CN116424315A

CN116424315A - Collision detection method, collision detection device, electronic equipment, automatic driving vehicle and medium

Info

Publication number: CN116424315A
Application number: CN202310337863.7A
Authority: CN
Inventors: 王丕阁
Original assignee: Apollo Zhilian Beijing Technology Co Ltd
Current assignee: Apollo Zhilian Beijing Technology Co Ltd
Priority date: 2023-03-31
Filing date: 2023-03-31
Publication date: 2023-07-14

Abstract

The disclosure relates to the field of computers, in particular to the technical fields of automatic driving, intelligent transportation and the like, and specifically relates to a collision detection method, a collision detection device, electronic equipment, an automatic driving vehicle and a medium. The specific implementation scheme is as follows: determining a current collision detection point of the target object; determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and the obstacle based on a pre-constructed distance field; and obtaining a collision detection result of the target object on the current collision detection point according to the position relation. By adopting the method and the device, the collision detection efficiency of the target object can be improved.

Description

Collision detection method, collision detection device, electronic equipment, automatic driving vehicle and medium

Technical Field

The disclosure relates to the field of computers, in particular to the technical fields of automatic driving, intelligent transportation and the like, and specifically relates to a collision detection method, a collision detection device, electronic equipment, an automatic driving vehicle and a medium.

Background

With the rise of automatic driving technology, the problems of driving planning, parking planning and the like which take an automatic driving vehicle as a research object are more and more emphasized, and the problems are mainly related to the collision detection technology. At present, collision detection is generally realized by adopting a grid space coverage enumeration method, a geometric boundary collision detection method and other technologies, and the detection efficiency is low.

Disclosure of Invention

The disclosure provides a collision detection method, a collision detection device, electronic equipment, an automatic driving vehicle and a medium.

According to an aspect of the present disclosure, there is provided a collision detection method including:

determining a current collision detection point of the target object;

determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and an obstacle based on a pre-constructed distance field;

and obtaining a collision detection result of the target object on the current collision detection point according to the position relation.

According to a second aspect of the present disclosure, there is provided a collision detection apparatus including:

a detection point determining unit configured to determine a current collision detection point of the target object;

a detection unit, configured to determine a positional relationship between a collision detection area corresponding to the target object when the target object travels to the current collision detection point and an obstacle, based on a pre-constructed distance field;

and a result acquisition unit for acquiring a collision detection result of the target object at the current collision detection point according to the position relation.

According to a third aspect of the present disclosure, there is provided an electronic device comprising:

At least one processor;

a memory communicatively coupled to the at least one processor;

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method provided in the first aspect.

According to a fourth aspect of the present disclosure, there is provided an autonomous vehicle comprising the electronic device provided by the third aspect.

According to a fifth aspect of the present disclosure there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method provided by the first aspect.

According to a sixth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method provided by the first aspect.

By adopting the method and the device, the collision detection efficiency of the target object can be improved.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

fig. 1 is a schematic flow chart of a collision detection method according to an embodiment of the disclosure;

fig. 2A and fig. 2B are schematic diagrams of a method for constructing a circular filling area according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of another method for constructing a circular filling area according to an embodiment of the present disclosure;

fig. 4A, fig. 4B, and fig. 4C are schematic diagrams of a distributed construction method of a circular filling area according to an embodiment of the present disclosure;

fig. 5 is a schematic diagram of a method for constructing a circular surrounding area according to an embodiment of the disclosure;

fig. 6 is a schematic diagram of a detection point sampling method according to an embodiment of the disclosure;

fig. 7 is a schematic diagram of a detection point sampling method according to an embodiment of the disclosure;

fig. 8 is a complete flow chart of a collision detection method according to an embodiment of the disclosure;

fig. 9 is a schematic application scenario diagram of a collision detection method according to an embodiment of the present disclosure;

FIG. 10 is a schematic block diagram of a collision detection apparatus provided in an embodiment of the present disclosure;

fig. 11 is a schematic block diagram of an electronic device according to an embodiment of the disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

As described in the background art, for collision detection, at present, a grid space coverage enumeration method, a geometric boundary collision detection method and other technologies are generally adopted for implementation.

The grid space coverage enumeration method is to disperse the running space of the target object into grids, and the grids where the obstacle is located are marked as occupied states. Then, by enumerating the grids corresponding to the target object, whether the grids contain the grids in the occupied state is judged to realize collision detection. The detection efficiency of the method is greatly influenced by the grid resolution, but the grid resolution cannot be greatly improved based on a conventional low-calculation-force platform. Therefore, the detection efficiency is low.

Geometric boundary collision detection is by constructing the target object and the obstacle into specific geometric shapes, for example, circles, straight lines, triangles, rectangles, convex polygons, or the like, respectively. Then, whether the geometric shape corresponding to the target object and the geometric shape corresponding to the obstacle are overlapped is judged to realize collision detection. This method is greatly affected by the complexity of the obstacle, and in the case of higher complexity of the obstacle, more geometric shapes are usually required to fully cover the obstacle, resulting in a significant decrease in detection efficiency.

Based on this, the embodiment of the present disclosure provides a collision detection method that can be applied to an electronic apparatus. In the following, a collision detection method provided in an embodiment of the present disclosure will be described with reference to a flowchart shown in fig. 1. It should be noted that although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in other orders.

Step S101, determining a current collision detection point of a target object;

step S102, determining the position relationship between a collision detection area corresponding to a target object when the target object runs to a current collision detection point and an obstacle based on a pre-constructed distance field;

step S103, obtaining a collision detection result of the target object on the current collision detection point according to the position relation.

Wherein the target object may be an autonomous vehicle, the vehicle type (model, size, brand) of which is not particularly limited by the disclosed embodiments.

In the embodiment of the disclosure, the current collision detection point may be a collision detection point determined by performing detection point sampling on a driving planning path according to a preset sampling strategy after determining the driving planning path of the target object. Based on this, it can be appreciated that in the embodiment of the present disclosure, the current collision detection point has corresponding pose information to characterize the position and the vehicle body pose corresponding to when the target object travels to the current collision detection point. Wherein the body attitude may be a body orientation. Then, after determining the current collision detection point of the target object, as a specific example of the embodiment of the present disclosure, a collision detection area to which the target object corresponds when traveling to the current collision detection point may be determined based on pose information to which the current collision detection point corresponds. The collision detection area may include a plurality of circular filling areas for filling an area occupied by the target object when traveling to the current collision detection point.

