Collision risk assessment method and device, electronic equipment and autonomous vehicle

Abstract

The invention provides a collision risk assessment method and device, electronic equipment and an automatic driving vehicle, and relates to the technical field of computers, in particular to the application fields of automatic driving, vehicle-road cooperation, intelligent transportation and the like. The specific implementation scheme is as follows: determining a current sampling point related to a target object; determining a target evaluation area corresponding to the target object when the target object travels to the current sampling point; based on a preset distance field, obtaining a position relationship between the target evaluation area and an obstacle; and obtaining a collision risk assessment result corresponding to the current sampling point based on the position relationship.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technologies, and in particular, to application fields such as autonomous driving, vehicle-road cooperation, and intelligent transportation. Specifically, it relates to a collision risk assessment method, apparatus, electronic device, and autonomous vehicle. Background Art

[0002] With the booming rise of autonomous driving technology, the problems of driving planning and parking planning for autonomous vehicles are increasingly becoming the focus of research, and the core is concentrated on the in-depth application of collision risk assessment technology. The collision risk assessment technology aims to enhance the safety of autonomous vehicles during driving and parking by analyzing potential collision risks in real time to ensure a more reliable autonomous driving experience. Summary of the Invention

[0003] The present disclosure provides a collision risk assessment method, apparatus, electronic device, and autonomous vehicle.

[0004] According to a first aspect of the present disclosure, there is provided a collision risk assessment method, including:

[0005] Determine a current sampling point related to a target object;

[0006] Determine a target evaluation area corresponding to the target object when it travels to the current sampling point;

[0007] Based on a preset distance field, obtain the positional relationship between the target evaluation area and an obstacle;

[0008] Based on the positional relationship between the target evaluation area and the obstacle, obtain a collision risk assessment result corresponding to the current sampling point.

[0009] According to a second aspect of the present disclosure, there is provided a collision risk assessment apparatus, including:

[0010] A sampling point determination unit for determining a current sampling point related to a target object;

[0011] An area determination unit for determining a target evaluation area corresponding to the target object when it travels to the current sampling point;

[0012] A positional relationship acquisition unit for obtaining the positional relationship between the target evaluation area and an obstacle based on a preset distance field;

[0013] An evaluation result acquisition unit for obtaining a collision risk assessment result corresponding to the current sampling point based on the positional relationship between the target evaluation area and the obstacle.

[0014] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0015] At least one processor;

[0016] A memory communicatively connected to at least one processor;

[0017] The memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to execute the method provided in the first aspect of the present disclosure.

[0018] According to a fourth aspect of the present disclosure, there is provided a self-driving vehicle including an electronic device.

[0019] According to a fifth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method provided in the first aspect of the present disclosure.

[0020] According to a sixth aspect of the present disclosure, there is provided a computer program product including a computer program, and the computer program implements the method provided in the first aspect of the present disclosure when executed by a processor.

[0021] Adopting the present disclosure can improve the accuracy of the collision risk assessment result on the basis of ensuring the acquisition efficiency of the collision risk assessment result.

[0022] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0024] Figure 1 It is a schematic flowchart of a collision risk assessment method provided by an embodiment of the present disclosure;

[0025] Figure 2 It is a schematic diagram of a determination method of a current sampling point provided by an embodiment of the present disclosure;

[0026] Figure 3 It is a schematic diagram of another determination method of a current sampling point provided by an embodiment of the present disclosure;

[0027] Figure 4 It is an explanatory diagram of a construction method of an inscribed circular region provided by an embodiment of the present disclosure;

[0028] Figure 5 It is an explanatory diagram of a construction method of a corner circular region provided by an embodiment of the present disclosure;

[0029] Figure 6 It is an explanatory diagram of another construction method of an inscribed circular region provided by an embodiment of the present disclosure;

[0030] Figure 7 It is an auxiliary explanatory diagram of the complete process of a collision risk assessment method provided by an embodiment of the present disclosure;

[0031] Figure 8 It is a schematic diagram of an application scenario of a collision risk assessment method provided by an embodiment of the present disclosure;

[0032] Figure 9 It is a schematic structural block diagram of a collision risk assessment device provided by an embodiment of the present disclosure;

[0033] Figure 10 It is a schematic structural block diagram of an electronic device provided by an embodiment of the present disclosure. Detailed implementation manners

[0034] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted below.

[0035] Currently, collision risk assessment technologies are usually implemented based on a collision determination method based on geometric profiles or a rasterized space traversal analysis method.

[0036] Specifically, the collision determination method based on geometric profiles is used to abstract the target object and obstacles into specific geometric figures, such as rectangles, circles, line segments, triangles, or three-dimensional polygons, etc. By determining whether there is spatial overlap between the geometric figure abstracted from the target object and the geometric figure abstracted from the obstacle when the target object travels to the current sampling point, it is determined whether there is a risk of collision with the obstacle when the target object travels to the current sampling point. However, when using this method, in the face of highly complex collision objects, in order to improve the accuracy of the collision risk assessment result, it is usually necessary to use more categories and more numbers of geometric figures to represent these obstacles, which will greatly increase the computational burden of the autonomous driving system, thus severely restricting the acquisition efficiency of the collision risk assessment result. Therefore, it is impossible to improve the accuracy of the collision risk assessment result on the basis of ensuring the acquisition efficiency of the collision risk assessment result.

[0037] As for the rasterized space traversal analysis method, it is used to divide the driving space of the target object into multiple discrete grid cells, mark the grid cells occupied by obstacles as blocked states, and then determine whether there is a risk of collision with obstacles when the target object travels to the current sampling point by detecting whether the grid cell occupied by the target object when it travels to the current sampling point is in a blocked state. However, when using this method, in order to improve the accuracy of the collision risk assessment result, it is usually necessary to increase the fineness of the grid cells, which will also seriously restrict the acquisition efficiency of the collision risk assessment result. Therefore, it is also impossible to improve the accuracy of the collision risk assessment result on the basis of ensuring the acquisition efficiency of the collision risk assessment result.

[0038] In view of the above problems, the embodiments of the present disclosure provide a collision risk assessment method, which can be applied to an electronic device. Among them, the electronic device can be an autonomous driving controller. Hereinafter, with reference to Figure 1 the following flow schematic diagram, a collision risk assessment method provided by the embodiments of the present disclosure will be described. It should be noted that although the logical order is shown in the flow schematic diagram, in some cases, the steps shown or described in the flowchart may also be executed in other orders.

[0039] Step S101, determine the current sampling point related to the target object.

[0040] Among them, the target object can be an autonomous driving vehicle, a robot, a drone, etc. Here, the autonomous driving vehicle can be a partially automated (L2)-level, conditionally automated (L3)-level, highly automated (L4)-level or fully automated (L5)-level autonomous driving vehicle.

[0041] In addition, in the embodiments of the present disclosure, the current sampling point can be located on the driving planning path of the target object, and the current sampling point has corresponding pose information to represent the body position and body attitude of the target object when it travels to the current sampling point. Among them, the body attitude can be the body orientation.

[0042] Step S102, determine the target evaluation area corresponding to the target object when it travels to the current sampling point.

[0043] Among them, at least part of the target evaluation area can be located within the safety surrounding area corresponding to the target object when it travels to the current sampling point, and the boundary distance from the safety surrounding area meets the preset safety distance requirement. Here, the safety surrounding area can be a polygon constructed around the target object and having a certain interval distance from the target object, for example, a rectangle, an octagon, etc.; the preset safety distance requirement can be that the boundary distance is less than or equal to the preset distance threshold, and the preset distance threshold can be set according to specific application requirements (for example, driving scenario, safety performance requirements, etc.), and the embodiments of the present disclosure do not limit this.

