Unmanned vehicle obstacle bypassing method, device, unmanned vehicle and storage medium

Abstract

The embodiment of the present invention discloses an unmanned vehicle obstacle bypassing method, device, unmanned vehicle and storage medium, the method comprising: if a first obstacle marked as a left bypass is detected, at least two left boundary points of the first obstacle in a target coordinate system are determined; if a second obstacle marked as a right bypass is detected, at least two right boundary points of the second obstacle in a target coordinate system are determined, wherein the target coordinate system includes a first coordinate axis perpendicular to the centerline of the road and a second coordinate axis distributed along the centerline of the road and in the direction of the accumulated distance; according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, the bypass constraint boundary of the unmanned vehicle is determined; and the unmanned vehicle is controlled to travel within the bypass constraint boundary. The problem that the prior art cannot accurately determine the bypass constraint boundary of the unmanned vehicle is solved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of unmanned vehicle control, and in particular to an unmanned vehicle obstacle circumvention method, device, unmanned vehicle and storage medium. Background Art

[0002] The current mobile robot technology is developing rapidly. With the continuous expansion of robot application scenarios and modes in recent years, various mobile robots have emerged one after another, and unmanned vehicles are one of them. At present, the path planning algorithm for unmanned vehicles is usually divided into two steps: one is the generation of rough paths and steering strategies, and the other is to obtain the detour constraint boundary required by the optimizer according to the steering strategy, and then generate the final path that meets the motion constraints through the optimizer and the detour constraint boundary. In order to improve the safety of unmanned vehicle driving and simplify the steps of obtaining the detour constraint boundary, the existing technology expands the range of obstacles when determining the detour constraint boundary. Although the expansion of the obstacle range improves the safety of the detour constraint boundary, it brings about the reduction of the detour constraint boundary, thereby increasing the difficulty of the optimizer to optimize the path, resulting in unsolvable situations in some scenarios, and reducing the speed and accuracy of path optimization.

[0003] In summary, in the process of implementing the present invention, the inventors found that the prior art has at least the following technical problems: the safety of the detour constraint boundary is improved by expanding the range of obstacles, which increases the difficulty of the optimizer to perform path optimization, resulting in unsolvable situations in certain scenarios, and reducing the speed and accuracy of path optimization. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for unmanned vehicle obstacle bypassing, an unmanned vehicle and a storage medium, which solve the problem that the prior art cannot accurately determine the unmanned vehicle bypass constraint boundaries.

[0005] In a first aspect, an embodiment of the present invention provides a method for unmanned vehicle to bypass obstacles, comprising:

[0006] If a first obstacle marked with a left detour is detected, at least two left boundary points of the first obstacle in the target coordinate system are determined; if a second obstacle marked with a right detour is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in a direction of accumulated distance;

[0007] Determining a detour constraint boundary of the unmanned vehicle according to a maximum constraint boundary of the road, a minimum constraint boundary of the road, a left boundary of the obstacle defined by the at least two left boundary points, and / or a right boundary of the obstacle defined by the at least two right boundary points;

[0008] The unmanned vehicle is controlled to travel within the detour constraint boundary to detour through the corresponding obstacle.

[0009] In a second aspect, an embodiment of the present invention further provides an obstacle circumvention device for an unmanned vehicle, comprising:

[0010] A boundary point module, for determining at least two left boundary points of a first obstacle in a target coordinate system if a first obstacle marked as a left detour is detected; and for determining at least two right boundary points of the second obstacle in the target coordinate system if a second obstacle marked as a right detour is detected, wherein the target coordinate system includes a first coordinate axis perpendicular to a center line of the road and a second coordinate axis distributed along the center line of the road and in a direction of accumulated distance;

[0011] A determination module, configured to determine a detour constraint boundary of the unmanned vehicle according to a maximum constraint boundary of the road, a minimum constraint boundary of the road, and a left boundary of the obstacle defined by the at least two left boundary points and / or a right boundary of the obstacle defined by the at least two right boundary points;

[0012] The driving module is used to control the unmanned vehicle to drive within the detour constraint boundary to detour through the corresponding obstacle.

[0013] In a third aspect, an embodiment of the present invention further provides an unmanned vehicle, comprising:

[0014] Vehicle body;

[0015] A traveling mechanism, arranged on the vehicle body, and used for driving the vehicle body to travel;

[0016] A controller is disposed in the vehicle body and is used to determine at least two left boundary points of the first obstacle in a target coordinate system if a first obstacle marked with a left detour is detected; and to determine at least two right boundary points of the second obstacle in the target coordinate system if a second obstacle marked with a right detour is detected; determine the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, and control the vehicle body to travel within the detour constraint boundary to detour the current obstacle, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in the direction of accumulated distance.

[0017] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the unmanned vehicle obstacle bypassing method as described in any embodiment.

[0018] The technical solution of the obstacle bypassing method for an unmanned vehicle provided in an embodiment of the present invention is as follows: if a first obstacle marked with a left bypass is detected, at least two left boundary points of the first obstacle in a target coordinate system are determined; if a second obstacle marked with a right bypass is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined. Since the at least two left boundary points or the at least two right boundary points are both on the obstacle boundary, the obstacle left boundary defined by the at least two left boundary points and / or the obstacle right boundary defined by the at least two right boundary points, as well as the maximum constraint boundary and the minimum constraint boundary of the road, the obstacle area and the non-obstacle area in the road can be accurately divided, so that the maximum bypass constraint boundary of the unmanned vehicle on the road can be accurately determined. The maximum bypass constraint boundary can greatly reduce the difficulty of the optimizer in performing path optimization, and improve the speed and accuracy of the optimizer's path optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a flow chart of an obstacle bypassing method for an unmanned vehicle provided in Embodiment 1 of the present invention;

[0021] Figure 2 This is a schematic diagram of an unmanned vehicle bypassing an obstacle provided by the first embodiment of the present invention;

[0022] Figure 3 is a flow chart of an obstacle bypassing method for an unmanned vehicle provided in Embodiment 2 of the present invention;