Further, in the disclosed embodiments, the distance field may be a Euclidean distance field (Euclidean Signed Distance Field, ESDF), a truncated signed distance function field (Truncated Signed Distance Function, TSDF), a Manhattan distance field, or the like. For an ESDF, it has a corresponding distance dataset for storing the nearest obstacle distance corresponding to each location point within the ESDF to characterize the separation distance between that location point and its nearest obstacle. For the TSDF, because it can be mutually converted with the ESDF, in the embodiment of the present disclosure, when the distance field is the TSDF, the TSDF may be converted into the ESDF, and then based on the ESDF, the positional relationship between the obstacle and the collision detection area corresponding to the target object when the target object travels to the current collision detection point is determined. For a Manhattan distance field, it also has a corresponding distance dataset for storing the nearest obstacle distance for each location point within the Manhattan distance field to characterize the Manhattan distance between the location point and the target volume. Wherein the target may be its nearest obstacle.

Based on the above description, in the embodiments of the present disclosure, when it is determined that a driving obstacle region exists in a plurality of circular filling regions based on a distance field, a positional relationship between a collision detection region corresponding to a target object when driving to a current collision detection point and an obstacle may be determined as an inclusion relationship, so as to represent that the obstacle is located inside the collision detection region, thereby determining that a collision detection result of the target object at the current collision detection point is a collision risk, otherwise, determining that a positional relationship between the collision detection region corresponding to the target object when driving to the current collision detection point and the obstacle is a non-inclusion relationship, so as to represent that the obstacle is located outside the collision detection region, and thus determining that a collision detection result of the target object at the current collision detection point is a collision risk. Wherein "determining that there is a travel obstacle region in the plurality of circular filling regions based on the distance field" may include: for each circular filling area, determining a target circle center position of the circular filling area in a distance field; inquiring a target obstacle distance corresponding to the position of the center of a target circle from a distance data set of a distance field; in the case where the target obstacle distance is smaller than the radius length of the circular filling area, the circular filling area is taken as a travel obstacle area to determine that the travel obstacle area exists among the plurality of circular filling areas.

The collision detection method provided by the embodiment of the disclosure can determine the current collision detection point of the target object; determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and the obstacle based on a pre-constructed distance field; according to the position relation, a collision detection result of the target object at the current collision detection point is obtained. Because the distance field is provided with a corresponding distance data set and is used for storing the nearest obstacle distance corresponding to each position point in the field, the position relationship between the corresponding collision detection area and the obstacle when the target object runs to the current collision detection point can be rapidly determined based on the distance field, and then the collision detection result of the target object on the current collision detection point is obtained according to the position relationship, so that the collision detection efficiency of the target object is improved.

In some alternative embodiments, "determining the positional relationship between the collision detection region and the obstacle, to which the target object corresponds when traveling to the current collision detection point, based on the pre-constructed distance field" may include the steps of:

constructing a rectangular surrounding area corresponding to the target object when the target object runs to the current collision detection point;

Constructing a plurality of circular filling areas on the rectangular surrounding area; wherein the collision detection region includes a plurality of circular filling regions;

in the case that a driving obstacle region including an obstacle exists in the plurality of circular filling regions based on the distance field, determining that the position relationship between the collision detection region corresponding to the target object when driving to the current collision detection point and the obstacle is an inclusion relationship so as to represent that the obstacle is positioned inside the collision detection region.

In the embodiment of the disclosure, based on pose information corresponding to the current collision detection point, a minimum circumscribed rectangle corresponding to the target object when the target object travels to the current collision detection point can be constructed and used as a rectangle surrounding area to represent an area occupied by the target object when the target object travels to the current collision detection point. After constructing the rectangular bounding region, a plurality of circular fill regions may be constructed over the rectangular bounding region such that each circular fill region intersects (including being tangent to) a side of the rectangular bounding region.

Referring to fig. 2A, for the construction of a plurality of circular filling regions, as a specific example of the embodiment of the present disclosure, a plurality of first inscribed circle filling regions 202 may be constructed along four sides of the rectangular surrounding region 201, and the diameter length of the first inscribed circle filling regions 202 is smaller than the width of the rectangular surrounding region 201. In addition, it should be noted that, in this example, the first corner round filling regions 203 (as shown in fig. 2B) may also be constructed by increasing the number of the first inscribed circle filling regions 202 so that the spacing distance between any two adjacent first inscribed circle filling regions 202 is smaller, and/or respectively constructing the four corner points of the rectangular surrounding region 201, so as to reduce the unfilled region of the rectangular surrounding region 201, thereby increasing the detection accuracy.

In the above example, the plurality of rounded fill regions may include a plurality of first inscribed rounded fill regions 202 and a plurality of first corner rounded fill regions 203.

In the embodiment of the disclosure, after a plurality of circular filling areas are constructed on the rectangular surrounding area, when it is determined that a driving obstacle area exists in the plurality of circular filling areas based on the distance field, a positional relationship between the obstacle and the collision detection area corresponding to the target object when driving to the current collision detection point is determined to be an inclusion relationship so as to represent that the obstacle is located inside the collision detection area, and whether the positional relationship between the obstacle and the collision detection area corresponding to the target object when driving to the current collision detection point is determined to be a non-inclusion relationship so as to represent that the obstacle is located outside the collision detection area. Wherein "determining that there is a travel obstacle region in the plurality of circular filling regions based on the distance field" may include: for each circular filling area, determining a target circle center position of the circular filling area in a distance field; inquiring a target obstacle distance corresponding to the position of the center of a target circle from a distance data set of a distance field; in the case where the target obstacle distance is smaller than the radius length of the circular filling area, the circular filling area is taken as a travel obstacle area to determine that the travel obstacle area exists among the plurality of circular filling areas. Furthermore, it should be noted that, in the embodiment of the present disclosure, the foregoing steps may be performed in a serial manner, that is, according to a preset determination sequence, only whether one circular filling area is a driving obstacle area at a time is determined, so that the collision detection method provided by the embodiment of the present disclosure may be applicable to a lower computing platform, for example, an automatic waiting parking (Automated Valet Parking, AVP) platform, or may be performed in a parallel manner, that is, whether a plurality of circular filling areas are driving obstacle areas is determined synchronously, so as to further improve the collision detection efficiency of the target object.

