more clearly express features of data (or patterns) associated with resonance. Based on the feature-converted data, a basis spin RPM (e.g., the basis spin RPM obtained in operation 710 of FIG. 7) causing resonance may be obtained.

According to one or more embodiments, the dimension reduction/feature conversion module 822a may input pre-processed data including the feature-converted data and the label data to the training module 830a.

According to one or more embodiments, the training module 830a may train the AI model based on the pre-processed data using a designated training method (e.g., a supervised training method).

According to one or more embodiments, the first AI model (e.g., the first AI model of FIGS. 6A and 6B), the second AI model (e.g., the second AI model of FIG. 6B), and/or the third AI model (e.g., the third AI model of FIG. 6C) may have a multi-layer perceptron (MLP) structure, but are not limited thereto. For example, a structure of a convolutional neural network (CNN) or a recurrent neural network (RNN) may be used for a corresponding AI model.

According to one or more embodiments, when the first AI model, the second AI model, and/or the third AI model have an MLP structure, the AI model may include an input layer, one or more hidden layers, and an output layer.

Table 2 illustrates an example of a configuration of an MLP applicable to the first AI model, the second AI model, and/or the third AI model. However, this is merely an example, and another configuration may be applied to the corresponding AI model.

TABLE 2

input_shape=(Input_dimension= A(e.g., 120));

Dense(Hidden_layer_Size= B(e.g., 240), activation= C(e.g., ‘relu’));

...

Dropout(D (e.g., 0.5));

Dense(output_dimension= E (e.g., 2))

Referring to Table 2, the MLP structure applicable to the first AI model, the second AI model, and/or the AI model may include one input layer, three hidden layers, and one output layer.

According to one or more embodiments, the input data of the input layer may be defined as a vector having a designated number (e.g., 120) of features.

According to one or more embodiments, the hidden layers may include at least one dense layer (e.g., a first dense layer, a second dense layer, and a dropout layer).

According to one or more embodiments, the first dense layer is a fully connected layer, which may be a layer in which both the input and the output are connected, may have a designated number (e.g., 240) of neurons (nodes), may combine characteristics of input data to convert into a vector of a designated dimension (e.g., 240 dimensions), and may use, e.g., a rectified linear unit (ReLU) activation function as the activation function. The ReLU activation function is a non-linear activation function, e.g., if the output is less than 0, 0 may be output, and if the output is larger than 0, the value may be output as it is. The second dense layer may receive and process the output of the first dense layer as an input. Like the first dense layer, the second dense layer may have a designated number (e.g., 240) of neurons (nodes) and may use an ReLU activation function. According to one or more embodiments, the dropout layer may be used to prevent overfitting, and for example, a designated value (e.g., 0.5 (50%)) may be used as the dropout ratio. When 0.5 is used as the dropout ratio, 50% of neurons may be randomly deactivated (dropped) for each update step during training, which may increase generalization performance by preventing the AI model from overly relying on a specific neuron.

According to one or more embodiments, the output layer may have n (e.g., two) neurons (nodes). This means that the final output is a two-dimensional vector, and the corresponding output may be used for binary classification (2) and multiple classification (2 or more) prediction according to the size of the output vector to be distinguished. For example, the output of the first AI model may be set to information about the floor state for predicting the floor state (e.g., information set to a first value indicating that the floor state is a soft floor or a second value indicating that the floor is a hard floor). For example, the output of the second AI model and the output of the third AI model may be set to information about the resonance associated with the floor for predicting whether the resonance associated with the floor is generated (e.g., information set to a first value indicating that the resonance associated with the floor is generated or a second value indicating that the resonance associated with the floor is not generated).

Referring to FIG. 8B, e.g., the inference processing module 800b for performing inference using the AI model trained through the training processing module 800a of FIG. 8A may include a data collection module 810b, a pre-processing module 820b, and/or an inference module 830b. According to one or more embodiments, the inference processing module 800b may be included in a server (e.g., the AI server 210 of FIG. 2) and/or a washer (e.g., the washer 1 of FIG. 2).

According to one or more embodiments, the inference processing module 800b may be set for each trained AI model.

According to one or more embodiments, the data collection module 810b may collect an inference data set for performing inference using the trained AI model. The inference data set of the data collection module 810b may be, e.g., a data set obtained through actual use of the washer. As the driving configuration information (or driving profile) is updated through the inference process using the actual use data, the driving profile (or driving configuration information) may be maintained in a state optimized for the user's environment.