[0044] In addition, in the embodiments of the present disclosure, after determining the current sampling point related to the target object, based on the pose information corresponding to the current sampling point, a safety surrounding area corresponding to the target object when traveling to the current sampling point can be determined, and a target evaluation area can be determined on the safety surrounding area. When determining the target evaluation area, the following area determination principles need to be followed: ensuring that at least part of the target evaluation area is located within the safety surrounding area corresponding to the target object when traveling to the current sampling point, and ensuring that the boundary distance between this at least part of the area located within the safety surrounding area and the boundary of the safety surrounding area meets the preset safety distance requirement.

[0045] Step S103: Obtain the positional relationship between the target evaluation area and the obstacle based on the preset distance field.

[0046] Among them, the preset distance field can be a spatial field with a corresponding distance data set, and the distance data set is used to store the in-field interval distance between each point in the preset distance field and other points, or is used to store the nearest obstacle distance corresponding to each point in the preset distance field; the positional relationship between the target evaluation area and the obstacle is used to characterize whether the obstacle is located within the target evaluation area.

[0047] In one example, based on a preset distance field, the interval distance between a first target point representing a target evaluation area and a second target point representing an obstacle can be obtained as a first reference distance, and based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained. Among them, the preset distance field can be a Manhattan distance field, an Euclidean Signed Distance Field (ESDF), a Truncated Signed Distance Function (TSDF), etc. Here, the Manhattan distance field has a corresponding distance data set for storing the in-field interval distance from each point in the Manhattan distance field to other points, specifically the Manhattan distance. Therefore, based on the Manhattan distance field, the interval distance between a first target point representing a target evaluation area and a second target point representing the nearest obstacle (i.e., the obstacle closest to the first target point) can be obtained (here, specifically the Manhattan distance from the first target point to the second target point) as the first reference distance, and based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained; the ESDF also has a corresponding distance data set for storing the nearest obstacle distance corresponding to each point in the ESDF. Therefore, based on the ESDF, the interval distance between a first target point representing a target evaluation area and a second target point representing the nearest obstacle can be obtained (here, specifically the nearest obstacle distance corresponding to the first target point) as the first reference distance, and based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained; for the TSDF, since it can be converted into the ESDF, when the distance field is the TSDF, it can be converted into the ESDF, and then based on the ESDF, the positional relationship between the target evaluation area and the obstacle can be obtained.

[0048] In a specific example, "based on the first reference distance, obtaining the positional relationship between the target evaluation area and the obstacle" may include: when the first reference distance is less than or equal to the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located within the target evaluation area; or, when the first reference distance is greater than the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located outside the target evaluation area. Among them, the second reference distance may be the distance value from the first target point to the boundary of the target evaluation area.

[0049] Step S104, based on the positional relationship between the target evaluation area and the obstacle, obtain a collision risk assessment result corresponding to the current sampling point.

[0050] For example, when the positional relationship between the obtained target evaluation area and the obstacle is that the obstacle is within the target evaluation area, the obtained collision risk assessment result corresponding to the current sampling point is that there is a collision risk, which is used to characterize that there is a risk of collision between the target object and the obstacle when the target object travels to the current sampling point; for another example, when the positional relationship between the obtained target evaluation area and the obstacle is that the obstacle is outside the target evaluation area, the obtained collision risk assessment result corresponding to the current sampling point is that there is no collision risk, which is used to characterize that there is no risk of collision between the target object and the obstacle when the target object travels to the current sampling point.

[0051] By using the collision risk assessment method provided in the embodiments of the present disclosure, after determining the current sampling point related to the target object, the target evaluation area corresponding to the target object when traveling to the current sampling point can be further determined, and based on the preset distance field, the positional relationship between the target evaluation area and the obstacle can be obtained. Then, based on the positional relationship between the target evaluation area and the obstacle, the collision risk assessment result corresponding to the current sampling point can be obtained. In this process, on the one hand, the preset distance field has a corresponding distance data set. Therefore, based on the preset distance field, the interval distance between the first target point representing the target evaluation area and the second target point representing the nearest obstacle can be quickly obtained as the first reference distance. And based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained, and then based on the positional relationship between the target evaluation area and the obstacle, the collision risk assessment result corresponding to the current sampling point can be obtained, thereby improving the acquisition efficiency of the collision risk assessment result; on the other hand, when determining the target evaluation area, only the area determination principle needs to be followed, that is, "ensuring that at least part of the target evaluation area is within the safety surrounding area corresponding to the target object when traveling to the current sampling point, and ensuring that the boundary distance between this at least part of the area within the safety surrounding area and the boundary of the safety surrounding area meets the preset safety distance requirement". In this way, the rationality of the target evaluation area can be greatly improved, that is, it can be avoided that the target evaluation area is too small to result in insufficient coverage of the safety surrounding area, and at the same time, it can be avoided that the target evaluation area is too large to result in unreasonable large-scale coverage of the safety surrounding area, thereby ensuring the accuracy of the collision risk assessment result.

[0052] In some alternative embodiments, step S101, that is, "determine the current sampling point related to the target object" may include:

[0053] When the driving planned path of the target object is a straight-line path, determine the current sampling point related to the target object from multiple candidate sampling points according to the preset traversal order.

[0054] Among them, multiple candidate sampling points can be obtained by sampling points on the driving planning path at a first sampling interval greater than the conventional sampling interval. Here, the conventional sampling interval can be set according to specific application requirements. For example, it can be set to 0.2 meters (m), and the embodiments of the present disclosure do not limit this. Based on this, in the embodiments of the present disclosure, the first sampling interval can be:

[0055] Lmin < L ≤ Lmax

[0056] Among them, L is used to represent the first sampling interval; Lmin is used to represent the conventional sampling interval; Lmax is used to represent the length of the target object, for example, 4m. It can be understood that in the embodiments of the present disclosure, when the first sampling interval is equal to the length of the target object, the spatial overlap degree of any two adjacent candidate sampling points is exactly 0m. Therefore, the acquisition efficiency of the collision risk assessment result can be further improved when the driving planning path is a straight path.

[0057] In addition, it should be noted that in the embodiments of the present disclosure, the preset traversal order can be the order from near to far relative to the target object. Please combine Figure 2 , for example, sampling points on the driving planning path at the first sampling interval, and the multiple candidate sampling points obtained include candidate sampling point A1, candidate sampling point A2, candidate sampling point A3, candidate sampling point A4, and candidate sampling point A5. Then, according to the preset traversal order, candidate sampling point A1, candidate sampling point A2, candidate sampling point A3, candidate sampling point A4, and candidate sampling point A5 can be sequentially determined as the current sampling points related to the target object.

[0058] In the above optional implementation manner, step S101, that is, "determine the current sampling point related to the target object" may also include:

[0059] When the driving planning path of the target object is a curved path, determine the current sampling point related to the target object from multiple candidate sampling points according to the breadth-first search (BFS) algorithm.

[0060] Among them, multiple candidate sampling points can be obtained by sampling points on the driving planning path at a second sampling interval determined based on the curvature parameter of the driving planning path. For example, the second sampling interval can be negatively correlated with the curvature parameter of the driving planning path, that is, the larger the curvature parameter of the driving planning path, the smaller a value is set as the second sampling interval; the smaller the curvature parameter of the driving planning path, the larger a value is set as the second sampling interval.

[0061] Please combine Figure 3, Exemplarily, according to the second sampling interval, point sampling is performed on the driving planned path, and the obtained multiple candidate sampling points include candidate sampling point B1, candidate sampling point B2, candidate sampling point B3, candidate sampling point B4, and candidate sampling point B5. Then, according to the BFS algorithm, candidate sampling point B1, candidate sampling point B5, candidate sampling point B3, candidate sampling point B2, and candidate sampling point B4 can be sequentially determined as the current sampling points related to the target object.

[0062] In the above manner, in the embodiments of the present disclosure, not only can a reasonable number of candidate sampling points be determined by optimizing the sampling interval to reduce the computational burden of the autonomous driving system, thereby achieving the purpose of further improving the acquisition efficiency of the collision risk assessment result, but also when the driving planned path of the target object is a curved path, according to the BFS algorithm, the current sampling points related to the target object can be determined from multiple candidate sampling points, that is, by means of layer-by-layer search, the current sampling points related to the target object are determined from multiple candidate sampling points to further reduce the computational burden of the autonomous driving system. In this way, the purpose of further improving the acquisition efficiency of the collision risk assessment result can also be achieved.