Through the above steps of determining the positional relationship between the collision detection area corresponding to the target object when driving to the current collision detection point and the obstacle based on the pre-constructed distance field, in the embodiment of the present disclosure, the area occupied by the target object when driving to the current collision detection point may be represented by the rectangular surrounding area, and then a plurality of circular filling areas are constructed on the rectangular surrounding area. Based on the distance field, whether a driving obstacle region containing an obstacle exists in the plurality of circular filling regions or not can be rapidly determined, so that the position relationship between the collision detection region corresponding to the target object when the target object drives to the current collision detection point and the obstacle is determined, and the collision detection efficiency of the target object is further improved.

In some alternative embodiments, the "building a plurality of circular filled regions over a rectangular surrounding region" may comprise the steps of:

constructing a plurality of longitudinal inscribed circle filling areas in the length direction of the rectangular surrounding area;

respectively constructing a plurality of corner inscribed circle filling areas in four corner areas of the rectangular surrounding area;

wherein the plurality of circular filling areas comprises a plurality of longitudinal inscribed circular filling areas and a plurality of corner inscribed circular filling areas.

Referring to fig. 3, for the construction of a plurality of circular filling regions, as another specific example of the embodiment of the present disclosure, a plurality of longitudinal inscribed circle filling regions 302 may be constructed in the length direction of the rectangular surrounding region 301, and the diameter length of the longitudinal inscribed circle filling region 302 is equal to the width of the rectangular surrounding region 301, and a plurality of corner inscribed circle filling regions 303 may be constructed in four corner regions of the rectangular surrounding region 301, respectively. In addition, in this example, the number of the vertical inscribed circle filling regions 302 may be increased to make the spacing distance between any two adjacent vertical inscribed circle filling regions 302 smaller and/or the number of the corner inscribed circle filling regions 303 may be increased to make the spacing distance between any two adjacent corner inscribed circle filling regions 303 smaller, so as to reduce the unfilled region of the rectangular surrounding region 301, thereby increasing the detection accuracy.

In the above example, the plurality of circular filling areas may include a plurality of longitudinal inscribed circular filling areas 302 and a plurality of corner inscribed circular filling areas 303.

By the above steps included in the "constructing a plurality of circular filling areas on a rectangular surrounding area", in the embodiment of the present disclosure, a plurality of longitudinal inscribed circle filling areas may be constructed in the length direction of the rectangular surrounding area, and a plurality of corner inscribed circle filling areas may be constructed in four corner areas of the rectangular surrounding area, respectively, and then the plurality of longitudinal inscribed circle filling areas and the plurality of corner inscribed circle filling areas may be used together as a plurality of circular filling areas, so that compared with other construction methods (for example, the construction methods shown in fig. 2A and 2B) of the plurality of circular filling areas, the unfilled area of the rectangular surrounding area may be reduced with a small number of circular filling areas as possible, thereby reducing the subsequent data calculation amount of the collision detection method, and further improving the collision detection efficiency of the target object.

In some alternative embodiments, the "building a plurality of circular filling areas on a rectangular surrounding area" may also comprise the steps of:

a plurality of circular filling areas are constructed on the rectangular surrounding area by adopting a halving method.

In the embodiment of the disclosure, in the process of constructing a plurality of circular filling areas on a rectangular surrounding area by adopting a halving method, at least one circular filling area can be constructed once every halving, the at least one circular filling area forms a detection layer, and the number of the detection layer corresponding to the detection layer is positively correlated with the halving times corresponding to the construction of the detection layer.

Hereinafter, the above steps will be described taking as an example a case where the plurality of round filling regions includes a plurality of longitudinal inscribed round filling regions and a plurality of corner inscribed round filling regions.

(1) And constructing a plurality of longitudinal inscribed circle filling areas in the length direction of the rectangular surrounding area by adopting a halving method.

Referring to fig. 4A, first, a longitudinal inscribed circle filling region 4021 may be constructed at a first end of a rectangular surrounding region 401, and a longitudinal inscribed circle filling region 4022 may be constructed at a second end of the rectangular surrounding region 401, where the longitudinal inscribed circle filling region 4021 and the longitudinal inscribed circle filling region 4022 form a detection layer, and a detection layer number corresponding to the detection layer may be 1. After that, the longitudinal inscribed circle filling area 4023 is constructed by taking the center point A1 of the longitudinal inscribed circle filling area 4021 and the middle point A3 of the center point A2 of the longitudinal inscribed circle filling area 4022 as center points, the longitudinal inscribed circle filling area 4023 forms a detection layer, and the number of the detection layer corresponding to the detection layer can be 2. Finally, the center of the longitudinal inscribed circle filling region 4021 and the middle point A4 of the center of the longitudinal inscribed circle filling region 4023 are taken as center positions, and the longitudinal inscribed circle filling region 4024 is constructed; the longitudinal inscribed circle filling area 4025 is constructed by taking the center point A2 of the longitudinal inscribed circle filling area 4022 and the middle point A5 of the center point A3 of the longitudinal inscribed circle filling area 4023 as center points, the longitudinal inscribed circle filling area 4024 and the longitudinal inscribed circle filling area 4025 form a detection layer, and the number of the detection layer corresponding to the detection layer can be 3. If more detection layers need to be built on the rectangular surrounding area 401 as well, and so on.

(2) And a halving method is adopted to respectively construct a plurality of corner inscribed circle filling areas in four corner areas of the rectangular surrounding area.

Referring to fig. 4B, first, a corner inscribed circle filling area 4031 may be constructed by using a corner position A6 of the rectangular surrounding area 401 as a center position, and a corner inscribed circle filling area 4032 (which may be the same inscribed circle filling area as the longitudinal inscribed circle filling area 4021) may be constructed at one end near the corner position A6 in the rectangular surrounding area 401, where the corner inscribed circle filling area 4031 and the corner inscribed circle filling area 4032 form a detection layer, and the number of the detection layer corresponding to the detection layer may be 1. After that, the corner inscribed circle filling area 4033 is formed by half-folding the corner inscribed circle filling area 4031 and the middle point A7 of the center position A1 of the corner inscribed circle filling area 4032, the corner inscribed circle filling area 4033 forms a detection layer, and the number of the detection layer corresponding to the detection layer may be 2. Finally, the corner inscribed circle filling area 4034 is constructed by halving once again and taking the center point A6 of the corner inscribed circle filling area 4031 and the middle point A8 of the center point A7 of the corner inscribed circle filling area 4033 as center positions; the corner inscribed circle filling area 4035 is constructed by taking the center point A1 of the corner inscribed circle filling area 4033 and the middle point A9 of the center point A7 of the corner inscribed circle filling area 4033 as center points, the corner inscribed circle filling area 4034 and the corner inscribed circle filling area 4035 form a detection layer, and the number of the detection layer corresponding to the detection layer can be 3. If more detection layers associated with corner position A6 need to be built on rectangular surrounding area 401, and so on.