For example, when the AI model is, e.g., the first AI model of FIGS. 6A and 6B, the inference data set for the first AI model may include a plurality of pieces of first inference data, and each piece of first inference data may include first washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle). The first washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed adjacent to the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

For example, when the AI model is, e.g., the second AI model of FIG. 6B, the inference data set for the second AI model may include a plurality of pieces of second inference data, and each piece of second inference data may include information about the floor state obtained through the first AI model and second washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle). The second washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed on the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

For example, when the AI model is, e.g., the third AI model of FIG. 6C, the inference data set for the third AI model may include a plurality of pieces of third inference data, and each piece of third inference data may include third washer data including RPM data and vibration data obtained during a spin cycle (e.g., the first spin cycle). The third washer data may further include, e.g., data related to laundry obtained through a third sensor (e.g., a weight sensor or a torque sensor) while the spin cycle (e.g., the first spin cycle) is performed, vibration data obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) disposed adjacent to the housing (e.g., the housing 10 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed, and/or additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed on the tub (e.g., the tub 30 of FIG. 1B) while the spin cycle (e.g., the first spin cycle) is performed.

According to one or more embodiments, the pre-processing module 820b may pre-process each piece of inference data included in the corresponding inference data set to generate pre-processed data. For example, the pre-processing module 820b may input pre-processed data obtained by performing pre-processing for each piece of inference data to the inference module 830b for a single inference. An example of the inference data may be as shown in FIG. 11A. For example, the inference data may include RPM data having the pattern 1110a of FIG. 11A and first vibration data (e.g., MEMS data) having the first pattern 1101a of FIG. 11A.

According to one or more embodiments, the pre-processing module 820b may include a filtering module 821b and/or a dimension reduction/feature conversion module 822b. The operation of the filtering module 821a may be performed, e.g., before the operation of the dimension reduction/feature conversion module 822b is performed.

According to one or more embodiments, the pre-processing module 820b may further perform an optional down-sampling operation before performing the operation of the filtering module 821b. For example, when the inference data is data obtained using a first sampling rate (e.g., a sampling rate of 10 Hz), the down-sampling operation may include down-sampling the corresponding sampling data at a sampling rate (e.g., a sampling rate of 5 Hz) lower than the first sampling rate. Through this down-sampling process, the amount of data to be processed by the inference processing module 800b is reduced by about half, but performance within an acceptable error range may be provided.

According to one or more embodiments, the filtering module 821b may filter the input inference data (e.g., raw data or down-sampled data) to obtain noise-removed filtering data. For example, the filtering module 821b may perform filtering using a butter worth filter, but is not limited thereto. The butter worth filter may have, e.g., a filter order of 5th-order, a cut-off frequency of 0.1 Hz, and a sampling frequency of 5 Hz, but is not limited thereto. An example of the filtering data may be the same as shown in FIG. 11B. For example, the filtering data may include filtering data of RPM data having the pattern 1110b of FIG. 11B and filtering data of first vibration data (e.g., MEMS data) having the first pattern 1101b of FIG. 11B.

According to one or more embodiments, the dimension reduction/feature conversion module 822b may obtain the dimension-reduced and/or feature-converted data by applying the dimension reduction function and/or the feature conversion function to the input filtering data. For example, the dimensional reduction/feature conversion module 822b may obtain feature-converted data by applying the feature conversion function to the input filtering data.

According to one or more embodiments, the dimension reduction/feature conversion module 822b may obtain the feature-converted data, based on filtering data of RPM data and filtering data of vibration data. For example, the dimension reduction/feature conversion module 822b may set data obtained by multiplying filtering data of RPM data and filtering data of vibration data corresponding to the same time (e.g., the same sampling time) to the feature-converted data. For example, the dimension reduction/feature conversion module 822b may set data obtained by multiplying filtering data of RPM data and filtering data of vibration data corresponding to the same time (e.g., the same sampling time) and divided by a designated value (e.g., 5252) as the feature-converted data.

According to one or more embodiments, the dimension reduction/feature conversion module 822b may input the feature-converted data to the inference module 830b.

According to one or more embodiments, the inference module 830b may perform inference using the trained AI model.

According to one or more embodiments, when the first AI model of FIGS. 6A and 6B, the second AI model of FIG. 6B, and/or the third AI model of FIG. 6C have an MLP structure, the AI model may include an input layer, one or more hidden layers, and an output layer. An example of a configuration of an MLP applicable to the first AI model, the second AI model, and/or the third AI model may be equal to the example shown in Table 2 described above.