[0063] In an optional implementation manner, step S102, that is, "determine the target evaluation area corresponding to the target object when it travels to the current sampling point" may include:

[0064] Construct a plurality of circular evaluation areas on the safety surrounding area;

[0065] Based on the plurality of circular evaluation areas, determine the target evaluation area.

[0066] Among them, at least some of the plurality of circular evaluation areas are located within the safety surrounding area, and the boundary distance from the safety surrounding area (here, specifically, the maximum boundary distance) meets the preset safety distance requirement, that is, at least some of the plurality of circular evaluation areas are located within the safety surrounding area, and the boundary distance from the safety surrounding area is less than or equal to the preset distance threshold.

[0067] In an example, after constructing a plurality of circular evaluation areas on the safety surrounding area, each circular evaluation area in the plurality of circular evaluation areas can be used as the target evaluation area to obtain a plurality of target evaluation areas.

[0068] After that, when performing step S103, the positional relationship between each target evaluation area among the multiple target evaluation areas and the obstacle can be obtained based on the preset distance field. Specifically, for each target evaluation area among the multiple target evaluation areas, the center point of the target evaluation area can be used as the first target point position for representing the target evaluation area, and the interval distance between the first target point position and the second target point position for representing the nearest obstacle can be obtained as the first reference distance. Then, based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained. For example, when the first reference distance is less than or equal to the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located within the target evaluation area; or, when the first reference distance is greater than or equal to the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located outside the target evaluation area. Among them, the second reference distance can be the distance value from the first target point position to the boundary of the target evaluation area, specifically, it can be the radius value of the target evaluation area. In this way, the reliability of the first reference distance and the second reference distance can be ensured, so as to accurately obtain the positional relationship between each target evaluation area among the multiple target evaluation areas and the obstacle, thereby further improving the accuracy of the collision risk assessment result.

[0069] In the above example, finally, a plurality of positional relationships corresponding one by one to the multiple target evaluation areas can be obtained. After that, step S104 can be entered. And when performing step S104, when at least one of the multiple positional relationships is that the obstacle is located within the target evaluation area, the collision risk assessment result corresponding to the current sampling point can be determined to be at risk of collision, and the current sampling point can be determined as a risk sampling point; or, when each of the multiple positional relationships is that the obstacle is located outside the target evaluation area, the collision risk assessment result corresponding to the current sampling point can be determined to be not at risk of collision, and the current sampling point can be determined as a non-risk sampling point.

[0070] In another example, after constructing a plurality of circular evaluation areas on the safety surrounding area, the plurality of circular evaluation areas can be divided into a plurality of evaluation area groups, and each evaluation area group among the plurality of evaluation area groups includes at least one circular evaluation area. Then, according to a specific evaluation order, a target area group can be selected from the plurality of evaluation area groups, and each circular evaluation area in the target area group can be used as a target evaluation area to obtain at least one target evaluation area. Among them, the division method of the evaluation area group and the specific evaluation order can be set according to specific application requirements, and the embodiments of the present disclosure do not limit this.

[0071] Thereafter, when performing step S103, the positional relationship between each target evaluation area in at least one target evaluation area and the obstacle can be obtained based on a preset distance field. Specifically, for each target evaluation area in at least one target evaluation area, the center point of the target evaluation area can be used as the first target point for representing the target evaluation area, and the interval distance between the first target point and the second target point for representing the nearest obstacle can be obtained as the first reference distance. Then, based on the first reference distance, the positional relationship between the target evaluation area and the obstacle can be obtained. For example, when the first reference distance is less than or equal to the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located within the target evaluation area; or, when the first reference distance is greater than or equal to the second reference distance, the obtained positional relationship between the target evaluation area and the obstacle is that the obstacle is located outside the target evaluation area. Among them, the second reference distance can be the distance value from the first target point to the boundary of the target evaluation area, specifically, it can be the radius value of the target evaluation area. In this way, the reliability of the first reference distance and the second reference distance can be ensured, so as to accurately obtain the positional relationship between each target evaluation area in at least one target evaluation area and the obstacle, thereby further improving the accuracy of the collision risk assessment result.

[0072] In the above example, finally, at least one positional relationship corresponding one-to-one to at least one target evaluation area can be obtained. Thereafter, step S104 can be entered, and when performing step S104, when at least one of the at least one positional relationship is that the obstacle is located within the target evaluation area, the target area group can be determined as the risk area group; or, when each of the at least one positional relationship is that the obstacle is located outside the target evaluation area, the target area group can be determined as the non-risk area group. Then, when the target area group is determined as the risk area group, the collision risk assessment result corresponding to the current sampling point can be determined as having a collision risk, and the current sampling point can be determined as the risk sampling point; or, when the target area group is determined as the non-risk area group, according to a specific evaluation order, the next target area group can be selected from multiple evaluation area groups, and it can be determined whether the next target area group is the risk area group until after a certain evaluation area group among the multiple evaluation area groups is determined as the risk area group, the collision risk assessment result corresponding to the current sampling point is determined as having a collision risk, and the current sampling point is determined as the risk sampling point; or, until after each of the multiple evaluation area groups is determined as the non-risk area group, the collision risk assessment result corresponding to the current sampling point is determined as having no collision risk, and the current sampling point is determined as the non-risk sampling point.

[0073] Further, in the embodiments of the present disclosure, the multiple circular evaluation regions may include multiple inscribed circular regions and multiple corner circular regions. Based on this, in the embodiments of the present disclosure, "constructing multiple circular evaluation regions on the safety surrounding region" in step S102 may include:

[0074] Constructing multiple inscribed circular regions in the length direction of the safety surrounding region.

[0075] Among them, the multiple inscribed circular regions are located inside the safety surrounding region, and the distance from the first boundary of the safety surrounding region meets the preset safety distance requirement. Here, the first boundary distance may be the maximum boundary distance between the multiple inscribed circular regions and the safety surrounding region, that is, the distance value from the intersection point of any two adjacent inscribed circular regions in the multiple inscribed circular regions to the length boundary line of the safety surrounding region; the preset safety distance requirement may include that the first boundary distance is less than or equal to the first preset distance threshold, and the first preset distance threshold may be set according to specific application requirements. For example, it may be set to 8 centimeters (Cm), and the embodiments of the present disclosure do not limit this. Based on this, in the embodiments of the present disclosure, "constructing multiple inscribed circular regions in the length direction of the safety surrounding region" may include:

[0076] Obtaining a first radius value based on the overall region width of the safety surrounding region;

[0077] Using the first radius value and the first preset distance threshold to obtain a first interval distance;

[0078] Constructing multiple inscribed circular regions in the length direction of the safety surrounding region according to the first radius value and the first interval distance.

[0079] Among them, the first radius value may be the radius value of each inscribed circular region in the multiple inscribed circular regions. In one example, "obtaining a first radius value based on the overall region width of the safety surrounding region" may be: taking half of the overall region width of the safety surrounding region as the first radius value.

[0080] After obtaining the first radius value based on the overall region width of the safety surrounding region, the Pythagorean theorem can be applied to use the first radius value and the first preset distance threshold to obtain the first interval distance. Among them, the first interval distance may be the distance value between the center points of any two adjacent inscribed circular regions in the multiple inscribed circular regions in the length direction of the safety surrounding region. After obtaining the first radius value based on the overall region width of the safety surrounding region and using the first radius value and the first preset distance threshold to obtain the first interval distance, multiple inscribed circular regions can be constructed in the length direction of the safety surrounding region according to the first radius value and the first interval distance.