It can be appreciated that in the embodiment of the present disclosure, the detection layer related to the corner position a10, the detection layer related to the corner position a12, and the detection layer related to the corner position a13 may also be constructed on the rectangular surrounding area 401 according to the above construction method, which is not described herein.

Finally, a plurality of circular filled regions as shown in fig. 4C, including a plurality of longitudinal inscribed circular filled regions and a plurality of corner inscribed circular filled regions, can be obtained.

Through the above steps of "constructing a plurality of circular filling areas on a rectangular surrounding area", in the embodiment of the present disclosure, a halving method may be adopted to construct a plurality of circular filling areas on a rectangular surrounding area, so that any two adjacent similar inscribed circular filling areas have equal spacing distances therebetween, thereby improving the collision detection effect of the target object.

In some alternative embodiments, the "using halving to construct a plurality of circular fill areas over a rectangular surrounding area" may include the steps of:

determining the current driving scene of the target object;

acquiring a halving parameter corresponding to a current driving scene;

and constructing a plurality of circular filling areas on the rectangular surrounding area according to the halving parameter by adopting a halving method.

The driving scene of the target object may include a driving scene and a parking scene, and the driving scene may be subdivided into a high-speed driving scene, a low-speed driving scene, and the like.

The halving parameter is used for controlling the number of detection layers constructed on the rectangular surrounding area. For example, in the case of a halving parameter of 2, a halving method will be used to construct 2 detection layers on a rectangular surrounding area; in case of a halving parameter of 3, a halving method will be used to construct 3 detection layers on a rectangular surrounding area.

In the embodiment of the present disclosure, a halving parameter corresponding to each driving scenario may be preset. For example, setting a halving parameter corresponding to a high-speed driving scene to be 1; setting the halving parameter corresponding to the low-speed driving scene as 2; setting the halving parameter corresponding to the parking scene as 3. It should be noted that in the embodiment of the present disclosure, the half-conversion parameters corresponding to each driving scenario may be specifically set in combination with the platform calculation force and the actual service requirement, which is not specifically limited in the embodiment of the present disclosure.

In addition, in the embodiment of the present disclosure, in the case where the plurality of circular filling areas includes the plurality of longitudinal inscribed circle filling areas and the plurality of corner inscribed circle filling areas, since the longitudinal length of the target object is generally greater than the transverse length, the longitudinal inscribed circle filling areas and the corner inscribed circle filling areas may also be constructed according to different halving parameters in the same driving scene. For example, under a high-speed driving scene, a longitudinal inscribed circle filling area is constructed according to a halving parameter '2', namely, 2 detection layers are constructed in the length direction of a rectangular surrounding area, and a corner inscribed circle filling area is constructed according to a halving parameter '1', namely, 1 detection layer is respectively constructed in four corner areas of the rectangular surrounding area; under a low-speed driving scene, constructing a longitudinal inscribed circle filling area according to a halving parameter '3', namely constructing 3 detection layers in the length direction of a rectangular surrounding area, and constructing corner inscribed circle filling areas according to a halving parameter '2', namely respectively constructing 2 detection layers in four corner areas of the rectangular surrounding area; in the parking scene, a longitudinal inscribed circle filling area is constructed according to a halving parameter '4', namely, 4 detection layers are constructed in the length direction of a rectangular surrounding area, and a corner inscribed circle filling area is constructed according to a halving parameter '3', namely, 3 detection layers are respectively constructed in four corner areas of the rectangular surrounding area. Also, it should be noted that in the embodiment of the present disclosure, in each driving scenario, when the longitudinal inscribed circle filling area and the corner inscribed circle filling area are constructed, the adopted halving parameters may be specifically set in combination with the platform calculation force and the actual service requirement, which is not specifically limited in the embodiment of the present disclosure.

By adopting the halving method to construct a plurality of circular filling areas on the rectangular surrounding area, it can be appreciated that in the embodiment of the disclosure, halving parameters corresponding to each driving scene can be preset, so that after the current driving scene of the target object is determined, halving parameters corresponding to the current driving scene can be directly obtained, and then adopting the halving method to construct a plurality of circular filling areas on the rectangular surrounding area according to the halving parameters, thereby reducing time consumption for constructing a plurality of circular filling areas and further improving the collision detection efficiency of the target object.

In addition, in the embodiment of the disclosure, the number of detection layers may also be dynamically controlled during the process of constructing a plurality of circular filling areas on the rectangular surrounding area. For example, in the case where the plurality of circular filling regions includes a plurality of longitudinal inscribed circle filling regions and a plurality of corner inscribed circle filling regions, the halving method is adopted, and in the process of constructing the plurality of circular filling regions on the rectangular surrounding region, the intersection distance between two adjacent inscribed circle filling regions of the same kind and the rectangular surrounding region is detected once every halving, and when the intersection distance is smaller than a preset distance threshold value, the next halving operation is stopped, and the number of final detection layers is determined. The preset distance threshold may be set in combination with the platform computing power and the actual service requirement, which is not specifically limited in the embodiments of the present disclosure.

In the embodiment of the disclosure, each circular filling area has a corresponding detection layer number, and the detection layer number corresponding to each circular filling area is positively correlated with the corresponding halving times when the circular filling area is constructed. The detection layer number corresponding to each circular filling area is the detection layer number corresponding to the detection layer where the circular filling area is located. Based on this, in some alternative embodiments, "determining that a travel obstacle region including an obstacle exists in a plurality of circular filling regions based on a distance field" may include the steps of:

selecting at least one target filling area with the same detection layer number from a plurality of circular filling areas according to the sequence of the detection layer numbers from low to high;

for each target filling area, determining a first circle center position of the target filling area in a distance field;

inquiring a first obstacle distance corresponding to the first circle center position from a distance data set of a distance field;

in the case where the first obstacle distance is smaller than the radius length of the target filling area, the target filling area is taken as a travel obstacle area to determine that the travel obstacle area exists among the plurality of circular filling areas.