[0023] Figure 4 is a schematic diagram of the relationship between the Frenet coordinate system and the Cartesian coordinate system provided in the second embodiment of the present invention;

[0024] Figure 5 is a structural block diagram of an obstacle circumvention device for an unmanned vehicle provided in Embodiment 3 of the present invention;

[0025] Figure 6 This is a structural block diagram of an unmanned vehicle provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described through implementation methods with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1

[0028] Figure 1 is a flow chart of the unmanned vehicle obstacle bypassing method provided by the first embodiment of the present invention. The technical solution of this embodiment is applicable to the case where the unmanned vehicle's bypassable range is increased by accurately determining the obstacle boundary. The method can be executed by the unmanned vehicle obstacle bypassing device provided by the embodiment of the present invention, which can be implemented in software and / or hardware and configured in a processor for application. Figure 1 As shown, the method specifically comprises the following steps:

[0029] S101. If a first obstacle marked with a left detour is detected, at least two left boundary points of the first obstacle in a target coordinate system are determined; if a second obstacle marked with a right detour is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined, wherein the target coordinate system includes a first coordinate axis perpendicular to a centerline of the road and a second coordinate axis distributed along the centerline of the road and in a direction of accumulated distance.

[0030] Among them, the second coordinate axis of the target coordinate system is always distributed along the center line of the road, and the direction is the direction of distance accumulation, that is, as the distance traveled by the unmanned vehicle increases, the coordinate value of the unmanned vehicle on the second coordinate axis also increases accordingly. The first coordinate axis of the target coordinate system is perpendicular to the second coordinate axis (center line of the road) and points to the left. It can be understood that for a given obstacle, the coordinate value of its left boundary is greater than the coordinate value of its right boundary.

[0031] It should be noted that the technical solution of the present application is applicable to a target coordinate system that complies with the right-hand rule. If the present solution is adaptively adjusted to comply with a coordinate system that complies with the left-hand rule, then the adjusted technical solution is an equivalent technical solution of the present application.

[0032] The target coordinate system is preferably but not limited to the Frenet coordinate system. This embodiment uses the Frenet coordinate system as an example to illustrate the technical solution. In the Frenet coordinate system, the S axis (the second coordinate axis) is distributed along the center line of the road, and the L axis (the first coordinate axis) is perpendicular to the center line of the road.

[0033] like Figure 2As shown, the road that the unmanned vehicle is traveling on is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, that is, between two curbs. There are one or more obstacles distributed on the road, each obstacle is pre-marked with a detour direction, and the detour directions corresponding to each obstacle may be the same or different.

[0034] When the obstacle is bypassed in the left direction, see Figure 2 When the unmanned vehicle is near obstacles M1, M2 and M3, it needs to execute the left detour strategy. When the detour direction is right, refer to Figure 2 When the unmanned vehicle drives near obstacles M4, M4 and M5, it needs to execute the right bypass strategy.

[0035] In order to improve the driving safety of the unmanned vehicle, this embodiment obtains the detour label of each obstacle in the obstacle set in real time to obtain the latest detour direction of each obstacle in the obstacle set.

[0036] If a first obstacle marked with a left detour is detected, at least two left boundary points of the obstacle in the target coordinate system are determined. Figure 2 The M1 in , whose detour direction is left, thus determines at least two left boundary points in the Frenet coordinate system, that is, at least two boundary points on the side where the point p1 is located.

[0037] Similarly, if a second obstacle marked with a right detour is detected, at least two right boundary points of the obstacle in the target coordinate system are determined. Figure 2 The M4 in the figure has a detour direction of right, so at least two right boundary points of the M4 in the Frenet coordinate system are determined, that is, at least two boundary points on the side where the point p1' is located.

[0038] S102. Determine a detour constraint boundary of the unmanned vehicle according to a maximum constraint boundary of the road, a minimum constraint boundary of the road, a left boundary of the obstacle defined by the at least two left boundary points, and / or a right boundary of the obstacle defined by the at least two right boundary points.

[0039] It can be understood that the left boundary of the obstacle can be determined based on the at least two left boundary points, and as the number of the determined left boundary points of the obstacle increases, the determined left boundary of the obstacle becomes more accurate; similarly, the right boundary of the obstacle can be determined based on the at least two right boundary points, and as the number of the determined right boundary points of the obstacle increases, the determined right boundary of the obstacle becomes more accurate.

[0040] If the determined left boundary of the obstacle is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, the left boundary and the maximum constraint boundary of the road are used as the detour constraint boundary of the unmanned vehicle. Figure 2 In M1, the left boundary of the obstacle, that is, the boundary where point P1 is located, is larger than the minimum constraint boundary of the road, but smaller than the maximum constraint boundary of the road. At this time, the unmanned vehicle can bypass the obstacle in the area between the left boundary of the obstacle and the maximum constraint boundary of the road. Therefore, the left boundary of the obstacle and the maximum constraint boundary of the road are used as the bypass constraint boundaries of the unmanned vehicle.

[0041] Similarly, if the right boundary of the determined obstacle is between the maximum constraint boundary and the minimum constraint boundary of the road, the right boundary and the minimum constraint boundary of the road are used as the detour constraint boundary of the unmanned vehicle. Figure 2 In M4, the right boundary of the obstacle is larger than the minimum constraint boundary of the road and smaller than the maximum constraint boundary of the road. At this time, the unmanned vehicle can perform a right detour around the obstacle in the area between the right boundary of the obstacle and the minimum constraint boundary of the road. Therefore, the right boundary of the obstacle and the minimum constraint boundary of the road are used as the detour constraint boundaries of the unmanned vehicle.

[0042] If the determined left boundary of the obstacle is smaller than the minimum constraint boundary of the road, the minimum constraint boundary of the road and the maximum constraint boundary of the road are used as the detour constraint boundary of the unmanned vehicle. Figure 2 In M2, the left boundary of the obstacle is smaller than the minimum constraint boundary of the road. At this time, the unmanned vehicle can travel on the entire road. Therefore, the minimum constraint boundary of the road and the maximum constraint boundary of the road are used as the detour constraint boundaries of the unmanned vehicle.