[0081] Please combine Figure 4, Exemplarily, the safety surrounding area is an octagon constructed around the target object with a certain distance (e.g., 15 Cm) from the target object. That is, the length boundary line of the safety surrounding area is separated from the width boundary line of the safety surrounding area, and they are connected by diagonal lines at the corners to form a chamfered area at the corner positions of the safety surrounding area.

[0082] When constructing multiple inscribed circular areas in the length direction of the safety surrounding area, half of the overall area width of the safety surrounding area ( Figure 4 denoted as 2r in Figure 4 can be used as the first radius value (

[0083]

[0084] denoted as r in

[0085]

[0086] That is, it serves as the radius value of each inscribed circular area among the multiple inscribed circular areas. By applying the Pythagorean theorem and using the first radius value and the first preset distance threshold, the first interval distance can be obtained. This process can be represented by the first calculation logic: max1 where S max1 is used to represent the first interval distance; r is used to represent the first radius value; e

[0087] is used to represent the first preset distance threshold.

[0088]

[0089] Based on the transformed first calculation logic, it can be understood that in the above example, as long as multiple inscribed circular areas are constructed in the length direction of the safety surrounding area according to the first radius value and any actual spacing less than or equal to the first interval distance, it can be ensured that the multiple inscribed circular areas are located within the safety surrounding area and the distance from the first boundary of the safety surrounding area meets the preset safety distance requirement. That is, it is ensured that the multiple inscribed circular areas are located within the safety surrounding area and the distance from the first boundary of the safety surrounding area is less than or equal to the first preset distance threshold. Moreover, when constructing multiple inscribed circular areas in the length direction of the safety surrounding area according to the first radius value and the first interval distance, it can be ensured to construct the minimum number of inscribed circular areas. Among them, the minimum number of inscribed circular areas can be obtained through the second calculation logic: 1 where n max1 is used to represent the minimum number of inscribed circular areas; Ceil is used to represent rounding up; S

[0090] In the above manner, in the embodiments of the present disclosure, a plurality of inscribed circular regions can be constructed in the length direction of the safety surrounding region. Moreover, when constructing a plurality of inscribed circular regions in the length direction of the safety surrounding region, based on the overall region width of the safety surrounding region, a first radius value can be obtained, and using the first radius value and a first preset distance threshold, a first interval distance can be obtained. Then, according to the first radius value and the first interval distance, a plurality of inscribed circular regions are constructed in the length direction of the safety surrounding region, so as to ensure that the plurality of inscribed circular regions are located within the safety surrounding region and the distance from the first boundary of the safety surrounding region meets the preset safety distance requirement, and on this basis, ensure that the minimum number of inscribed circular regions are constructed, so as to further reduce the computational burden of the autonomous driving system, thereby achieving the purpose of further improving the acquisition efficiency of the collision risk assessment result.

[0091] In the above embodiments, "constructing a plurality of circular evaluation regions on the safety surrounding region" in step S102 may further include:

[0092] Constructing a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively.

[0093] Specifically, a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively can be constructed at the four corner positions of the safety surrounding region respectively.

[0094] Among them, at least some of the plurality of corner circular regions are located within the safety surrounding region and the distance from the second boundary of the safety surrounding region meets the preset safety distance requirement. Here, the second boundary distance may be the maximum boundary distance between the plurality of corner circular regions and the safety surrounding region, that is, the distance value from the intersection point of any two adjacent corner circular regions in the plurality of corner circular regions to the length boundary line of the safety surrounding region; the preset safety distance requirement may include that the second boundary distance is less than or equal to a second preset distance threshold, and the second preset distance threshold can be set according to specific application requirements. For example, it can be set to 10 Cm, and the embodiments of the present disclosure do not limit this. Based on this, in the embodiments of the present disclosure, "constructing a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively" may include:

[0095] Obtaining a second radius value;

[0096] Using the second radius value and the second preset distance threshold, obtaining a second interval distance;

[0097] Based on the second radius value and the second interval distance, constructing a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively.

[0098] Among them, the second radius value can be the radius value of the corner circular region closest to the corner position of the safety surrounding region among multiple corner circular regions. In one example, the safety surrounding region is an octagon constructed around the target object with a certain distance interval from the target object. That is, the length boundary line of the safety surrounding region is separated from the width boundary line of the safety surrounding region and is connected by corner diagonal lines to form a chamfered region at the corner position of the safety surrounding region. Based on this, in the above example, the length value of the chamfered region can be obtained, and the second radius value can be obtained by using the length value of the chamfered region and the second preset distance threshold. For example, the Pythagorean theorem can be applied to obtain the second radius value by using the length value of the chamfered region and the second preset distance threshold.

[0099] After obtaining the second radius value, the Pythagorean theorem can be continuously applied to obtain the second interval distance by using the second radius value and the second preset distance threshold. Among them, the second interval distance can be the distance value between the center points of any two adjacent corner circular regions among multiple corner circular regions in the length direction of the safety surrounding region. After obtaining the second radius value and using the second radius value and the second preset distance threshold to obtain the second interval distance, multiple corner circular regions tangent to the length boundary line and the width boundary line of the safety surrounding region can be constructed based on the second radius value and the second interval distance.

[0100] Please combine Figure 5 and Figure 6 , Exemplarily, the safety surrounding region is an octagon constructed around the target object with a certain distance interval (for example, 15 Cm) from the target object. That is, the length boundary line of the safety surrounding region is separated from the width boundary line of the safety surrounding region and is connected by corner diagonal lines to form a chamfered region at the corner position of the safety surrounding region.

[0101] When constructing multiple corner circular regions tangent to the length boundary line and the width boundary line of the safety surrounding region, the length value of the chamfered region can be obtained, and the Pythagorean theorem can be applied to obtain the second radius value by using the length value of the chamfered region and the second preset distance threshold. This process can be represented by the third calculation logic:

[0102] r 1 2 =(r 1 -e max2 ) 2 +(r 1 -d) 2

[0103] Perform step-by-step transformation on the third calculation logic to obtain the transformed third calculation logic:

[0104] (r 1 -(d+e max2 ))2 = 2de max2

[0105]

[0106] where r 1 is used to represent the second radius value; d is used to represent the length value of the chamfered area; e max2 is used to represent the second preset distance threshold.

[0107] After obtaining the second radius value, the Pythagorean theorem can be continuously applied, and using the second radius value and the second preset distance threshold, the second interval distance can be obtained. This process can be represented by the fourth calculation logic:

[0108]

[0109] By gradually transforming the fourth calculation logic, the transformed fourth calculation logic is obtained:

[0110]

[0111] where S max2 is used to represent the second interval distance; r 1 is used to represent the second radius value; r 2 is used to represent the radius value of the other corner circular area adjacent to the corner circular area at the corner position closest to the safety surrounding area among multiple corner circular areas; e max2 is used to represent the second preset distance threshold.

[0112] Based on the transformed fourth calculation logic, it can be understood that in the above example, as long as multiple corner circular areas tangent to the length boundary line and width boundary line of the safety surrounding area are constructed based on the second radius value and any actual spacing less than or equal to the second interval distance, it can be ensured that at least some of the multiple corner circular areas are located within the safety surrounding area and the second boundary distance from the safety surrounding area meets the preset safety distance requirement, that is, it is ensured that at least some of the multiple corner circular areas are located within the safety surrounding area and the second boundary distance from the safety surrounding area is less than or equal to the second preset distance threshold, and moreover, when constructing multiple corner circular areas tangent to the length boundary line and width boundary line of the safety surrounding area based on the second radius value and the second interval distance, it can be ensured to construct the minimum number of corner circular areas. Among them, the minimum number of corner circular areas can be obtained through the fifth calculation logic:

[0113]

[0114] where n 2 is used to represent the minimum number of corner circular areas; Ceil is used to represent rounding up; Smax2 is used to characterize the second interval distance; w is used to characterize the overall area width of the safety surrounding area; r 1 is used to characterize the second radius value.