In the embodiment of the present disclosure, after at least one target filling area with the same detection layer number is selected from a plurality of circular filling areas according to the order of the detection layer numbers from low to high, the subsequent steps may be performed in a serial manner, that is, according to a preset judging order, only whether one target filling area is a driving obstacle area at a time is judged, so that the collision detection method provided by the embodiment of the present disclosure may be applicable to a lower power platform, or may be performed in a parallel manner, that is, whether the at least one target filling area is a driving obstacle area is synchronously judged, so as to further improve the collision detection efficiency of the target object.

In addition, in the embodiment of the disclosure, taking the distance field as an ESDF as an example, after determining the first center position of the target filling area in the ESDF, the nearest obstacle distance corresponding to the first center position may be queried as the first obstacle distance from the distance dataset of the ESDF with the time complexity of O (1), and denoted as D1. In the case where the first obstacle distance D1 is smaller than the radius length R1 of the target filling area, that is, D1 < R1, the target filling area is taken as the travel obstacle area to determine that the travel obstacle area exists among the plurality of circular filling areas.

Since the distribution characteristics of the obstacles are generally continuously distributed in large blocks, by the above steps included in the "determining that the driving obstacle region including the obstacle exists in the plurality of circular filling regions based on the distance field", in the embodiment of the present disclosure, it may be determined whether the driving obstacle region including the obstacle exists in the plurality of circular filling regions in the order of the detection layer numbers from low to high. Even in a low-power platform where it is impossible to determine whether or not a plurality of circular filling areas are travel obstacle areas simultaneously, the collision detection efficiency of the target object can be relatively improved.

In some optional embodiments, "determining, based on the pre-constructed distance field, a positional relationship between the collision detection area corresponding to the target object when traveling to the current collision detection point and the obstacle", may further include the steps of:

constructing a circular surrounding area outside the rectangular surrounding area; wherein the collision detection region includes a circular surrounding region;

in case it is determined that no obstacle is contained in the circular surrounding area based on the distance field, the positional relationship is determined as a non-containing relationship to characterize that the obstacle is located outside the collision detection area, otherwise the step of constructing a plurality of circular filling areas on the rectangular surrounding area is performed.

Referring to fig. 5, in the embodiment of the disclosure, the circular surrounding area 501 may be the smallest circumscribing circle of the rectangular surrounding area 502, that is, the center position B of the circular surrounding area 501 is the center point of the rectangular surrounding area 502 (to represent the center position of the target object), and the radius length of the circular surrounding area 501 is half of the diagonal length of the rectangular surrounding area.

In the embodiment of the disclosure, after the circular surrounding area is constructed outside the rectangular surrounding area, if it is determined that the circular surrounding area does not include an obstacle based on the distance field, determining that the position relationship between the collision detection area corresponding to the target object when the target object travels to the current collision detection point and the obstacle is a non-inclusion relationship, so as to characterize that the obstacle is located outside the collision detection area, otherwise, executing the step of constructing a plurality of circular filling areas on the rectangular surrounding area.

By determining the positional relationship between the collision detection area corresponding to the target object when the target object travels to the current collision detection point and the obstacle based on the pre-constructed distance field, including the above steps, in the embodiment of the present disclosure, a circular surrounding area may be constructed outside the rectangular surrounding area, and then whether the circular surrounding area contains the obstacle may be determined based on the distance field, so that in the case that it is determined that the circular surrounding area does not contain the obstacle, the positional relationship between the collision detection area corresponding to the target object when the target object travels to the current collision detection point and the obstacle is determined to be a non-containing relationship, and the execution of the subsequent steps is stopped, thereby further improving the collision detection efficiency of the target object.

In some alternative embodiments, "determining that no obstacle is contained in the circular surrounding area based on the distance field" may include the steps of:

determining a second center position of the circular surrounding area in the distance field;

inquiring a second obstacle distance corresponding to the second circle center position from a distance data set of the distance field;

in the case where the second obstacle distance is greater than the radius length of the circular surrounding area, it is determined that no obstacle is included in the circular surrounding area.

Taking the distance field as an ESDF as an example, after determining the second center position of the circular surrounding area in the ESDF, the nearest obstacle distance corresponding to the second center position can be queried as the second obstacle distance from the distance data set of the ESDF with the time complexity of O (1), and is denoted as D2. In case the second obstacle distance D2 is larger than the radius length R2 of the circular surrounding area, i.e. D2 > R2, it is determined that no obstacle is contained in the circular surrounding area.

Through the above steps of determining that the circular surrounding area does not include an obstacle based on the distance field, in the embodiment of the present disclosure, after determining the second center position of the circular surrounding area in the distance field, the second obstacle distance corresponding to the second center position may be quickly queried from the distance dataset of the distance field, so that the circular surrounding area is determined to not include an obstacle when the second obstacle distance is greater than the radius length of the circular surrounding area, thereby further improving the collision detection efficiency of the target object.

In some alternative embodiments, "determining the current collision detection point of the target object" may include the steps of:

under the condition that the running planning path of the target object is a straight path, determining a current collision detection point from a plurality of sampling points to be detected according to a preset traversal sequence; the method comprises the steps that a plurality of sampling points to be detected are obtained by sampling detection points of a running planning path according to a first sampling interval which is larger than a conventional sampling threshold;

and/or determining a current collision detection point from a plurality of sampling points to be detected according to a breadth first search (Breadth First Search, BFS) algorithm under the condition that the running planning path of the target object is a curve path; the plurality of sampling points to be detected are obtained by sampling detection points of the running planning path according to the second sampling interval.

Based on the above steps, it may be understood that in the embodiment of the present disclosure, for the case that the running planned path is a straight path, the running planned path may be sampled according to the first sampling interval greater than the conventional sampling threshold value to obtain a plurality of to-be-detected sampling points. Wherein the conventional sampling threshold may be 0.2M. Based on this, in the embodiment of the present disclosure, the value of the first sampling interval L may be: l is more than 0.2 and less than or equal to Lmax. Where Lmax may be the length of the target object, for example, may be 4.5M. In the case where the first sampling interval L is equal to the length Lmax of the target object, that is, l=lmax, the overlap interval between any two adjacent collision detection points is just 0, and therefore, the collision detection efficiency of the target object can be greatly improved when the travel planned path is a straight path.