[0043] Similarly, if the right boundary of the determined obstacle is larger than the maximum constraint boundary of the road, the minimum constraint boundary of the road and the maximum constraint boundary of the road are used as the detour constraint boundary of the unmanned vehicle. Figure 2 In M5, the right boundary of the obstacle is larger than the maximum constraint boundary of the road. At this time, the unmanned vehicle can travel on the entire road. Therefore, the minimum constraint boundary of the road and the maximum constraint boundary of the road are used as the detour constraint boundaries of the unmanned vehicle.

[0044] It is understandable that if the left boundary of the obstacle exceeds the maximum constraint boundary of the road, it is considered that the unmanned vehicle cannot complete the left bypass operation of the obstacle on the current road, so it is determined that there is no bypass constraint boundary for the unmanned vehicle on the current road. Similarly, if the right boundary of the obstacle exceeds the minimum constraint boundary of the road, it is determined that there is no bypass constraint boundary for the unmanned vehicle on the current road.

[0045] S103: Control the unmanned vehicle to travel within the detour constraint boundary to detour through the corresponding obstacle.

[0046] After the detour constraint boundary is determined, the path optimizer can optimize the detour path of the unmanned vehicle according to the detour constraint boundary. Since the detour constraint boundary is the maximum boundary determined based on the outer boundary of the obstacle, the path optimizer can relatively easily optimize the detour path of the unmanned vehicle based on the detour constraint boundary, thereby significantly improving the optimization efficiency of the detour path. It can be understood that the detour path is located inside the detour constraint boundary.

[0047] After the detour path is determined, the unmanned vehicle controller can control the unmanned vehicle to travel along the detour path to bypass the corresponding obstacles along the road.

[0048] The technical solution of the obstacle bypassing method for an unmanned vehicle provided in an embodiment of the present invention is as follows: if a first obstacle marked with a left bypass is detected, at least two left boundary points of the first obstacle in the target coordinate system are determined; if a second obstacle marked with a right bypass is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined, and the at least two left boundary points or the right boundary points are all on the obstacle boundary. Therefore, according to the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, as well as the maximum constraint boundary of the road and the minimum constraint boundary of the road, the obstacle area and the non-obstacle area in the road can be accurately divided, so that the maximum bypass constraint boundary of the unmanned vehicle on the road can be accurately determined, and the maximum bypass constraint boundary can greatly reduce the difficulty of the optimizer in performing path optimization, and improve the speed and accuracy of the optimizer's path optimization.

[0049] Embodiment 2

[0050] Figure 3 1 is a flow chart of the obstacle bypassing method for an unmanned vehicle provided in the second embodiment of the present invention. Based on the above embodiment, the present embodiment refines the method for determining the left boundary point and the right boundary point of the obstacle.

[0051] Accordingly, the method of this embodiment includes:

[0052] S201. In an initial coordinate system, determine a perpendicular line perpendicular to a road centerline where a current target point in a target point set is located, wherein each target point in the target point set is distributed on the road centerline, and each obstacle corresponds to at least two target points.

[0053] Based on the target coordinate system or the initial coordinate system, multiple target points are determined on the road centerline at preset intervals, such as 0.1 meters, and the determined target points constitute a target point set. The target coordinate system is preferably, but not limited to, the Frenet coordinate system, and the initial coordinate system is preferably, but not limited to, the Cartesian coordinate system. The preset interval can be based on the length of each obstacle along the road centerline, so that each obstacle corresponds to at least two target points.

[0054] Traverse each target point in the target point set, and determine the perpendicular line perpendicular to the center line of the road where the current target point is located in the initial coordinate system. When determining the perpendicular line perpendicular to the center line of the road where the current target point is located, first determine the direction of the current target point, and then make a perpendicular line perpendicular to the direction and passing through the current target point. The direction of the target point is the vector direction of the target point.

[0055] It can be understood that if each target point in the target point set is determined in the target coordinate system, the position of each target point in the initial coordinate system is determined, and the vertical line perpendicular to the center line of the road where each target point is located is determined according to the position of each target point in the initial coordinate system. If each target point in the target point set is determined in the initial coordinate system, the vertical line perpendicular to the center line of the road where each target point is located is directly determined in the initial coordinate system.

[0056] The length of the vertical line determines the distribution range of obstacles that can be detected. The longer the vertical line is, the larger the distribution range of obstacles that can be detected is. The length of the vertical line is at least equal to the road width. In one embodiment, the length of the vertical line is three times the road width.

[0057] S202: If there are two intersection points between the vertical line and at least one obstacle, obtain a circumventing direction of the at least one obstacle.

[0058] Check whether the vertical line of the current target point passes through at least one obstacle and has two intersections with the passed obstacle. If so, obtain the detour direction of the obstacle that has two intersections with the vertical line based on the coordinates of the current target point and the projection range of each obstacle on the center line of the road.

[0059] For example, see Figure 2 As shown, the vertical line AB where the target point r is located passes through the obstacle M1 and the obstacle M4, and has two intersection points with the obstacle M1 and the obstacle M4, so the detour directions of the obstacle M1 and the obstacle M4 are obtained respectively.

[0060] S203: If the obtained detour direction of the obstacle is left detour, the intersection point with the largest coordinate value among the two intersection points of the vertical line and the obstacle boundary is taken as the left boundary point, and the coordinate value of the left boundary point in the target coordinate system is determined.

[0061] like Figure 2As shown, the L-axis (first coordinate axis) coordinate value of the left boundary of the obstacle is greater than the L-axis coordinate value of its right boundary. Therefore, if the detour direction corresponding to the obstacle through which the vertical line passes is left detour, the intersection point with the largest coordinate value among the two intersection points of the vertical line and the boundary of the obstacle is determined, and the intersection point is used as the left boundary point of the obstacle, and then the coordinate value of the left boundary point in the target coordinate system is determined according to the coordinate value of the left boundary point in the initial coordinate system. It can be understood that for a given obstacle, the smaller the preset interval between each target point in the target point set, the more target points the obstacle corresponds to; and the more target points there are, the more left boundary points corresponding to the target points; and the more corresponding left boundary points there are, the more accurate the left boundary of the obstacle determined based on these left boundary points.