[0115] In the above way, in the embodiments of the present disclosure, a plurality of corner circular regions tangent to the length boundary line and the width boundary line of the safety surrounding area can be constructed. Moreover, when constructing a plurality of corner circular regions tangent to the length boundary line and the width boundary line of the safety surrounding area, the second radius value can be obtained, and the second interval distance can be obtained by using the second radius value and the second preset distance threshold. Then, based on the second radius value and the second interval distance, a plurality of corner circular regions tangent to the length boundary line and the width boundary line of the safety surrounding area are constructed, so as to ensure that at least part of the plurality of corner circular regions is located within the safety surrounding area and the second boundary distance from the safety surrounding area meets the preset safety distance requirement, and on this basis, ensure that the minimum number of corner circular regions is constructed, so as to further reduce the computational burden of the autonomous driving system, thereby achieving the purpose of further improving the acquisition efficiency of the collision risk assessment result.

[0116] Based on the above content, in the embodiments of the present disclosure, the plurality of circular evaluation regions include a plurality of inscribed circular regions and a plurality of corner circular regions. Therefore, the rationality of the target evaluation region can be improved to a greater extent, so as to further improve the accuracy of the collision risk assessment result. Moreover, based on the above content, in the embodiments of the present disclosure, it is possible to ensure that the minimum number of inscribed circular regions is constructed on the basis that the plurality of inscribed circular regions are located within the safety surrounding area and the first boundary distance from the safety surrounding area meets the preset safety distance requirement. At the same time, on the basis that at least part of the plurality of corner circular regions is located within the safety surrounding area and the second boundary distance from the safety surrounding area meets the preset safety distance requirement, ensure that the minimum number of corner circular regions is constructed. Finally, it is possible to construct the minimum number of circular evaluation regions (including a plurality of inscribed circular regions and a plurality of corner circular regions), specifically, it can be n (here, n = n 1 + 4n 2 ) circular evaluation regions, so as to achieve the purpose of further improving the acquisition efficiency of the collision risk assessment result.

[0117] In some alternative embodiments, before performing step S102, that is, "determining the target evaluation region corresponding to the target object when it travels to the current sampling point", the collision risk assessment method may further include:

[0118] Determining the pre-evaluation region corresponding to the target object when it travels to the current sampling point;

[0119] Based on the preset distance field, obtaining the positional relationship between the pre-evaluation region and the obstacle.

[0120] Among them, the coverage range of the pre-evaluation area over the safety surrounding area is greater than that of the target evaluation area over the safety surrounding area.

[0121] In one example, based on the distance field, the interval distance between the third target point representing the pre-evaluation area and the second target point representing the obstacle can be obtained as the third reference distance, and based on the third reference distance, the positional relationship between the pre-evaluation area and the obstacle can be obtained. Among them, the preset distance field can be a Manhattan distance field, ESDF, TSDF, etc. Here, the Manhattan distance field has a corresponding distance data set for storing the Manhattan distance from each point in the Manhattan distance field to other points. Therefore, based on the Manhattan distance field, the interval distance between the third target point representing the pre-evaluation area and the second target point representing the nearest obstacle (here, specifically, the Manhattan distance from the third target point to the second target point) can be obtained as the third reference distance, and based on the third reference distance, the positional relationship between the pre-evaluation area and the obstacle can be obtained; ESDF also has a corresponding distance data set for storing the nearest obstacle distance corresponding to each point in ESDF. Therefore, based on ESDF, the interval distance between the third target point representing the pre-evaluation area and the second target point representing the nearest obstacle (here, specifically, the nearest obstacle distance corresponding to the third target point) can be obtained as the third reference distance, and based on the third reference distance, the positional relationship between the pre-evaluation area and the obstacle can be obtained; for TSDF, since it can be converted into ESDF, when the distance field is TSDF, it can be converted into ESDF, and then based on ESDF, the positional relationship between the pre-evaluation area and the obstacle can be obtained.

[0122] In a specific example, "obtaining the positional relationship between the pre-evaluation area and the obstacle based on the third reference distance" may include: when the third reference distance is less than or equal to the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located within the pre-evaluation area; or, when the third reference distance is greater than the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located outside the pre-evaluation area. Among them, the fourth reference distance may be the distance value from the third target point to the boundary of the pre-evaluation area.

[0123] Furthermore, in the embodiments of the present disclosure, the pre-evaluation area may be the circumscribed circular area of the safety surrounding area, that is, the circumscribed circular area of the safety surrounding area can be constructed, and the circumscribed circular area is determined as the pre-evaluation area corresponding to the target object when it travels to the current sampling point.

[0124] Thereafter, when obtaining the positional relationship between the pre-evaluation area and the obstacle based on the preset distance field, the center point of the pre-evaluation area can be used as the third target point for representing the pre-evaluation area, and the interval distance between the third target point and the second target point for representing the nearest obstacle is obtained as the third reference distance. Then, based on the third reference distance, the positional relationship between the pre-evaluation area and the obstacle is obtained. For example, when the third reference distance is less than or equal to the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located within the pre-evaluation area; or, when the third reference distance is greater than or equal to the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located outside the pre-evaluation area. Among them, the fourth reference distance can be the distance value from the third target point to the boundary of the pre-evaluation area, specifically, the radius value of the pre-evaluation area. In this way, the reliability of the third reference distance and the fourth reference distance can be ensured, so as to accurately obtain the positional relationship between each pre-evaluation area and the obstacle among multiple pre-evaluation areas, thereby further improving the accuracy of the collision risk assessment result.

[0125] Thereafter, when the positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located within the pre-evaluation area, step S102 is entered, that is, "determine the target evaluation area corresponding to the target object when driving to the current sampling point"; when the positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located outside the pre-evaluation area, the collision risk assessment result corresponding to the current sampling point is determined to be no collision risk, and the current sampling point is determined as a non-risk sampling point.

[0126] Through the above method, in the embodiments of the present disclosure, after determining the current sampling point related to the target object, the pre-evaluation area corresponding to the target object when driving to the current sampling point can be determined first, and based on the preset distance field, the positional relationship between the pre-evaluation area and the obstacle is obtained, so that when the positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located outside the pre-evaluation area, the collision risk assessment result corresponding to the current sampling point is directly determined to be no collision risk, and the current sampling point is determined as a non-risk sampling point, without continuing to execute step S102, step S103, and step S104, thereby improving the acquisition efficiency of the collision risk assessment result in the safe driving scenario.

[0127] In some alternative embodiments, the collision risk assessment method may further include:

[0128] When there are risk sampling points among multiple candidate sampling points, the driving planned path is determined as a risk path;

[0129] Or, when there are no risk sampling points among multiple candidate sampling points, the driving planned path is determined as a safe path.

[0130] Among them, the collision risk assessment result corresponding to the risk sampling point indicates that there is a collision risk.

[0131] As described above, in the embodiments of the present disclosure, when the driving planned path of the target object is a straight path, the current sampling point related to the target object can be determined from multiple candidate sampling points according to a preset traversal order; or, when the driving planned path of the target object is a curved path, the current sampling point related to the target object can be determined from multiple candidate sampling points according to the BFS algorithm. After determining the current sampling point related to the target object from multiple candidate sampling points, a collision risk assessment result corresponding to the current sampling point will be obtained, which is used to characterize whether there is a risk of collision with an obstacle when the target object travels to the current sampling point. When the collision risk assessment result indicates that there is a risk of collision with an obstacle when the target object travels to the current sampling point, the current sampling point is determined as the risk sampling point, and it is determined that there is a risk sampling point among multiple candidate sampling points, and then the driving planned path is determined as the risk path; when the collision risk assessment result indicates that there is no risk of collision with an obstacle when the target object travels to the current sampling point, a new current sampling point related to the target object is determined from multiple candidate sampling points, and a collision risk assessment result corresponding to the new current sampling point is obtained, until after a certain candidate sampling point among multiple candidate sampling points is determined as the risk sampling point, it is determined that there is a risk sampling point among multiple candidate sampling points, and the driving planned path is determined as the risk path; or, until after each candidate sampling point among multiple candidate sampling points is determined as a non-risk sampling point, it is determined that there is no risk sampling point among multiple candidate sampling points, and the driving planned path is determined as a non-risk path.