After the detection points of the running planning path are sampled according to the first sampling interval to obtain a plurality of to-be-detected sampling points, the current collision detection points can be determined from the plurality of to-be-detected sampling points according to a preset traversal sequence. As a specific example of an embodiment of the present disclosure, the preset traversal order may be a near-to-far traversal order with respect to the target object. Based on this, please refer to fig. 6, in the embodiment of the disclosure, it is assumed that, according to the first sampling interval, the detection point is sampled on the driving planning path, and the obtained plurality of to-be-detected sampling points include to-be-detected sampling point C1, to-be-detected sampling point C2, to-be-detected sampling point C3, to-be-detected sampling point C4 and to-be-detected sampling point C5. Then, according to the preset traversal sequence, the to-be-detected sampling point C1, the to-be-detected sampling point C2, the to-be-detected sampling point C3, the to-be-detected sampling point C4 and the to-be-detected sampling point C5 may be sequentially used as current collision detection points.

Based on the above steps, it may be further understood that in the embodiment of the present disclosure, for the case that the running planned path is a curved path, the detecting point may be sampled according to the second sampling interval, so as to obtain a plurality of to-be-detected sampling points. The second sampling distance may be inversely related to the curvature parameter of the planned travel path, i.e., the larger the curvature parameter of the planned travel path, the smaller the second sampling distance.

After the detection points of the running planning path are sampled according to the second sampling interval to obtain a plurality of to-be-detected sampling points, the current collision detection points can be determined from the plurality of to-be-detected sampling points according to a BFS algorithm based on the distribution characteristics of the obstacle. Referring to fig. 7, in the embodiment of the disclosure, it is assumed that, according to the second sampling interval, the detection points of the driving planning path are sampled, and the obtained plurality of to-be-detected sampling points include to-be-detected sampling point D1, to-be-detected sampling point D2, to-be-detected sampling point D3, to-be-detected sampling point D4 and to-be-detected sampling point D5. Then, according to the preset traversal sequence, the to-be-detected sampling point D1, the to-be-detected sampling point D5, the to-be-detected sampling point D3, the to-be-detected sampling point D2 and the to-be-detected sampling point D4 may be sequentially used as current collision detection points.

Through the above steps included in the determination of the current collision detection point of the target object, in the embodiment of the present disclosure, the running planning path may be combined with continuity and distribution characteristics of the obstacle, the sampling interval is adaptively increased in the case that the running planning path of the target object is a straight path, and the BFS algorithm is selected to determine the current collision detection point from a plurality of sampling points to be detected in the case that the running planning path of the target object is a curved path, so as to further improve the collision detection efficiency of the target object.

In some optional implementations, the collision detection method provided by the embodiments of the present disclosure may further include the following steps:

determining a next collision detection point from a running planning path of the target object under the condition that a collision detection result corresponding to the current collision detection point is collision risk-free;

acquiring a collision detection result corresponding to the next collision detection point;

under the condition that a risk detection point exists on the running planning path, determining the running planning path as a risk path; the collision detection result corresponding to the risk detection point is collision risk.

As described above, in the embodiment of the present disclosure, in the case where the traveling planned path of the target object is a straight path, the next collision detection point may be determined from a plurality of sampling points to be detected according to a preset traversal sequence; the method comprises the steps that a plurality of sampling points to be detected are obtained by sampling detection points of a running planning path according to a first sampling interval which is larger than a conventional sampling threshold; and/or under the condition that the running planning path of the target object is a curve path, determining a next collision detection point from a plurality of sampling points to be detected according to a BFS algorithm; the plurality of sampling points to be detected are obtained by sampling detection points of the running planning path according to the second sampling interval.

Thereafter, according to the steps described in step S102 and step S103, a collision detection result corresponding to the next collision detection point is obtained, which is not described in detail in the embodiments of the present disclosure. And under the condition that a risk detection point exists on the running planning path, determining the running planning path as a risk path, and otherwise, determining the running planning path as a safety path. The collision detection result corresponding to the risk detection point is collision risk.

Through the steps, the collision detection method provided by the embodiment of the disclosure can also be used for performing collision detection on the running planning path, so that the applicable range of the collision detection method is increased.

A complete flow of a collision detection method according to an embodiment of the present disclosure will be described below with reference to fig. 8.

And sampling detection points of the running planning path to obtain a plurality of sampling points to be detected.

Traversing the plurality of sampling points to be detected, and determining the current collision detection point. Specifically, under the condition that the running planning path of the target object is a straight path, determining a current collision detection point from a plurality of sampling points to be detected according to a preset traversal sequence; and under the condition that the running planning path of the target object is a curve path, determining the current collision detection point from a plurality of sampling points to be detected according to a BFS algorithm.

It is determined whether an obstacle is included in the circular surrounding area. In particular, a second center position of the circular surrounding area may be determined in the distance field; inquiring a second obstacle distance corresponding to the second circle center position from a distance data set of the distance field; and if the second obstacle distance is larger than the radius length of the circular surrounding area, determining that no obstacle is contained in the circular surrounding area, otherwise, determining that the obstacle is contained in the circular surrounding area. The round surrounding area is the minimum circumcircle of the rectangular surrounding area corresponding to the target object when the target object runs to the current collision detection point.

Under the condition that the circular surrounding area contains the obstacle, selecting at least one target filling area with the same corresponding detection layer number from a plurality of circular filling areas constructed on the rectangular surrounding area according to the sequence of the detection layer numbers from low to high; for each target filling area, determining a first circle center position of the target filling area in a distance field; inquiring a first obstacle distance corresponding to the first circle center position from a distance data set of a distance field; and under the condition that the first obstacle distance is smaller than the radius length of the target filling area, taking the target filling area as a driving obstacle area to determine that the driving obstacle area exists in the plurality of circular filling areas, otherwise, continuing to detect the circular filling area of the next layer until the number of the detection layer reaches the maximum. When the number of the detection layer reaches the maximum, if the driving obstacle area is still not determined from the plurality of round filling areas, determining that the collision detection result corresponding to the current collision detection point is no collision risk, namely, determining that the current collision detection point is a safety detection point, otherwise, determining that the collision detection result corresponding to the current collision detection point is collision risk, namely, determining that the current collision detection point is a risk detection point.