[0062] Before determining the coordinate value of the left boundary point in the target coordinate system according to the coordinate value of the left boundary point in the initial coordinate system, the conversion relationship between the coordinate value of the discrete point in the initial coordinate system and the coordinate value in the target coordinate system is first determined. The details are as follows:

[0063] like Figure 4 As shown in the figure, the Cartesian coordinate system is XY, and the Frenet coordinate system is SL. In the Frenet coordinate system, the S axis is along the centerline of the road, and the L axis is perpendicular to the centerline of the road. The centerline of the road is composed of a series of discrete points. The coordinates of point p in the Cartesian coordinate system can be expressed as p(x p ,y p ). Find the two points on the road centerline that are closest to point p, namely point s and point e. If the coordinates of point s in the Frenet coordinate system are (s s ,0), the coordinates of point e in the Frenet coordinate system are (s e ,0), then according to the coordinate value of point p in the initial coordinate system (Cartesian coordinate system), the steps to determine the coordinate value of point p in the target coordinate system (Frenet coordinate system) are as follows:

[0064] The vector from point s to point p can be expressed as:

[0065]

[0066] The vector from point e to point p can be expressed as:

[0067]

[0068] The vector is calculated by the following formula With vector Conversion factor between

[0069]

[0070]

[0071] Therefore, p It can be expressed as:

[0072]

[0073] l p It can be expressed as:

[0074]

[0075] S204: If the obtained detour direction of the obstacle is right detour, the intersection point with the smallest coordinate value among the two intersection points of the vertical line and the obstacle boundary is taken as the right boundary point, and the coordinate value of the right boundary point in the target coordinate system is determined.

[0076] like Figure 2 As shown, the L-axis (first coordinate axis) coordinate value of the right boundary of the obstacle is smaller than the L-axis coordinate value of its left boundary. If the detour direction corresponding to the obstacle through which the vertical line passes is right detour, the intersection point with the smallest coordinate value among the two intersection points of the vertical line and the boundary of the obstacle is determined, and the intersection point is used as the right boundary point of the obstacle, and then the coordinate value of the right boundary point in the target coordinate system is determined according to the coordinate value of the right boundary point in the initial coordinate system.

[0077] It can be understood that for a given obstacle, the smaller the preset interval between each target point in the target point set, the more target points the obstacle corresponds to; and the more target points there are, the more right boundary points the target points correspond to; and the more corresponding right boundary points there are, the more accurate the right boundary of the obstacle determined based on these right boundary points.

[0078] S205. Determine the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points.

[0079] The left boundary of the autonomous vehicle may be completely larger than the minimum constraint boundary of the road (see Figure 2 M1 in ), or it may be completely smaller than the minimum road constraint boundary (see Figure 2 M2 in ), it is also possible that part of it is smaller than the minimum road constraint boundary, and part of it is larger than the minimum road constraint boundary (see Figure 2M3 in the figure). To this end, the present embodiment compares any determined left boundary point in real time with the size of the corresponding road minimum constraint boundary point and the corresponding road maximum constraint boundary point. If any determined left boundary point is smaller than the corresponding road minimum constraint boundary point, the corresponding road maximum constraint boundary point and the corresponding road minimum constraint boundary point are used as the detour constraint boundary points of the unmanned vehicle; if any determined left boundary point is between the corresponding road maximum constraint boundary and the corresponding road minimum constraint boundary, the corresponding road maximum constraint boundary point and the left boundary point are used as the detour constraint boundary points of the unmanned vehicle. Among them, the road maximum constraint boundary point is located at the road maximum constraint boundary, and the road minimum constraint boundary point is located at the road minimum constraint boundary.

[0080] Similarly, this embodiment compares any determined right boundary point in real time with the corresponding road minimum constraint boundary point and the corresponding road maximum constraint boundary point. If any determined right boundary point is larger than the corresponding road maximum constraint boundary point, the corresponding road maximum constraint boundary point and the corresponding road minimum constraint boundary point are used as the bypass constraint boundary points of the unmanned vehicle; if any determined right boundary point is between the corresponding road maximum constraint boundary point and the road minimum constraint boundary point, the corresponding right boundary point and the corresponding road minimum constraint boundary point are used as the bypass constraint boundary points of the unmanned vehicle.

[0081] It can be understood that if the currently determined left boundary point is greater than the corresponding road maximum constraint boundary point, it is determined that the unmanned vehicle cannot complete the left bypass of the obstacle on the current driving path, that is, there is no bypass constraint boundary; similarly, if the currently determined right boundary point is less than the corresponding road minimum constraint boundary point, it is determined that the unmanned vehicle cannot complete the right bypass of the obstacle on the current driving path, that is, there is no bypass constraint boundary.

[0082] In some embodiments, the left boundary point of the obstacle marked with a left detour and the right boundary point of the obstacle marked with a right detour may be obtained at the same time, for example Figure 2 The current target point r in corresponds to the left boundary point of obstacle M1 and the right boundary point of obstacle M4. At this time, determine whether the right boundary point is greater than the left boundary point and less than the corresponding maximum constraint boundary point of the road, and whether the left boundary point is greater than the corresponding minimum constraint boundary point of the road. If so, the right boundary point and the left boundary point are used as the bypass constraint boundary points of the unmanned vehicle, that is, the unmanned vehicle needs to complete the current bypass operation between the left boundary point and the right boundary point.

[0083] In some embodiments, for obstacles marked with a left detour, see Figure 2 For the obstacle M1, after determining the two intersection points of the vertical line perpendicular to the center line and the obstacle where the current target point is located, the right boundary of the detour constraint boundary is determined by the following formula:

[0084] l 右 =std::clamp(max(l1,l2),l min ,l max )

[0085] Among them, l1 and l2 are the L-axis (first coordinate axis) coordinate values ​​of the intersection point p1 and the intersection point p2 respectively. The maximum L-axis coordinate value of the two, that is, the L-axis coordinate value of point p1, is taken as the L-axis coordinate value of the current left boundary point, denoted as l. min <l max , so when l≤l min When l is min , that is, the minimum constraint boundary point of the road, at l min <l<l max When l≥l max When l is max , which is the maximum constraint boundary point of the road. The left boundary of the detour constraint boundary is the maximum constraint boundary point of the road. The left boundary and the right boundary together form the road constraint boundary, and when l≥l max , there is no detour constraint boundary point.