[0132] Through the above method, the collision risk assessment method provided by the embodiments of the present disclosure can also be used to perform collision risk assessment on the driving planned path, thereby improving the applicable range of the collision risk assessment method.

[0133] Next, combined with Figure 7 , the complete process of the collision risk assessment method provided by the embodiments of the present disclosure will be described.

[0134] Step S701, perform point sampling on the driving planned path to obtain multiple candidate sampling points.

[0135] In one example, when the driving planned path of the target object is a straight path, the multiple candidate sampling points can be obtained by performing point sampling on the driving planned path at a first sampling interval greater than the conventional sampling interval; or, when the driving planned path of the target object is a curved path, the multiple candidate sampling points can be obtained by performing point sampling on the driving planned path at a second sampling interval determined based on the curvature parameter of the driving planned path.

[0136] Step S702: Determine the current sampling point related to the target object from the multiple candidate sampling points.

[0137] In one example, when the driving planned path of the target object is a straight path, the current sampling point related to the target object can be determined from the multiple candidate sampling points in a preset traversal order; or, when the driving planned path of the target object is a curved path, the current sampling point related to the target object can be determined from the multiple candidate sampling points according to the BFS algorithm.

[0138] Step S703: Determine the pre-evaluation area corresponding to the target object when driving to the current sampling point, and based on the preset distance field, obtain the positional relationship between the pre-evaluation area and the obstacle.

[0139] In one example, the pre-evaluation area can be the circumscribed circular area of the safety surrounding area, that is, the circumscribed circular area of the safety surrounding area can be constructed and determined as the pre-evaluation area corresponding to the target object when driving to the current sampling point, and then based on the preset distance field, the positional relationship between the pre-evaluation area and the obstacle is obtained. Specifically, when obtaining the positional relationship between the pre-evaluation area and the obstacle based on the preset distance field, the center point of the pre-evaluation area can be used as the third target point representing the pre-evaluation area, and the interval distance between the third target point and the second target point representing the nearest obstacle is obtained as the third reference distance, and then based on the third reference distance, the positional relationship between the pre-evaluation area and the obstacle is obtained. For example, when the third reference distance is less than or equal to the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located within the pre-evaluation area; or, when the third reference distance is greater than or equal to the fourth reference distance, the obtained positional relationship between the pre-evaluation area and the obstacle is that the obstacle is located outside the pre-evaluation area. Among them, the fourth reference distance can be the distance value from the third target point to the boundary of the pre-evaluation area, specifically, the radius value of the pre-evaluation area.

[0140] When the positional relationship between the pre-evaluation area and the obstacle is such that the obstacle is within the pre-evaluation area, step S704 is entered; when the positional relationship between the pre-evaluation area and the obstacle is such that the obstacle is outside the pre-evaluation area and there are remaining sampling points among the multiple candidate sampling points that have not been selected as the current sampling points related to the target object, step S702 is returned.

[0141] Step S704: Determine the target evaluation area corresponding to the target object when it travels to the current sampling point, obtain the positional relationship between the target evaluation area and the obstacle based on the preset distance field, and then obtain the collision risk assessment result corresponding to the current sampling point based on the positional relationship between the target evaluation area and the obstacle.

[0142] In one example, multiple circular evaluation areas can be constructed on the safety surrounding area, and the target evaluation area is determined based on the multiple circular evaluation areas. Among them, at least some of the multiple circular evaluation areas are within the safety surrounding area and the boundary distance (here, specifically, it can be the maximum boundary distance) from the safety surrounding area meets the preset safety distance requirement, that is, at least some of the multiple circular evaluation areas are within the safety surrounding area and the boundary distance from the safety surrounding area is less than or equal to the preset distance threshold.

[0143] In a specific example, the multiple circular evaluation areas can include multiple inscribed circular areas and multiple corner circular areas. That is, "constructing multiple circular evaluation areas on the safety surrounding area" can include:

[0144] Construct multiple inscribed circular areas in the length direction of the safety surrounding area.

[0145] Among them, the multiple inscribed circular areas are within the safety surrounding area and the first boundary distance from the safety surrounding area meets the preset safety distance requirement. Here, the first boundary distance can be the maximum boundary distance between the multiple inscribed circular areas and the safety surrounding area, that is, the distance value from the intersection point of any two adjacent inscribed circular areas among the multiple inscribed circular areas to the length boundary line of the safety surrounding area; the preset safety distance requirement can include that the first boundary distance is less than or equal to the first preset distance threshold, and the first preset distance threshold can be set according to specific application requirements. For example, it can be set to 8 Cm, and the embodiments of the present disclosure do not limit this.

[0146] "Constructing multiple circular evaluation areas on the safety surrounding area" can also include:

[0147] Construct multiple corner circular areas that are tangent to the length boundary line and the width boundary line of the safety surrounding area respectively.

[0148] Among them, at least some of the multiple corner circular regions are located within the safety surrounding region, and the distance from the second boundary of the safety surrounding region meets the preset safety distance requirement. Here, the second boundary distance can be the maximum boundary distance between the multiple corner circular regions and the safety surrounding region, that is, the distance value from the intersection point of any two adjacent corner circular regions among the multiple corner circular regions to the length boundary line of the safety surrounding region; the preset safety distance requirement can include that the second boundary distance is less than or equal to the second preset distance threshold, and the second preset distance threshold can be set according to specific application requirements. For example, it can be set to 10 Cm, and the embodiments of the present disclosure do not limit this.

[0149] In a specific example, "determining the target evaluation region based on multiple circular evaluation regions" can be: dividing the multiple circular evaluation regions into multiple evaluation region groups, and each evaluation region group in the multiple evaluation region groups includes at least one circular evaluation region, and then selecting a target region group from the multiple evaluation region groups according to a specific evaluation order, and then taking each circular evaluation region in the target region group as the target evaluation region to obtain at least one target evaluation region. Among them, the division method of the evaluation region groups and the specific evaluation order can be set according to specific application requirements, and the embodiments of the present disclosure do not limit this.

[0150] Thereafter, based on the preset distance field, the positional relationship between each target evaluation region in at least one target evaluation region and the obstacle can be obtained. Specifically, for each target evaluation region in at least one target evaluation region, the center point of the target evaluation region can be used as the first target point for representing the target evaluation region, and the interval distance between the first target point and the second target point for representing the nearest obstacle can be obtained as the first reference distance, and then based on the first reference distance, the positional relationship between the target evaluation region and the obstacle can be obtained. For example, when the first reference distance is less than or equal to the second reference distance, the obtained positional relationship between the target evaluation region and the obstacle is that the obstacle is within the target evaluation region; or, when the first reference distance is greater than or equal to the second reference distance, the obtained positional relationship between the target evaluation region and the obstacle is that the obstacle is outside the target evaluation region. Among them, the second reference distance can be the distance value from the first target point to the boundary of the target evaluation region, specifically, it can be the radius value of the target evaluation region.

[0151] In the above examples, finally, at least one positional relationship corresponding one-to-one to at least one target evaluation area can be obtained. Thereafter, when at least one of the at least one positional relationships is such that the obstacle is within the target evaluation area, the target area group can be determined as a risk area group; or, when each of the at least one positional relationships is such that the obstacle is outside the target evaluation area, the target area group can be determined as a non-risk area group. Then, when the target area group is determined as a risk area group, the collision risk assessment result corresponding to the current sampling point can be determined as having a collision risk, and the current sampling point can be determined as a risk sampling point; or, when the target area group is determined as a non-risk area group and there are still remaining area groups among the multiple evaluation area groups that have not been selected as the target area group, the next target area group can be selected from the multiple evaluation area groups in a specific evaluation order, and it can be determined whether the next target area group is a risk area group, until after determining that a certain evaluation area group among the multiple evaluation area groups is a risk area group, the collision risk assessment result corresponding to the current sampling point is determined as having a collision risk, and the current sampling point is determined as a risk sampling point; or, until after determining that each of the multiple evaluation area groups is a non-risk area group, the collision risk assessment result corresponding to the current sampling point is determined as not having a collision risk, and the current sampling point is determined as a non-risk sampling point.