According to one or more embodiments, the inference module 830b of the first AI model may input the pre-processed data obtained by pre-processing the first washer data, as input data, and output information about the floor state for predicting the floor state (e.g., information set to a first value indicating that the floor state is a soft floor or a second value indicating that the floor is a hard floor).

According to one or more embodiments, the inference module 830b of the second AI model may input, as input data, pre-processed data obtained by pre-processing the information about the state of the floor output by the first AI model and the second washer data and output information about resonance associated with the floor for predicting whether the resonance associated with the floor is generated (e.g., information set to the first value indicating that the resonance associated with the resonance is generated or the second value indicating that the resonance associated with the floor is not generated).

According to one or more embodiments, the inference module 830b of the third AI model may input, as input data, pre-processed data obtained by pre-processing the third washer data and output information about resonance associated with the floor for predicting whether the resonance associated with the floor is generated (e.g., information set to the first value indicating that the resonance associated with the resonance is generated or the second value indicating that the resonance associated with the floor is not generated).

FIG. 9 illustrates a configuration of an electronic device providing an AI function according to one or more embodiments of the disclosure.

In the embodiment of FIG. 9, the electronic device 900 providing an AI function may be included in, e.g., the AI server 210 of FIG. 2, but is not limited thereto. For example, the electronic device 900 providing the AI function may be included in a washer (e.g., the washer 1 of FIG. 2).

According to one or more embodiments, the electronic device 900 may include a data collection module 910, an AI function module 920, a distribution module 930, and/or a database (DB) 940.

According to one or more embodiments, the data collection module 910 may include a data set for training an AI model or performing inference using the trained AI model. The data collection module 910 may perform, e.g., all or some of the functions of the data collection module 810a of FIG. 8A and/or all or some of the functions of the data collection module 810b of FIG. 8B. The data collection module 910 may transfer the collected data to the AI function module 920.

According to one or more embodiments, the AI function module 920 may train the AI model and/or perform inference using the trained AI model. The AI function module 920 may perform, e.g., all or some of the functions of the pre-processing module 820a and the training module 830a of FIG. 8A and/or all or some of the functions of the pre-processing module 820b and the inference module 830b of FIG. 8B. The AI function module 920 may transfer the trained AI model to the distribution module 930.

According to one or more embodiments, the AI function module 920 may retrain the trained AI model, based on a retraining trigger signal 901 received from another electronic device (e.g., the washer 1, the electronic device 210, or the external server 230 of FIG. 2). For example, the AI function module 920 may retrain the trained AI model in response to the retraining trigger signal 901 received from the other electronic device. The retraining may lead to an updated, performance-enhanced AI model.

According to one or more embodiments, the external server 230 may generate the retraining trigger signal 901 using the complaint data based on the resonance associated with the floor (e.g., the floor state), and may transmit the generated retraining trigger signal 901 to the electronic device 900. For example, the external server 230 may generate the retraining trigger signal 901 when a designated number (e.g., 1000 or more) of complaints based on resonance associated with the floor state are generated during a designated period (e.g., one month).

According to one or more embodiments, the washer 1 may generate a retraining trigger signal 901 based on the frequency of generation of resonance associated with the floor (e.g., the floor state), and may transmit the generated retraining trigger signal 901 to the electronic device 900. For example, the washer 1 may generate the retraining trigger signal 901 when the resonance associated with the floor state is generated at a designated frequency or more (e.g., when the resonance associated with the floor state is generated more than 6 times during 10 spin cycles).

According to one or more embodiments, the washer 1 may generate a retraining trigger signal 901 based on a user input, and may transmit the generated retraining trigger signal 901 to the electronic device 900. For example, the washer 1 may generate a retraining trigger signal 901 based on reception of a user input for retraining the AI model.

According to one or more embodiments, the electronic device 220 may generate a retraining trigger signal 901 based on a user input and may transmit the generated retraining trigger signal 901 to the electronic device 900. For example, the electronic device 220 may generate a retraining trigger signal 901 based on reception of a user input for retraining the AI model. When the AI model is retrained through the user input, the user may retrain the AI model before entrance of a complaint related to resonance associated with the floor state or processing of the complaint. Accordingly, the AI model may be quickly retrained, resolving the inconvenience that it takes long to reflect the user's complaint to retrain and redistribute the corresponding AI model.

According to one or more embodiments, the distribution module 930 may transmit the trained (or retrained) AI model to another electronic device (e.g., the washer 1 of FIG. 2). The transmission of the AI model may include, e.g., transmission of the AI model itself or transmission of information about parameters constituting the AI model.