[0086] Similarly, the left boundary of the detour constraint boundary with the right detour obstacle can be determined by the following formula:

[0087] l 左 =std::clamp(min(l1,l2),l min ,l max )

[0088] Among them, l1 and l2 are the L-axis (first coordinate axis) coordinate values ​​of the intersection point p′1 and the intersection point p′2 respectively. The maximum L-axis coordinate value of the two, that is, the L-axis coordinate value of point p′1, is taken as the L-axis coordinate value of the current left boundary point, denoted as l. min <l max , so when l>l max When l is max , that is, the maximum constraint boundary point of the road, at l min <l<l max When l≤l min When l is max , which is the minimum constraint boundary point of the road. The right boundary of the detour constraint boundary is the minimum constraint boundary point of the road. The left boundary and the right boundary together form the road constraint boundary, and when l≤l min , there is no detour constraint boundary point.

[0089] S206: Control the unmanned vehicle to travel within the detour constraint boundary to detour through the corresponding obstacle.

[0090] Compared with the prior art, the technical solution provided by the embodiment of the present invention determines a vertical line perpendicular to the center line of the road where the current target point is located in the initial coordinate system. When the obstacle detour direction is left, the intersection point with the largest coordinate value among the intersection points of the vertical line and the obstacle is used as the left boundary point of the obstacle corresponding to the current target point; when the obstacle detour direction is right, the intersection point with the smallest coordinate value among the intersection points of the vertical line and the obstacle is used as the right boundary point of the obstacle corresponding to the current target point. Since each obstacle corresponds to at least two target points, each obstacle corresponds to at least two left boundary points or at least two right boundary points. Since the at least two left boundary points are located on the left boundary of the obstacle, and the at least two right boundary points are located on the right boundary of the obstacle, the decomposition line of the obstacle area and the non-obstacle area on the road can be accurately determined according to the at least two left boundary points and / or the at least two right boundary points, so that the determined detour constraint boundary is the maximum boundary that the unmanned vehicle can detour, which reduces the difficulty of the path optimizer to perform path optimization according to the detour constraint boundary, and improves the speed of path optimization and the accuracy of path optimization results.

[0091] Embodiment 3

[0092] Figure 5 : is a structural block diagram of an unmanned vehicle obstacle bypassing device provided in an embodiment of the present invention. The device is used to execute the unmanned vehicle obstacle bypassing method provided in any of the above embodiments, and the device can be implemented in software or hardware. The device includes:

[0093] The boundary point module 11 is used to determine at least two left boundary points of the first obstacle in the target coordinate system if a first obstacle marked with a left detour is detected; and to determine at least two right boundary points of the second obstacle in the target coordinate system if a second obstacle marked with a right detour is detected, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in a direction of accumulated distance;

[0094] A determination module 12, configured to determine a detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points;

[0095] The driving module 13 is used to control the unmanned vehicle to drive within the detour constraint boundary to detour through the corresponding obstacles.

[0096] Optionally, the boundary point module 11 is used to determine, in the initial coordinate system, a vertical line perpendicular to the center line of the road where the current target point in the target point set is located, wherein each target point in the target point set is distributed on the center line of the road, and each obstacle corresponds to at least two target points; if there are two intersection points between the vertical line and at least one obstacle, the detour direction of the at least one obstacle is obtained; if the detour direction of the obstacle obtained is left detour, the intersection point with the largest coordinate value between the two intersection points of the vertical line and the boundary of the obstacle is taken as the left boundary point, and the coordinate value of the left boundary point in the target coordinate system is determined; if the detour direction of the obstacle obtained is right detour, the intersection point with the smallest coordinate value between the two intersection points of the vertical line and the boundary of the obstacle is taken as the right boundary point, and the coordinate value of the right boundary point in the target coordinate system is determined.

[0097] Optionally, the determination module 12 is used to, if any determined left boundary point exceeds the minimum constraint boundary of the road, then use the corresponding maximum constraint boundary point of the road and the corresponding minimum constraint boundary point of the road as the detour constraint boundary points of the unmanned vehicle, the maximum constraint boundary point of the road is located on the maximum constraint boundary of the road, and the minimum constraint boundary point of the road is located on the minimum constraint boundary of the road; if any determined left boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then use the corresponding maximum constraint boundary point of the road and the left boundary point as the detour constraint boundary points of the unmanned vehicle.

[0098] Optionally, the determination module 12 is used to, if any determined right boundary point exceeds the maximum constraint boundary of the road, then use the corresponding maximum constraint boundary point of the road and the corresponding minimum constraint boundary point of the road as the detour constraint boundary points of the unmanned vehicle, the maximum constraint boundary point of the road is located on the maximum constraint boundary of the road, and the minimum constraint boundary point of the road is located on the minimum constraint boundary of the road; if any determined right boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then use the corresponding minimum constraint boundary point of the road and the right boundary point as the detour constraint boundary points of the unmanned vehicle.

[0099] Optionally, the determination module 12 is also used to determine that the unmanned vehicle cannot bypass the current obstacle if any of the determined left boundary points exceeds the maximum constraint of the road; if any of the determined right boundary points exceeds the minimum constraint of the road, then determine that the unmanned vehicle cannot bypass the current obstacle.

[0100] Optionally, the determination module 12 is used to, when a left boundary point of an obstacle marked as a left detour and a right boundary point of an obstacle marked as a right detour are obtained at the same time, if the right boundary point is larger than the left boundary point and smaller than the corresponding maximum constraint boundary point of the road, and the left boundary point is larger than the corresponding minimum constraint boundary point of the road, use the right boundary point and the left boundary point as the detour constraint boundary points of the unmanned vehicle.

