122 can be to the extended lane line perpendicular to the lane line. In this manner, errors in metric computation can be avoided.

FIG. 5 graphically depicts example operations of the intrusion system 140 for an object in an intersection. At numeral 1, an intersection lane 410 is depicted with a bounding box 122 and vehicle 110 traveling upward toward the interaction. At numeral 2, the lane boundary extension component 250 extends the lane lines. At numeral 3, the vertex component 220 processes the vertices of the bounding box 122 by projecting the vertices to the line lines and computing one or more ratios associated with each vertex. At numeral 4, the metric component 230 generates a metric based on the output of the vertex component 220. As shown, intrusion from the left to the rightmost point is identified as approximately 80%, and intrusion from the right to the leftmost point is approximately 40%. The minimum of the two values can be selected as 40% and returned as the metric representing how much the object is intruding or occupying lane width. Alternatively, the minimum of 40% can be subtracted from 100% to return 60% as a metric representing available lane width space not occupied by the object.

Example Methods of Intrusion-Based Collision Avoidance

Turning to FIG. 6, an illustrative flow diagram depicts an example method 600 for collision avoidance or mitigation. The method 600 can be executed by the collision avoidance system 150 of FIG. 1.

The method 600 starts at block 610 by identifying an object or obstacle in a lane. An object can be identified based on data from one or more sensors. The sensors can correspond to cameras, light detection and ranging (LiDAR) systems, radar, or a combination thereof. Further, machine learning and computer vision technologies may be utilized to analyze data collected by sensors to identify objects such as other vehicles, pedestrians, and cyclists.

The method 600 continues at block 620 by determining a metric that describes how the object blocks the lane. The object may partially or fully block the lane. Additionally, an object may be small and block only a portion of the lane but may be centrally located, blocking a significant portion of the lane. In accordance with one embodiment, the metric can capture the available space in terms of lane width, given the object's presence, size, and position in the lane. In this manner, the metric can quantify how the object blocks a lane. Further details regarding the metric determination are presented in the following flow diagram of FIG. 7.

Next, the method 600 proceeds to block 630 with a decision regarding whether or not the metric satisfies one or more thresholds. One or more threshold values can be predetermined. The metric value can be compared to one or more threshold values to determine whether a threshold is satisfied or unsatisfied. For example, a threshold value can quantify that the object occupies more than 90% of a lane. The metric value can be compared to the threshold value to determine whether the threshold is satisfied or not. If the threshold is satisfied, the method 600 can continue to block 640. If the threshold is not satisfied, the method 600 can terminate.

The method 600 can proceed to block 640 when the threshold is satisfied. At block 640, the method 600 triggers intervention. The intervention can be in the form of a driver warning or alert or corrective action such as applying vehicle brakes or performing evasive steering, among other things. Different thresholds can be specified for different forms of intervention or combinations. For example, a threshold can be specified that is associated with alerting a driver, automatically applying a vehicle brake, or both, while a different threshold can be associated with evasive steering.

Note that FIG. 6 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.

FIG. 7 depicts an illustrative flow diagram of an example method 700 for determining a metric that describes how an object blocks a lane. The method 600 can be executed by the intrusion system 140 of FIGS. 1 and 2 including the bounding component 210, vertex component 220, metric component 230, and lane boundary extension component 250.

The method 700 starts at block 710, identifying the traffic lane in which a vehicle is currently traveling. In accordance with one embodiment, a global positioning system associated with the vehicle can be employed to receive coordinates that provide a precise geographic location. The coordinates can be analyzed with reference to a digital map that identifies traffic lanes and intersections, among other things, to identify the traffic lane. Additionally or alternatively, sensor data such as images or video can be analyzed to determine a particular traffic lane among many of a road, for example, based on analysis of lane lanes painted on a road.

The method 700 continues at block 720 with extending lane boundary lines. If a vehicle is about to enter an intersection, the lane portion through solely the intersection may not provide enough data to perform further processing accurately. Accordingly, the lane lines considered can be extended before and after the intersection. For example, the lane lines considered can be ten meters before and after the intersection.

The method 700 can proceed at block 730 with determining a bounding box for an object detected to be present in the lane the vehicle is traveling. The bounding box can be a rectangle, square, or any other polygon encompassing an object. The bounding box can capture an object's size, shape, and position relative to the lane. The bounding box can be generated by a machine learning model trained to detect objects in an image or video. Further sensor data, for example, from radar, LIDAR, or an imaging depth camera can also be utilized to refine the bounding box. The bounding box can be specified with respect to the center of an object as well as at least the object's length and width.

The method 700 continues at block 740 with determining ratios for each vertex of the bounding box. Each vertex can be projected to a left lane line and a right lane line substantially perpendicular thereto. Subsequently, a Euclidean distance computation can be performed to determine the distance between two points, namely the vertex and the left lane line projection and the vertex and the right lane line projection. A ratio can be determined based on the distance to quantify the extent to which an object intrudes into the lane as a percentage of the total lane width. More specifically, a left-to-right ratio can be determined that represents the distance between the left lane line and a vertex. In one instance, the vertices of a bounding box that reside outside a lane need not be considered or can be assigned a left-to-right ratio of 1.0 and a right-to-left ratio of 0.0 for vertices that are beyond the right lane line and vice versa for vertices that are beyond the left lane line. Further, a left-to-right ratio can be determined that represents a relative distance between the right lane line and a vertex. In accordance with one embodiment, one of the left-to-right or right-to-left ratios can be determined and the other ratio can be determined by subtracting the ratio from 1.0 or 100 if the ratio is represented as a percentage. For example, if the left-to-right ratio is 0.6 (or 60%), then the right-to-left ratio is “1.0-0.6” or 0.4 (or 40%).