FIG. 10 is a view illustrating a pattern of RPM data and a pattern of vibration data according to a floor state according to one or more embodiments of the disclosure.

Referring to FIG. 10, RPM data may have a pattern of stepwise increasing. For example, the RPM data may have a pattern of stepwise increasing to the final spin RPM (e.g., B RPM) through a low-speed spin section (e.g., a section equal to or lower than A RPM) and a high-speed spin section (e.g., a section equal to or higher than A RPM) and maintaining the final spin RPM for a designated duration. The final spin RPM (e.g., B RPM) may be larger than a reference RPM (e.g., A RPM) dividing the low-speed spin section and the high-speed spin section, but is not limited thereto. As illustrated in FIG. 10, RPM data shows a similar pattern in each floor state (e.g., a soft floor or a hard floor).

According to one or more embodiments, vibration data (e.g., MEMS data) shows a different pattern for each floor state. For example, in the soft floor, the vibration data peaks at about C RPM in the low-speed spin section. The peak RPM at which the peak occurs may be identified as a basis spin RPM (e.g., the basis spin RPM of operation 710 of FIG. 7) causing resonance.

FIGS. 11A to 11D are views illustrating a pattern of RPM data and a pattern of vibration data according to one or more embodiments of the disclosure.

Embodiments of FIGS. 11A to 11D illustrate a change in a pattern of RPM data and vibration data (e.g., MEMS data) included in training data used for one training or a change in a pattern of RPM data and vibration data (e.g., MEMS data) included in inference data used for one inference. Hereinafter, for convenience of description, an example in which a change in the pattern of FIGS. 11A to 11D is a change in the pattern of RPM data and vibration data (e.g., MEMS data) included in inference data is described. The same description may be applied to changes in patterns of RPM data and vibration data (e.g., MEMS data) included in training data.

The embodiment of FIG. 11A shows the pattern 1110a of the RPM data and the patterns 1101a, 1102a, and 1103a of the vibration data (e.g., MEMS data) included in the inference data (e.g., data of 500 samples) used for one inference or the down-sampling data (e.g., the data of 250 samples) obtained by down-sampling the corresponding inference data. The pattern 1101a of the first vibration data, the pattern 1102a of the second vibration data, and the pattern 1103a of the third vibration data may be, e.g., vibration data obtained by setting different states of laundry (e.g., dry cloth) which is subjected to a spin cycle in the same floor state.

Referring to FIG. 11A, the pattern 1110a of RPM data shows a pattern of stepwise increasing. The pattern 1101a of the first vibration data, the pattern 1102a of the second vibration data, and the pattern 1103a of the third vibration data show a pattern of data including noise.

The embodiment of FIG. 11B shows, e.g., patterns 1101b, 1101b, and 1103b of the filtering data obtained by performing a filtering process on the first vibration data, the second vibration data, and the third vibration data (e.g., a filtering process by the filtering module 821b of FIG. 8B) of the embodiment of FIG. 11A.

Referring to FIG. 11B, a pattern 1101b of filtered first vibration data obtained by filtering the first vibration data of FIG. 11A, a pattern 1102b of filtered second vibration data obtained by filtering the second vibration data, and a pattern 1103b of filtered third vibration data obtained by filtering the third vibration data show noise-removed patterns.

The embodiment of FIG. 11C shows the patterns 1101c, 1102c, and 1103c of the feature-converted data obtained by performing a feature conversion process (e.g., the feature conversion process by the dimension reduction/feature conversion module 822b of FIG. 8B) to clearly identify characteristics associated with the resonance of the region of interest (ROI) of the filtered first vibration data, the filtered second vibration data, and the filtered third vibration data of the embodiment of FIG. 11B.

Referring to FIG. 11C, the pattern 1101c of the feature-converted first vibration data obtained by feature-converting the filtered first vibration data of FIG. 11B, the pattern 1102c of the feature-converted second vibration data obtained by feature-converting the second vibration data, and the pattern 1103c of the feature-converted third vibration data obtained by feature-converting the third vibration data show patterns in which the peak RPM may clearly be identified in the region of interest.