[0101] The technical solution of the obstacle bypassing device for an unmanned vehicle provided in an embodiment of the present invention is as follows: if a first obstacle marked with a left bypass is detected, at least two left boundary points of the first obstacle in a target coordinate system are determined; if a second obstacle marked with a right bypass is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined. Since the at least two left boundary points or the at least two right boundary points are both on the obstacle boundary, the obstacle left boundary defined by the at least two left boundary points and / or the obstacle right boundary defined by the at least two right boundary points, as well as the maximum constraint boundary of the road and the minimum constraint boundary of the road, the obstacle area and the non-obstacle area in the road can be accurately divided, so that the maximum bypass constraint boundary of the unmanned vehicle on the road can be accurately determined. The maximum bypass constraint boundary can greatly reduce the difficulty of the optimizer in performing path optimization, and improve the speed and accuracy of the optimizer's path optimization.

[0102] The unmanned vehicle obstacle bypassing device provided in the embodiment of the present invention can execute the unmanned vehicle obstacle bypassing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0103] Embodiment 4

[0104] Figure 6 is a structural block diagram of an unmanned vehicle provided by an embodiment of the present invention. The controller of the unmanned vehicle is used to execute the unmanned vehicle obstacle bypassing method provided by any of the above embodiments. The unmanned vehicle includes: a vehicle body, a driving mechanism 21 and a controller 22. The driving mechanism 21 is arranged on the vehicle body, and is used to drive the vehicle body to travel. The controller 22 is arranged in the vehicle body, and is used to determine at least two left boundary points of the first obstacle in the target coordinate system if a first obstacle marked with a left bypass is detected; if a second obstacle marked with a right bypass is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined; according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, the bypass constraint boundary of the unmanned vehicle is determined, and the vehicle body is controlled to travel within the bypass constraint boundary to bypass the current obstacle, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in the direction of the accumulated distance.

[0105] Among them, the second coordinate axis of the target coordinate system is always distributed along the center line of the road, and the direction is the direction of distance accumulation, that is, as the distance traveled by the unmanned vehicle increases, the coordinate value of the unmanned vehicle on the second coordinate axis also increases accordingly. The first coordinate axis of the target coordinate system is perpendicular to the second coordinate axis (center line of the road) and points to the left. It can be understood that for a given obstacle, the coordinate value of its left boundary is greater than the coordinate value of its right boundary.

[0106] It should be noted that the technical solution of the present application is applicable to a target coordinate system that complies with the right-hand rule. If the present solution is adaptively adjusted to comply with a coordinate system that complies with the left-hand rule, then the adjusted technical solution is an equivalent technical solution of the present application.

[0107] The target coordinate system is preferably but not limited to the Frenet coordinate system. This embodiment uses the Frenet coordinate system as an example to illustrate the technical solution. In the Frenet coordinate system, the S axis (the second coordinate axis) is distributed along the center line of the road, and the L axis (the first coordinate axis) is perpendicular to the center line of the road.

[0108] like Figure 2 As shown, the road that the unmanned vehicle is traveling on is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, that is, between two curbs. There are one or more obstacles distributed on the road, each obstacle is pre-marked with a detour direction, and the detour directions corresponding to each obstacle may be the same or different.

[0109] When the obstacle is bypassed in the left direction, see Figure 2 When the unmanned vehicle is near obstacles M1, M2 and M3, it needs to execute the left detour strategy. When the detour direction is right, refer to Figure 2 When the unmanned vehicle drives near obstacles M4, M4 and M5, it needs to execute the right bypass strategy.

[0110] In order to improve the driving safety of the unmanned vehicle, this embodiment obtains the detour label of each obstacle in the obstacle set in real time to obtain the latest detour direction of each obstacle in the obstacle set.

[0111] If a first obstacle marked with a left detour is detected, at least two left boundary points of the obstacle in the target coordinate system are determined. Figure 2 The detour direction of M1 in is left, so at least two left boundary points of it in the Frenet coordinate system are determined, that is, at least two boundary points of the side where point p1 is located. Similarly, if a second obstacle marked with right detour is detected, at least two right boundary points of the obstacle in the target coordinate system are determined. For example, see Figure 2 The M4 in the figure has a detour direction of right, so at least two right boundary points of the M4 in the Frenet coordinate system are determined, that is, at least two boundary points on the side where the point p1' is located.

[0112] The at least two left boundary points and the at least two right boundary points are determined by the following steps.

[0113] Based on the target coordinate system or the initial coordinate system, multiple target points are determined on the road centerline at preset intervals, such as 0.1 meters, and the determined target points constitute a target point set. The target coordinate system is preferably, but not limited to, the Frenet coordinate system, and the initial coordinate system is preferably, but not limited to, the Cartesian coordinate system. The preset interval can be based on the length of each obstacle along the road centerline, so that each obstacle corresponds to at least two target points.

[0114] Traverse each target point in the target point set, and determine the perpendicular line perpendicular to the center line of the road where the current target point is located in the initial coordinate system. When determining the perpendicular line perpendicular to the center line of the road where the current target point is located, first determine the direction of the current target point, and then make a perpendicular line perpendicular to the direction and passing through the current target point. The direction of the target point is the vector direction of the target point.

[0115] It is understandable that if each target point in the target point set is determined in the target coordinate system, the position of each target point in the initial coordinate system is determined, and the vertical line perpendicular to the center line of the road where each target point is located is determined based on the position of each target point in the initial coordinate system. If each target point in the target point set is determined in the initial coordinate system, the vertical line perpendicular to the center line of the road where each target point is located is directly determined in the initial coordinate system. Among them, the length of the vertical line determines the distribution range of obstacles that can be detected. The longer the vertical line, the larger the distribution range of obstacles that can be detected. The length of the vertical line is at least equal to the width of the road. In one embodiment, the length of the vertical line is three times the width of the road.

[0116] Check whether the vertical line of the current target point passes through at least one obstacle and has two intersections with the passed obstacle. If so, obtain the detour direction of the obstacle that has two intersections with the vertical line according to the coordinates of the current target point and the projection range of each obstacle on the center line of the road. For example, see Figure 2 As shown, the vertical line AB where the target point r is located passes through the obstacle M1 and the obstacle M4, and has two intersection points with the obstacle M1 and the obstacle M4, so the detour directions of the obstacle M1 and the obstacle M4 are obtained respectively.