[0152] When the current sampling point is determined as a risk sampling point, it can be determined that there is a risk sampling point among the multiple candidate sampling points, and the driving planning path can be determined as a risk path; when the current sampling point is determined as a non-risk sampling point and there are still remaining sampling points among the multiple candidate sampling points that have not been selected as the current sampling point related to the target object, return to step S702; when the current sampling point is determined as a non-risk sampling point and there are no remaining sampling points among the multiple candidate sampling points that have not been selected as the current sampling point related to the target object, the driving planning path can be determined as a safe path.

[0153] For the specific functions and examples of the above steps, reference can be made to the relevant descriptions of the corresponding steps in the foregoing embodiments of the collision risk assessment method, which will not be elaborated here.

[0154] Please refer to Figure 8 , which is a schematic diagram of an application scenario of a collision risk assessment method provided by an embodiment of the present disclosure.

[0155] The collision risk assessment method provided by an embodiment of the present disclosure is applied to an electronic device. Among them, the electronic device can be an autonomous driving controller, and is used for:

[0156] Determine the current sampling point related to the target object;

[0157] Determine the current sampling point related to the target object;

[0158] Determine the target evaluation area corresponding to the target object when it travels to the current sampling point;

[0159] Based on the preset distance field, obtain the positional relationship between the target evaluation area and the obstacle;

[0160] Based on the positional relationship between the target evaluation area and the obstacle, obtain the collision risk assessment result corresponding to the current sampling point.

[0161] It should be noted that in the embodiments of the present disclosure, Figure 8 The schematic diagram of the application scenario shown is only illustrative and not restrictive. Those skilled in the art can make various obvious changes and / or substitutions based on Figure 8 the examples, and the obtained technical solutions still fall within the scope of the disclosure of the embodiments of the present disclosure.

[0162] To better implement the foregoing collision risk assessment method, the embodiments of the present disclosure further provide a collision risk assessment device, which can be integrated into an electronic device. Among them, the electronic device can be an autonomous driving controller. Hereinafter, with reference to Figure 9 the schematic structural block diagram shown, a collision risk assessment device 900 provided by the embodiments of the disclosure will be described.

[0163] The collision risk assessment device 900 includes:

[0164] A sampling point determination unit 901, configured to determine the current sampling point related to the target object;

[0165] An area determination unit 902, configured to determine the target evaluation area corresponding to the target object when it travels to the current sampling point;

[0166] A positional relationship acquisition unit 903, configured to obtain the positional relationship between the target evaluation area and the obstacle based on the preset distance field;

[0167] An evaluation result acquisition unit 904, configured to obtain the collision risk assessment result corresponding to the current sampling point based on the positional relationship between the target evaluation area and the obstacle.

[0168] In some alternative embodiments, at least part of the target evaluation area is located within the safety surrounding area corresponding to the target object when it travels to the current sampling point, and the distance from the boundary of the safety surrounding area meets the preset safety distance requirement.

[0169] In some alternative embodiments, the area determination unit 902 is configured to:

[0170] Construct a plurality of circular evaluation regions on the safety surrounding region; wherein, at least some of the plurality of circular evaluation regions are located within the safety surrounding region, and the distance from the boundary of the safety surrounding region meets the preset safety distance requirement;

[0171] Based on the plurality of circular evaluation regions, determine the target evaluation region.

[0172] In some alternative embodiments, the plurality of circular evaluation regions include a plurality of inscribed circular regions and a plurality of corner circular regions; the region determination unit 902 is configured to:

[0173] Construct a plurality of inscribed circular regions in the length direction of the safety surrounding region; wherein, the plurality of inscribed circular regions are located within the safety surrounding region, and the distance from the first boundary of the safety surrounding region meets the preset safety distance requirement;

[0174] Construct a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively; wherein, at least some of the plurality of corner circular regions are located within the safety surrounding region, and the distance from the second boundary of the safety surrounding region meets the preset safety distance requirement.

[0175] In some alternative embodiments, the preset safety distance requirement includes that the first boundary distance is less than or equal to the first preset distance threshold; the region determination unit 902 is configured to:

[0176] Based on the overall region width of the safety surrounding region, obtain a first radius value; wherein, the first radius value is the radius value of each of the plurality of inscribed circular regions to be constructed.

[0177] Using the first radius value and the first preset distance threshold, obtain a first interval distance;

[0178] Construct a plurality of inscribed circular regions in the length direction of the safety surrounding region according to the first radius value and the first interval distance.

[0179] In some alternative embodiments, the preset safety distance requirement includes that the second boundary distance is less than or equal to the second preset distance threshold; the region determination unit 902 is configured to:

[0180] Obtain a second radius value; wherein, the second radius value is the radius value of the corner circular region closest to the corner position of the safety surrounding region among the plurality of corner circular regions to be constructed.

[0181] Using the second radius value and the second preset distance threshold, obtain a second interval distance;

[0182] Based on the second radius value and the second interval distance, construct a plurality of corner circular regions that are tangent to the length boundary line and the width boundary line of the safety surrounding region respectively.

[0183] In some alternative embodiments, the length boundary line of the safety surrounding area is separated from the width boundary line of the safety surrounding area and is connected by a corner diagonal line to form a chamfered area at the corner position of the safety surrounding area; the area determination unit 902 is configured to:

[0184] Obtain the length value of the chamfered area;

[0185] Use the length value of the chamfered area and the second preset distance threshold to obtain a second radius value.

[0186] In some alternative embodiments, the sampling point determination unit 901 is configured to:

[0187] When the driving planned path of the target object is a straight path, determine the current sampling point related to the target object from multiple candidate sampling points according to a preset traversal order; wherein, the multiple candidate sampling points are obtained by sampling points on the driving planned path according to a first sampling interval greater than the conventional sampling interval;

[0188] Alternatively, when the driving planned path of the target object is a curved path, determine the current sampling point related to the target object from multiple candidate sampling points according to the breadth-first search algorithm; wherein, the multiple candidate sampling points are obtained by sampling points on the driving planned path according to a second sampling interval determined based on the curvature parameter of the driving planned path.

[0189] In some alternative embodiments, the collision risk assessment further includes a path assessment unit, configured to:

[0190] When there is a risk sampling point among the multiple candidate sampling points, determine the driving planned path as a risk path; wherein, the collision risk assessment result corresponding to the risk sampling point is that there is a collision risk;

[0191] Alternatively, when there is no risk sampling point among the multiple candidate sampling points, determine the driving planned path as a safe path.

[0192] In some alternative embodiments, the collision risk assessment device 900 further includes a pre-assessment unit, configured to:

[0193] Determine the pre-assessment area corresponding to the target object when driving to the current sampling point; wherein, the coverage range of the pre-assessment area on the safety surrounding area is greater than the coverage range of the target assessment area on the safety surrounding area;

[0194] Based on a preset distance field, obtain the positional relationship between the pre-assessment area and the obstacle;

[0195] The area determination unit 902, is configured to:

[0196] When the position relationship between the pre-evaluation area and the obstacle characterizes that the obstacle is within the pre-evaluation area, determine the target evaluation area corresponding to the target object when it travels to the current sampling point.

[0197] In some alternative embodiments, the pre-evaluation unit is configured to:

[0198] Construct a circumscribed circular area of the safety surrounding area;

[0199] Determine the circumscribed circular area as the pre-evaluation area corresponding to the target object when it travels to the current sampling point.