The method 700 continues at block 750 with identifying the ratio associated with the rightmost vertex. The ratio can be identified by locating the maximum left-to-right ratio amongst bounding box vertices.

Next, the method 700 proceeds at block 760 with identifying the ratio associated with the leftmost vertex. The ratio can be determined by identifying the maximum right-to-left ratio amongst vertices of the bounding box.

The method 700 continues to block 770 with selecting the minimum of the rightmost ratio and the leftmost ratio as the intrusion ratio. For example, consider an object entirely contained within the lane (e.g., no portion outside the lane) with a rightmost point ratio of 60% and a leftmost point ratio of 90%. In this scenario, the intrusion ratio is 60%, meaning the object occupies 40% of the lane width with 60% of the lane available. Subsequently, the method 700 can terminate.

Note that FIG. 7 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.

Example Intrusion Ratio Determinations

FIGS. 8A-D illustrated example intersections with object intrusions to aid clarity and understanding regarding determining an intrusion ratio. Each figure depicts an intersection 410 with extended lane lines and a particular bounding box 122. FIG. 8A depicts a bounding box 122 for an object that extends beyond intersection lane lines on both sides. In this scenario, the intrusion ratio as a percentage is 100%. FIG. 8B illustrates a bounding box 122 that appears halfway inside the right lane line of the intersection and halfway outside the right lane line of the intersection. Here, the intrusion ratio as a percentage of the lane width is 50%. FIG. 8C depicts a smaller bounding box than those in FIGS. 8A and B, but is centrally located in the middle of the intersection lane lines. The intrusion ratio as a percentage for this small centrally disposed bounding box 122 is also 50%. FIG. 8D depicts a bounding box 122 located on the left side of the intersection lane lines extending about halfway between the left and right lane lines. In this situation, the intrusion ratio as a percentage is also 50%.

For simplicity and clarity, bounding boxes have be depicted and discussed for a single object or obstacle often moving from right to left. It is to be appreciated that the metric is agnostic to the direction of intrusion. Further, the metric can be a composite metric accounting for more than one object or obstacle in a lane or intersection.

Example Processing System for Intrusion-Based Collision Avoidance

FIG. 9 depicts an example processing system configured to perform various aspects described herein, including, for example, methods as described above with respect to FIGS. 7 and 8.

To provide a context for the disclosed subject matter, FIG. 9, as well as the following discussion, are intended to provide a brief, general description of a suitable environment in which various aspects of the disclosed subject matter can be implemented. The suitable environment is solely an example and is not intended to suggest any limitation on the scope of use or functionality.

With reference to FIG. 9, illustrated is an example computing device 112. The computing device 112 includes one or more processor(s) 910, memory 920, bus 930, storage device(s) 940, input device(s) 950, output device(s) 960, and network interface(s) 970. The bus 930 communicatively couples at least the above system constituents. However, the computing device 112, in its simplest form, can include one or more processors 910 coupled to at least one memory 920, wherein the one or more processors 910 execute various computer-executable actions, instructions, and or components stored in the memory 920 and retrieved from storage devices 940.

The bus 930 may be formed from any medium capable of transmitting a signal, such as conductive wires, conductive traces, optical waveguides, connectors, or the like. In one embodiment, the bus 930 comprises a combination of conductive traces, conductive wires, connectors, and cooperate to permit the transmission of electrical data signals to components such as the processor(s) 910, memory 920, storage device(s) 940, input device(s) 950, output device(s) 960, and network interface(s) 970.

The processor(s) 910 can be implemented with a general-purpose processor, a microprocessor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor can be a microprocessor, but in the alternative, the processor can be any processor, controller, microcontroller, or state machine. The processor(s) 910 can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, multi-core processors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. In one embodiment, the processor(s) 910 can be a graphics processor unit (GPU) that performs calculations concerning digital image processing and computer graphics.

The computing device 112 can include or otherwise interact with a variety of computer-readable media to facilitate control of the computing device 112 to implement one or more aspects of the disclosed subject matter. The computer-readable media can be any available media accessible to the computing device 112 and includes volatile and nonvolatile media and removable and non-removable media. Computer-readable media can comprise two distinct and mutually exclusive types: storage media and communication media.

Storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology to store information such as computer-readable instructions, data structures, program modules, or other data. Storage media includes memory devices (e.g., random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM)), magnetic storage devices (e.g., hard disk, floppy disk, cassettes, tape), optical disks (e.g., compact disk (CD), digital versatile disk (DVD)), and solid-state devices (e.g., solid-state drive (SSD), flash memory drive (e.g., card, stick, key drive)), or any other like mediums that store, as opposed to transmit or communicate, the desired information accessible by the computing device 112. Accordingly, storage media excludes modulated data signals as well as that which is described with respect to communication media.

Communication media embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.