[0117] If the detour direction corresponding to the obstacle passed by the vertical line is left detour, then determine the intersection point with the largest coordinate value among the two intersection points of the vertical line and the boundary of the obstacle, and use the intersection point as the left boundary point of the obstacle, and then determine the coordinate value of the left boundary point in the target coordinate system based on the coordinate value of the left boundary point in the initial coordinate system. It can be understood that for a given obstacle, the smaller the preset interval between each target point in the target point set, the more target points the obstacle corresponds to; and the more target points there are, the more left boundary points corresponding to the target points; and the more corresponding left boundary points there are, the more accurate the left boundary of the obstacle determined based on these left boundary points.

[0118] If the detour direction corresponding to the obstacle through which the vertical line passes is right detour, then determine the intersection point with the smallest coordinate value among the two intersection points of the vertical line and the boundary of the obstacle, and use the intersection point as the right boundary point of the obstacle, and then determine the coordinate value of the right boundary point in the target coordinate system based on the coordinate value of the right boundary point in the initial coordinate system. It can be understood that for a given obstacle, the smaller the preset interval between each target point in the target point set, the more target points the obstacle corresponds to; and the more target points, the more right boundary points the target points correspond to; and the more corresponding right boundary points, the more accurate the right boundary of the obstacle determined based on these right boundary points.

[0119] After the at least two left boundary points of the obstacle are determined, the left boundary of the obstacle can be determined according to the at least two left boundary points; the detour constraint boundary of the unmanned vehicle is determined according to the size relationship between the left boundary of the obstacle and the maximum constraint boundary and the minimum constraint boundary of the road. Similarly, after the at least two right boundary points of the obstacle are determined, the right boundary of the obstacle can be determined according to the at least two right boundary points; the detour constraint boundary of the unmanned vehicle is determined according to the size relationship between the right boundary of the obstacle and the maximum constraint boundary and the minimum constraint boundary of the road.

[0120] Considering that the left boundary of the unmanned vehicle may be completely larger than the minimum constraint boundary of the road (see Figure 2 M1 in ), or it may be completely smaller than the minimum road constraint boundary (see Figure 2 M2 in ), it is also possible that part of it is smaller than the minimum road constraint boundary, and part of it is larger than the minimum road constraint boundary (see Figure 2M3 in). To this end, the present embodiment compares any determined left boundary point in real time with the size of the corresponding road minimum constraint boundary point and the corresponding road maximum constraint boundary point. If any determined left boundary point is smaller than the corresponding road minimum constraint boundary point, the corresponding road maximum constraint boundary point and the corresponding road minimum constraint boundary point are used as the detour constraint boundary points of the unmanned vehicle; if any determined left boundary point is between the corresponding road maximum constraint boundary and the corresponding road minimum constraint boundary, the corresponding road maximum constraint boundary point and the left boundary point are used as the detour constraint boundary points of the unmanned vehicle. Among them, the road maximum constraint boundary point is located at the road maximum constraint boundary, and the road minimum constraint boundary point is located at the road minimum constraint boundary. Similarly, the present embodiment compares any determined right boundary point in real time with the size of the corresponding road minimum constraint boundary point and the corresponding road maximum constraint boundary point. If any determined right boundary point is larger than the corresponding maximum constraint boundary point of the road, the corresponding maximum constraint boundary point of the road and the corresponding minimum constraint boundary point of the road are used as the detour constraint boundary points of the unmanned vehicle; if any determined right boundary point is between the corresponding maximum constraint boundary point of the road and the minimum constraint boundary point of the road, the corresponding right boundary point and the corresponding minimum constraint boundary point of the road are used as the detour constraint boundary points of the unmanned vehicle.

[0121] It can be understood that if the currently determined left boundary point is greater than the corresponding road maximum constraint boundary point, it is determined that the unmanned vehicle cannot complete the left bypass of the obstacle on the current driving path, that is, there is no bypass constraint boundary; similarly, if the currently determined right boundary point is less than the corresponding road minimum constraint boundary point, it is determined that the unmanned vehicle cannot complete the right bypass of the obstacle on the current driving path, that is, there is no bypass constraint boundary.

[0122] In some embodiments, the left boundary point of the obstacle marked with a left detour and the right boundary point of the obstacle marked with a right detour may be obtained at the same time, for example Figure 2 The current target point r in corresponds to the left boundary point of obstacle M1 and the right boundary point of obstacle M4. At this time, determine whether the right boundary point is greater than the left boundary point and less than the corresponding maximum constraint boundary point of the road, and whether the left boundary point is greater than the corresponding minimum constraint boundary point of the road. If so, the right boundary point and the left boundary point are used as the bypass constraint boundary points of the unmanned vehicle, that is, the unmanned vehicle needs to complete the current bypass operation between the left boundary point and the right boundary point.

[0123] The technical solution of the unmanned vehicle provided by the embodiment of the present invention determines at least two left boundary points of the first obstacle in the target coordinate system if a first obstacle marked with a left detour is detected; and determines at least two right boundary points of the second obstacle in the target coordinate system if a second obstacle marked with a right detour is detected. Since the at least two left boundary points or the at least two right boundary points are both on the obstacle boundary, the obstacle area and non-obstacle area in the road can be accurately divided according to the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, as well as the maximum constraint boundary of the road and the minimum constraint boundary of the road, so that the maximum detour constraint boundary of the unmanned vehicle on the road can be accurately determined. The maximum detour constraint boundary can greatly reduce the difficulty of the optimizer in performing path optimization, and improve the speed and accuracy of the optimizer's path optimization.