Under the condition that the current collision detection point is determined to be the safety detection point, continuing to detect the next collision detection point until the running planning path is determined to be a safety path under the condition that a plurality of collision detection points are determined to be the safety detection points; and under the condition that the current collision detection point is determined to be the risk detection point, determining the running planning path as a risk path.

Fig. 9 is a schematic application scenario diagram of a collision detection method according to an embodiment of the disclosure.

As described above, the collision detection method provided by the embodiment of the present disclosure is applied to an electronic device. Wherein the electronic device is intended to represent various forms of digital computers, such as automobile computers, or other suitable computers.

In the embodiment of the disclosure, the electronic device may be configured to perform a collision detection method:

determining a current collision detection point of the target object;

determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and the obstacle based on a pre-constructed distance field;

It should be noted that, in the embodiment of the present disclosure, the electronic device may be disposed in an autonomous vehicle.

In addition, it should be further noted that, in the embodiment of the present disclosure, the schematic view of the scenario shown in fig. 9 is merely illustrative and not restrictive, and those skilled in the art may make various obvious changes and/or substitutions based on the example of fig. 9, and the obtained technical solutions still fall within the scope of the embodiment of the present disclosure.

In order to better implement the above collision detection method, the embodiments of the present disclosure also provide a collision detection apparatus 1000, and the collision detection apparatus 1000 may be integrated in an electronic device. Hereinafter, a collision detecting apparatus 1000 according to the disclosed embodiment will be described with reference to a schematic structural diagram shown in fig. 10.

The collision detection apparatus 1000 includes:

a detection point determination unit 1001 for determining a current collision detection point of the target object;

a detection unit 1002 configured to determine a positional relationship between a collision detection area corresponding to the target object when traveling to the current collision detection point and the obstacle, based on a distance field constructed in advance;

a result acquisition unit 1003 for acquiring a collision detection result of the target object at the current collision detection point according to the positional relationship.

In some alternative embodiments, the detection unit 1002 is configured to:

in the case where it is determined that a travel obstacle region including an obstacle exists among the plurality of circular filling regions based on the distance field, the positional relationship is determined as an inclusion relationship to characterize that the obstacle is located inside the collision detection region.

In some alternative embodiments, the detection unit 1002 is configured to:

determining the current driving scene of the target object;

acquiring a halving parameter corresponding to a current driving scene;

In some alternative embodiments, the detection unit 1002 is configured to:

In some alternative embodiments, each circular filling area has a corresponding detection layer number, and the detection layer number corresponding to each circular filling area is positively correlated with the half-turn number corresponding to when the circular filling area is constructed, and the detection unit 1002 is configured to:

In some alternative embodiments, the detection unit 1002 is further configured to:

In some alternative embodiments, the detection unit 1002 is configured to:

In some alternative embodiments, the detection point determining unit 1001 is configured to:

under the condition that the running planning path of the target object is a straight path, determining a current collision detection point from a plurality of sampling points to be detected according to a preset traversal sequence; the plurality of sampling points to be detected are obtained by sampling detection points of the driving planning path according to a first sampling interval which is larger than a conventional sampling threshold value.

And/or determining a current collision detection point from a plurality of sampling points to be detected according to a BFS algorithm under the condition that the running planning path of the target object is a curve path; the plurality of sampling points to be detected are obtained by sampling detection points of the running planning path according to the second sampling interval.

In some alternative embodiments, the collision detection apparatus further comprises a path detection control unit for:

In the implementation, each module may be implemented as an independent entity, or may be combined arbitrarily, and implemented as the same entity or a plurality of entities, and the implementation of each module may be referred to the foregoing embodiment of the collision detection method, which is not described herein.

The collision detection device provided by the embodiment of the disclosure can determine the current collision detection point of the target object; determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and the obstacle based on a pre-constructed distance field; and obtaining a collision detection result of the target object on the current collision detection point according to the position relation. Because the distance field is provided with a corresponding distance data set and is used for storing the nearest obstacle distance corresponding to each position point in the field, the position relationship between the corresponding collision detection area and the obstacle when the target object runs to the current collision detection point can be rapidly determined based on the distance field, and then the collision detection result of the target object on the current collision detection point is obtained according to the position relationship, so that the collision detection efficiency of the target object is improved.

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related user personal information all conform to the regulations of related laws and regulations, and the public sequence is not violated.

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

Fig. 11 illustrates a schematic block diagram of an example electronic device 1100 that can be used to implement embodiments of the present disclosure.

As previously mentioned, in the presently disclosed embodiments, electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 11, the electronic device 1100 includes a computing unit 1101 that can perform various appropriate actions and processes according to a computer program stored in a Read-Only Memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access Memory (Random Access Memory, RAM) 1103. In the RAM1103, various programs and data required for the operation of the electronic device 1100 can also be stored. The computing unit 1101, ROM 1102, and RAM1103 are connected to each other by a bus 1104. An Input/Output (I/O) interface 1105 is also connected to bus 1104.

A number of components in the electronic device 1100 are connected to the I/O interface 1105, including: an input unit 1106, e.g., a keyboard, a mouse, etc.; an output unit 1107 such as various types of displays, speakers, and the like; a storage unit 1108, e.g., a magnetic disk, an optical disk, etc.; and a communication unit 1109 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 1109 allows the electronic device 1100 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunications networks.

The computing unit 1101 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1101 include, but are not limited to, a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), various dedicated artificial intelligence (Artificial Intelligence, AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (Digital Signal Process, DSP), and any suitable processors, controllers, microcontrollers, etc. The calculation unit 1101 performs the respective methods and processes described above, for example, a collision detection method. For example, in some alternative embodiments the collision detection methods may each be implemented as a computer software program, which is tangibly embodied on a non-transitory computer-readable storage medium, such as storage unit 1108. In some alternative embodiments, some or all of the computer programs may be loaded and/or installed onto electronic device 1100 via ROM 1102 and/or communication unit 1109. When the computer program is loaded into the RAM 1103 and executed by the computing unit 1101, one or more steps of the collision detection method described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform the collision detection method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (Field Programmable Gate Array, FPGAs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), application specific standard products (Application Specific Standard Product, ASSPs), systems On Chip (SOC), complex programmable logic devices (Complex Programmable Logic Device, CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute 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.