The embodiment of FIG. 11D is an enlarged view of the region of interest of FIG. 11C. Referring to FIG. 11D, the pattern 1101d corresponding to the feature-converted first vibration data of FIG. 11C shows having A RPM as the peak RPM (e.g., basis spin RPM), the pattern 1102d corresponding to the feature-converted second vibration data shows having B RPM (e.g., an RPM smaller than A RPM) as the peak RPM (e.g., basis spin RPM), and the pattern 1103d corresponding to the feature-converted third vibration data shows having CRPM (e.g., an RPM smaller than B RPM) as the peak RPM (e.g., basis spin RPM).

According to one or more embodiments, a washer may be provided. A washer may comprise a housing, a tub disposed in the housing, a drum disposed in the tub and configured to be rotatable with respect to the tub, a driver disposed in the housing to rotate the drum, at least one sensor including a first sensor (e.g., the first vibration sensor 151 of FIG. 1B), at least one processor, and a memory including one or more storage media storing instructions. The instructions may, when executed by the at least one processor, cause the washer to perform at least one operation. The at least one operation may include obtaining driving configuration information based on a driving profile of the washer. The driving configuration information may include information related to revolutions per minute (RPM) of the driver or the drum associated with a spin cycle of the washer. The at least one operation may include obtaining RPM data and vibration data while a first spin cycle is performed based on the driving configuration information. The RPM data may include RPM values of the driver or the drum obtained while the first spin cycle is performed, and the vibration data may include values related to a vibration of the washer obtained through the first sensor while the first spin cycle is performed. The at least one operation may include identifying whether resonance associated with a floor (e.g., a state of the floor) where the washer is disposed occurs, using at least one artificial intelligence (AI) model based on the RPM data and the vibration data. The at least one operation may include adjusting a value of the information related to RPM (e.g., RPM-related information) using an RPM value associated with the resonance based on identifying that the resonance associated with the state of the floor is generated. The adjusted RPM-related information may be used to perform a second spin cycle performed after the first spin cycle.

The at least one operation may include inputting input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model and identifying whether the resonance associated with the state of the floor is generated using a designated algorithm for filtering resonance by laundry (e.g., the eccentricity of the laundry) received in the drum, based on washer data including the RPM data and the vibration data and the information about the state of the floor. The information about the state of the floor may be set to one of a first value indicating that the floor is a soft floor or a second value indicating that the floor is a hard floor.

The designated algorithm may include a first algorithm configured to filter the resonance associated with the laundry using information about a pattern associated with the resonance by the laundry (e.g., the eccentricity of the laundry), a second algorithm configured to filter the resonance by the laundry (e.g., the eccentricity of the laundry) using the vibration data and additional vibration data obtained through a second sensor (e.g., the second vibration sensor 152 of FIG. 1B) disposed adjacent to the tub while the first spin cycle is performed, and a third algorithm configured to filter the resonance by the laundry (e.g., the eccentricity of the laundry) using information about a frequency of resonance generation.

The at least one operation may include inputting input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model, and inputting input data generated based on the information about the state of the floor and washer data including the RPM data and the vibration data to a second AI model and obtain information indicating whether the resonance associated with the state of the floor is generated as output data of the second AI model. The information indicating whether the resonance associated with the state of the floor is generated may be set to one of a first value indicating that the resonance associated with the state of the floor is generated or a second value indicating that the resonance associated with the state of the floor is not generated.

The at least one operation may include inputting input data generated based on washer data including the RPM data and the vibration data to a third AI model and obtaining information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the third AI model.

The RPM-related information may include information indicating a setting value of a final spin RPM to be used in the first spin cycle. The at least one operation may include identifying a basis spin RPM causing the resonance in response to identifying that the resonance associated with the floor (e.g., the state of the floor) is generated, obtaining a harmonic resonance RPM based on the basis spin RPN, and adjusting the final spin RPM based on the harmonic resonance RPM.

The RPM data and the vibration data may be obtained at the same sampling period while the first spin cycle is performed.

The at least one operation may include obtaining weight data related to a weight of the laundry through a third sensor while the first spin cycle is performed. The weight data may be further included in the washer data used to identify whether the resonance associated with the floor (e.g., the state of the floor) is generated.

The at least one AI model may be trained using a supervised learning scheme using training data including the vibration data obtained through the first sensor, the RPM data related to the RPM of the driver or the drum, the information related to the state of the floor, and label data.

The at least one AI model may be again trained based on a trigger signal received from a server, and the trigger signal may be generated by the server based on complaint data related to the resonance.

The driving profile may correspond to an initial driving profile selected based on information related to a position where the washer is installed among a plurality of driving profiles, and the plurality of driving profiles may be configured through statistical analysis of a residential environment.