[0200] In the embodiments of the present disclosure, for the specific functions and examples of each unit in the collision risk assessment device 900, reference may be made to the relevant descriptions of the corresponding steps in the foregoing embodiments of the collision risk assessment method, which will not be elaborated herein.

[0201] In the technical solution of the present disclosure, the acquisition, storage, application, etc. of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0202] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, an autonomous vehicle, a readable storage medium, and a computer program product.

[0203] Figure 10 FIG. shows a schematic structural block diagram of an exemplary electronic device 1000 that can be used to implement the embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as in-vehicle computing devices, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0204] Such as Figure 10As shown, the electronic device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to computer programs stored in a read-only memory (ROM) 1002 or computer programs loaded from a storage unit 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the electronic device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0205] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of renderers, speakers, etc.; a storage unit 1008, such as a disk, an optical disc, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0206] The computing unit 1001 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 1001 executes the various methods and processes described above, such as the collision risk assessment method. For example, in some embodiments, the collision risk assessment method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the collision risk assessment method described above can be executed. Alternatively, in other embodiments, the computing unit 1001 can be configured as the collision risk assessment method in any other appropriate manner (e.g., by means of firmware).

[0207] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on chip (SOC) systems, complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0208] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine, or entirely on the remote machine or server.

[0209] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a RAM, a ROM, an erasable programmable read-only memory (EPROM) or flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0210] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a rendering device for rendering information to the user (e.g., a cathode ray tube (CRT) renderer or a liquid crystal display (LCD) renderer); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0211] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0212] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0213] An embodiment of the present disclosure also provides an autonomous vehicle, including an electronic device.

[0214] An embodiment of the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute a collision risk assessment method.

[0215] An embodiment of the present disclosure also provides a computer program product, including a computer program which, when executed by a processor, implements a collision risk assessment method.

[0216] It should be understood that various forms of the processes shown above may be used, steps may be reordered, added or deleted. For example, the steps recited in the present disclosure may be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and this is not limited herein. In addition, in the present disclosure, relational terms such as "first", "second", "third", etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In addition, "a plurality" in the present disclosure may be understood as at least two.

[0217] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A collision risk assessment method, comprising: determining a current sampling point relative to a target object; Determining a target evaluation area corresponding to the target object when the target object travels to the current sampling point; Based on the preset distance field, obtaining the positional relationship between the target evaluation area and the obstacle; Based on the positional relationship between the target assessment area and the obstacle, a collision risk assessment result corresponding to the current sampling point is obtained.

2. The method according to claim 1, wherein: At least a portion of the target assessment area is located within a safety encirclement area corresponding to the target object when the target object travels to the current sampling point, and a distance from a boundary of the safety encirclement area meets a preset safety distance requirement.

3. The method according to claim 2, wherein: The determining the target evaluation area corresponding to the target object when the target object travels to the current sampling point includes: Constructing a plurality of circular evaluation areas on the safety encirclement area; wherein at least some of the plurality of circular evaluation areas are located within the safety encirclement area and the distance from the boundary of the safety encirclement area meets the preset safety distance requirement; Based on the plurality of circular evaluation areas, a target evaluation area is determined.

4. The method according to claim 3, wherein: The plurality of circular evaluation areas include a plurality of inscribed circular areas and a plurality of corner circular areas; The step of constructing a plurality of circular evaluation areas on the safety encirclement area comprises: Constructing the plurality of inscribed circular areas in the length direction of the safety encirclement area; wherein the plurality of inscribed circular areas are located within the safety encirclement area, and the distances from the first boundary of the safety encirclement area meet the preset safety distance requirement; Construct the multiple corner circular areas that are tangent to the length boundary line and the width boundary line of the safety encirclement area respectively; wherein at least part of the multiple corner circular areas are located within the safety encirclement area, and the distance from the second boundary of the safety encirclement area meets the preset safety distance requirement.

5. The method according to claim 4, wherein: The preset safety distance requirement includes that the first boundary distance is less than or equal to a first preset distance threshold; The step of constructing a plurality of inscribed circular areas in the length direction of the safety encirclement area comprises: Based on the overall area width of the safe encirclement area, a first radius value is obtained; wherein the first radius value is the radius value of each inscribed circular area among the multiple inscribed circular areas to be constructed; Obtaining a first interval distance using the first radius value and the first preset distance threshold; According to the first radius value and the first spacing distance, a plurality of inscribed circular areas are constructed in the length direction of the safety encirclement area.

6. The method according to claim 4, wherein: The preset safety distance requirement includes that the second boundary distance is less than or equal to a second preset distance threshold; The step of constructing a plurality of circular corner areas that are tangent to the length boundary line and the width boundary line of the safety encirclement area respectively includes: Obtain a second radius value; wherein the second radius value is the radius value of the corner circular area that is closest to the corner position of the safe encirclement area among the multiple corner circular areas to be constructed; Using the second radius value and the second preset distance threshold, a second interval distance is obtained; Based on the second radius value and the second spacing distance, a plurality of corner circular areas are constructed which are tangent to the length boundary line and the width boundary line of the safety enclosing area respectively.

7. The method according to claim 6, wherein: The length boundary line of the safety encirclement area is separated from the width boundary line of the safety encirclement area and connected by a corner oblique line to form a corner cutting area at the corner position of the safety encirclement area; The obtaining of the second radius value comprises: Obtaining the length value of the corner cutting area; A second radius value is obtained by using the length value of the corner cutting area and the second preset distance threshold.

8. The method according to claim 1, wherein: The determining of the current sampling point related to the target object comprises: When the planned driving path of the target object is a straight line path, determine a current sampling point related to the target object from a plurality of candidate sampling points according to a preset traversal order; wherein the plurality of candidate sampling points are obtained by performing point sampling on the planned driving path according to a first sampling interval greater than a conventional sampling interval; Alternatively, when the planned driving path of the target object is a curved path, a current sampling point related to the target object is determined from multiple candidate sampling points according to a breadth-first search algorithm; wherein the multiple candidate sampling points are obtained by performing point sampling on the planned driving path according to a second sampling spacing determined based on a curvature parameter of the planned driving path.

9. The method according to claim 8, further comprising: In the case where there is a risk sampling point among the multiple candidate sampling points, determining the driving plan path as a risk path; wherein the collision risk assessment result corresponding to the risk sampling point is that there is a collision risk; Alternatively, when there is no risky sampling point among the multiple candidate sampling points, the planned driving path is determined as a safe path.

10. The method according to any one of claims 2 to 7, further comprising: Determine a pre-evaluation area corresponding to the target object when it travels to the current sampling point; wherein the coverage of the pre-evaluation area to the safety encirclement area is greater than the coverage of the target evaluation area to the safety encirclement area; Based on the preset distance field, obtaining a positional relationship between the pre-evaluation area and the obstacle; The determining the target evaluation area corresponding to the target object when the target object travels to the current sampling point includes: In a case where the positional relationship between the pre-evaluation area and the obstacle indicates that the obstacle is located within the pre-evaluation area, a target evaluation area corresponding to the target object when the target object travels to the current sampling point is determined.

11. The method according to claim 10, wherein: The determining of the pre-evaluation area corresponding to the target object when the target object travels to the current sampling point includes: Constructing a circumscribed circular area of ​​the safe encirclement area; The circumscribed circular area is determined as a pre-evaluation area corresponding to the target object when the target object travels to the current sampling point.

12. A collision risk assessment device, comprising: A sampling point determination unit, used to determine a current sampling point associated with a target object; An area determination unit, used to determine a target evaluation area corresponding to the target object when the target object travels to the current sampling point; A position relationship acquisition unit, used to obtain the position relationship between the target evaluation area and the obstacle based on a preset distance field; An assessment result acquisition unit is used to obtain a collision risk assessment result corresponding to the current sampling point based on a positional relationship between the target assessment area and the obstacle.

13. An electronic device comprising: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 11.

14. An autonomous driving vehicle comprising the electronic device described in claim 13.

15. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 11.

16. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 11.

Images