In the context of this disclosure, a non-transitory computer readable storage medium may 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. The non-transitory computer readable storage medium may be a machine readable signal medium or a machine readable storage medium. The non-transitory computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a non-transitory computer readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: other types of devices may also be used to provide interaction with a user, for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback), and input from the user may be received in any form (including acoustic input, speech input, or tactile input).

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (Local Area Network, LAN), wide area network (Wide Area Network, WAN) and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

The disclosed embodiments also provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the foregoing collision detection method.

The disclosed embodiments also provide a computer program product comprising a computer program which, when executed by a processor, implements the aforementioned collision detection method.

The electronic equipment, the automatic driving vehicle, the medium and the computer program product provided by the embodiment of the disclosure can determine the current collision detection point of the target object; determining the position relationship between a collision detection area corresponding to the target object when the target object runs to the current collision detection point and the obstacle based on a pre-constructed distance field; and obtaining a collision detection result of the target object on the current collision detection point according to the position relation. Because the distance field is provided with a corresponding distance data set and is used for storing the nearest obstacle distance corresponding to each position point in the field, the position relationship between the corresponding collision detection area and the obstacle when the target object runs to the current collision detection point can be rapidly determined based on the distance field, and then the collision detection result of the target object on the current collision detection point is obtained according to the position relationship, so that the collision detection efficiency of the target object is improved.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein. Moreover, in this disclosure, relational terms such as "first," "second," "third," and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Furthermore, the term "plurality" in this disclosure may be understood as at least two.

The foregoing detailed description is not intended to limit the scope of the disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A collision detection method, comprising:

Determining a current collision detection point of the target object;

2. The method of claim 1, wherein the determining, based on the pre-constructed distance field, a positional relationship between a collision detection area and an obstacle to which the target object corresponds when traveling to the current collision detection point, comprises:

constructing a plurality of circular filling areas on the rectangular surrounding area; wherein the collision detection region includes the plurality of circular filling regions;

in the case where it is determined that a traveling obstacle region including the obstacle exists among the plurality of circular filling regions based on the distance field, the positional relationship is determined as an inclusion relationship to characterize that the obstacle is located inside the collision detection region.

3. The method of claim 2, wherein said constructing a plurality of circular fill areas on said rectangular surrounding area comprises:

And constructing a plurality of circular filling areas on the rectangular surrounding area by adopting a halving method.

4. A method according to claim 3, wherein said employing a halving process to construct a plurality of circular fill areas over said rectangular surrounding area comprises:

determining the current driving scene of the target object;

acquiring a halving parameter corresponding to the current driving scene;

5. The method of any of claims 2-4, wherein the constructing a plurality of circular fill areas on the rectangular surrounding area comprises:

wherein the plurality of circular filling areas includes the plurality of longitudinal inscribed circular filling areas and the plurality of corner inscribed circular filling areas.

6. A method according to claim 3, wherein each of the circular filling areas has a corresponding detection layer number, and the detection layer number corresponding to each of the circular filling areas is positively correlated with the corresponding halving times when the circular filling areas are constructed; the determining, based on the distance field, that a driving obstacle region including the obstacle exists in the plurality of circular filling regions includes:

Selecting at least one target filling area with the same detection layer number from the plurality of circular filling areas according to the sequence of the detection layer numbers from low to high;

for each of the target fill areas, determining a first center position of the target fill area in the distance field;

inquiring a first obstacle distance corresponding to the first circle center position from a distance data set of the distance field;

and in the case that the first obstacle distance is smaller than the radius length of the target filling area, taking the target filling area as the driving obstacle area to determine that the driving obstacle area exists in the plurality of circular filling areas.

7. The method of claim 2, wherein the determining, based on the pre-constructed distance field, a positional relationship between a collision detection area and an obstacle to which the target object corresponds when traveling to the current collision detection point, further comprises:

constructing a circular surrounding area outside the rectangular surrounding area; wherein the collision detection region includes the circular surrounding region;

and determining that the position relationship is a non-inclusion relationship in the case that the obstacle is not included in the circular surrounding area based on the distance field so as to represent that the obstacle is positioned outside the collision detection area, and otherwise, executing the step of constructing a plurality of circular filling areas on the rectangular surrounding area.

8. The method of claim 7, wherein the determining that the obstacle is not contained in the circular surrounding area based on the distance field comprises:

inquiring a second obstacle distance corresponding to the second circle center position from the distance data set of the distance field;

in the case where the second obstacle distance is greater than the radius length of the circular surrounding area, it is determined that the obstacle is not included in the circular surrounding area.

9. The method of claim 1, wherein the determining the current collision detection point of the target object comprises:

under the condition that the running planning path of the target object is a straight path, determining the current collision detection point from a plurality of sampling points to be detected according to a preset traversal sequence; the plurality of sampling points to be detected are obtained by sampling detection points of the running planning path according to a first sampling interval which is larger than a conventional sampling threshold;

and/or determining the current collision detection point from a plurality of sampling points to be detected according to a breadth-first search algorithm under the condition that the running planning path of the target object is a curve path; the plurality of sampling points to be detected are obtained by sampling detection points of the running planning path according to a second sampling interval.

10. The method of claim 1 or 9, further comprising:

under the condition that a risk detection point exists on the running planning path, determining the running planning path as a risk path; and the collision detection result corresponding to the risk detection point is collision risk.

11. A collision detection apparatus comprising:

12. The apparatus of claim 11, wherein the detection unit is configured to:

13. The apparatus of claim 12, wherein the detection unit is configured to:

14. The apparatus of claim 13, wherein the detection unit is configured to:

determining the current driving scene of the target object;

acquiring a halving parameter corresponding to the current driving scene;

15. The apparatus according to any one of claims 12 to 14, wherein the detection unit is configured to:

16. The apparatus of claim 13, wherein each of the circular fill areas has a corresponding detection layer number, and the detection layer number corresponding to each of the circular fill areas is positively correlated with a corresponding halving number when the circular fill areas are constructed; the detection unit is used for:

17. The apparatus of claim 12, wherein the detection unit is further configured to:

18. The apparatus of claim 17, wherein the detection unit is configured to:

19. The apparatus of claim 11, wherein the detection point determination unit is configured to:

20. The apparatus according to claim 11 or 19, further comprising a path detection control unit for:

21. An electronic device, comprising:

at least one processor;

a memory communicatively coupled to the at least one processor;

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 10.

22. An autonomous vehicle comprising the electronic device of claim 21.

23. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-10.

24. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 10.