According to one or more embodiments, a server may be provided. The server may comprise a communication device, at least one processor, and a memory including one or more storage media storing instructions. The instructions may, when executed by the at least one processor, cause the server to perform at least one operation. The at least one operation may include receiving RPM data and vibration data obtained while a first spin cycle is performed based on driving configuration information, from a washer through the communication device. The driving configuration information may be obtained based on a driving profile of the washer and may include information related to revolutions per minute (RPM) of the driver or the drum associated with a spin cycle of the washer. The RPM data may include RPM values of the driver or the drum obtained while the first spin cycle is performed. The vibration data may include values related to a vibration of the washer obtained through the first sensor (e.g., the first vibration sensor 151 of FIG. 1B) while the first spin cycle is performed. The at least one operation may include identifying whether resonance associated with a floor (e.g., a state of the floor) where the washer is disposed occurs, using at least one artificial intelligence (AI) model based on the RPM data and the vibration data. The at least one operation may include adjusting a value of the information related to RPM (e.g., RPM-related information) using an RPM value associated with the resonance based on identifying that the resonance associated with the state of the floor is generated and transmitting the adjusted RPM-related information to the washer through the communication device. The adjusted RPM-related information may be used to perform a second spin cycle performed after the first spin cycle.

The at least one operation may include inputting input data generated based on the RPM data and the vibration data to a first AI model and obtain information about a state of the floor as output data of the first AI model, and inputting input data generated based on the information about the state of the floor and washer data including the RPM data and the vibration data to a second AI model and obtain information indicating whether the resonance associated with the floor (e.g., the state of the floor) is generated as output data of the second AI model. The information indicating whether the resonance associated with the state of the floor is generated may be set to one of a first value indicating that the resonance associated with the state of the floor is generated or a second value indicating that the resonance associated with the state of the floor is not generated.

The RPM-related information may include information indicating a setting value of a final spin RPM to be used in the first spin cycle. The at least one operation may include identifying a basis spin RPM causing the resonance in response to identifying that the resonance associated with the state of the floor is generated, obtaining a harmonic resonance RPM based on the basis spin RPN, and adjusting the final spin RPM based on the harmonic resonance RPM.

The RPM data and the vibration data may be obtained at the same sampling period while the first spin cycle is performed.

The at least one operation may include receiving, from the washer, weight data related to a weight of the laundry obtained through a third sensor while the first spin cycle is performed. The weight data may be further included in the washer data used to identify whether the resonance associated with the state of the floor is generated.

FIG. 12 is a view illustrating an operation of providing a notification for generation of resonance associated with a floor state by a washer according to one or more embodiments of the disclosure.

Referring to FIG. 12, the washer 1 may identify that resonance associated with the floor state is generated. For a description of identifying that resonance (or vibration) associated with the floor state is generated by the washer 1, e.g., the description of FIGS. 4 to 7 may be referred to.

According to one or more embodiments, the washer 1 may display a notification message (visual notification message) through a display (e.g., the display 52 of FIG. 1C) of the washer 1 based on identifying that resonance associated with the floor state is generated.

According to one or more embodiments, the notification message may include warning information 1210 (e.g., “warning: washer floor unstable”) for providing a warning for resonance associated with the floor state and/or guide information 1220 (e.g., “adjust the position of the washer or place the washer on a hard floor”) for a method for solving the problem with the corresponding warning.

According to one or more embodiments, the warning information 1210 and the guide information 1220 may be displayed on the same screen or may be displayed on different screens, respectively. When the warning information 1210 and the guide information 1220 are displayed on separate screens, the warning information 1210 may be displayed before the guide information 1220. For example, the warning information 1210 may be displayed first, and if user identification for the warning information 1210 is identified, the guide information 1220 may be then displayed.

According to one or more embodiments, the washer 1 may provide warning information 1210 and/or guide information 1220 by voice through a speaker.

According to one or more embodiments, the washer 1 may provide the warning information 1210 and/or the guide information 1220 as a visual notification message, as a voice notification message, or as both the visual notification message and the voice message, based on selection of the notification method based on the user input.

According to one or more embodiments, the washer 1 may include a plurality of floor legs (e.g., four floor legs) for fitting the height and/or balancing the washer 1. For example, the washer 1 may include a plurality of floor legs automatically length-adjustable by the washer 1. In this case, the washer 1 may provide the warning information 1210 and automatically adjust the lengths of the plurality of floor legs based on identifying that the resonance associated with the floor state is generated, thereby preventing generation of resonance. In this case, the guide information 1220 may not be provided, but is not limited thereto.