[0124] Embodiment 5

[0125] An embodiment of the present invention further provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to execute a method for unmanned vehicle to bypass obstacles when executed by a computer processor, the method comprising:

[0126] If a first obstacle marked with a left detour is detected, at least two left boundary points of the first obstacle in the target coordinate system are determined; if a second obstacle marked with a right detour is detected, at least two right boundary points of the second obstacle in the target coordinate system are determined, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in a direction of accumulated distance;

[0127] Determining a detour constraint boundary of the unmanned vehicle according to a maximum constraint boundary of the road, a minimum constraint boundary of the road, a left boundary of the obstacle defined by the at least two left boundary points, and / or a right boundary of the obstacle defined by the at least two right boundary points;

[0128] The unmanned vehicle is controlled to travel within the detour constraint boundary to detour through the corresponding obstacle.

[0129] Of course, the storage medium containing computer executable instructions provided in an embodiment of the present invention is not limited to the method operations described above, and the computer executable instructions can also execute related operations in the unmanned vehicle obstacle bypassing method provided in any embodiment of the present invention.

[0130] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the unmanned vehicle obstacle bypassing method described in each embodiment of the present invention.

[0131] It is worth noting that in the embodiment of the above-mentioned unmanned vehicle obstacle bypassing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0132] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for unmanned vehicle to bypass obstacles. It is characterized in that include: If a first obstacle marked with a left detour is detected, determining at least two left boundary points of the first obstacle in the target coordinate system; If a second obstacle marked with a right detour is detected, determining at least two right boundary points of the second obstacle in the target coordinate system, wherein the target coordinate system includes a first coordinate axis perpendicular to the centerline of the road and a second coordinate axis distributed along the centerline of the road and in a direction of accumulated distance; Determining the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, including: if any determined left boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then using the corresponding maximum constraint boundary point of the road and the left boundary point as the detour constraint boundary points of the unmanned vehicle; if any determined right boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then using the corresponding minimum constraint boundary point of the road and the right boundary point as the detour constraint boundary points of the unmanned vehicle; if a left boundary point of an obstacle marked with a left detour and a right boundary point of an obstacle marked with a right detour are obtained at the same time, and the right boundary point is greater than the left boundary point and less than the corresponding maximum constraint boundary point of the road, and the left boundary point is greater than the corresponding minimum constraint boundary point of the road, then using the right boundary point and the left boundary point as the detour constraint boundary points of the unmanned vehicle; The unmanned vehicle is controlled to travel within the detour constraint boundary to detour through the corresponding obstacle.

2. The method according to claim 1, It is characterized in that If a first obstacle marked with a left detour is detected, determining at least two left boundary points of the first obstacle in the target coordinate system; If a second obstacle marked with a right detour is detected, determining at least two right boundary points of the second obstacle in the target coordinate system includes: In the initial coordinate system, determine a vertical line perpendicular to the center line of the road where the current target point in the target point set is located, wherein each target point in the target point set is distributed on the center line of the road, and each obstacle corresponds to at least two target points; If there are two intersection points between the vertical line and at least one obstacle, obtaining a circumventing direction of the at least one obstacle; If the obtained detour direction of the obstacle is left detour, the intersection point with the largest coordinate value among the two intersection points of the vertical line and the boundary of the obstacle is taken as the left boundary point, and the coordinate value of the left boundary point in the target coordinate system is determined; If the obtained detour direction of the obstacle is right detour, the intersection point with the smallest coordinate value among the two intersection points of the vertical line and the obstacle boundary is taken as the right boundary point, and the coordinate value of the right boundary point in the target coordinate system is determined.

3. The method according to claim 2, It is characterized in that The length of the vertical line is greater than 3 times the width of the road.

4. The method according to claim 1, It is characterized in that Determining the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points includes: If any of the determined left boundary points exceeds the minimum constraint boundary of the road, the corresponding maximum constraint boundary point of the road and the corresponding minimum constraint boundary point of the road will be used as the detour constraint boundary points of the unmanned vehicle. The maximum constraint boundary point of the road is located on the maximum constraint boundary of the road, and the minimum constraint boundary point of the road is located on the minimum constraint boundary of the road.

5. The method according to claim 1, It is characterized in that Determining the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the right boundary of the obstacle defined by the at least two right boundary points includes: If any determined right boundary point exceeds the maximum constraint boundary of the road, the corresponding maximum constraint boundary point of the road and the corresponding minimum constraint boundary point of the road will be used as the detour constraint boundary points of the unmanned vehicle. The maximum constraint boundary point of the road is located on the maximum constraint boundary of the road, and the minimum constraint boundary point of the road is located on the minimum constraint boundary of the road.

6. The method according to claim 4 or 5, It is characterized in that Also includes: If any of the determined left boundary points exceeds the maximum road constraint, it is determined that the unmanned vehicle cannot bypass the current obstacle; If any of the determined right boundary points exceeds the minimum road constraint, it is determined that the unmanned vehicle cannot bypass the current obstacle.

7. An obstacle bypassing device for unmanned vehicles, It is characterized in that include: A boundary point module, configured to determine at least two left boundary points of the first obstacle in a target coordinate system if a first obstacle marked with a left detour is detected; If the detection strike mark has a second obstacle for right detour, at least two right boundary points of the second obstacle in the target coordinate system are determined, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in the direction of distance accumulation; A determination module, for determining the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, including: if any determined left boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then the corresponding maximum constraint boundary point of the road and the left boundary point are used as detour constraint boundary points of the unmanned vehicle; if any determined right boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then the corresponding minimum constraint boundary point of the road and the right boundary point are used as detour constraint boundary points of the unmanned vehicle; if a left boundary point of an obstacle marked with a left detour and a right boundary point of an obstacle marked with a right detour are obtained at the same time, and the right boundary point is greater than the left boundary point and less than the corresponding maximum constraint boundary point of the road, and the left boundary point is greater than the corresponding minimum constraint boundary point of the road, then the right boundary point and the left boundary point are used as detour constraint boundary points of the unmanned vehicle; The driving module is used to control the unmanned vehicle to drive within the detour constraint boundary to detour through the corresponding obstacle.

8. An unmanned vehicle, It is characterized in that include: Vehicle body; A traveling mechanism, arranged on the vehicle body, and used for driving the vehicle body to travel; A controller is disposed in the vehicle body and is used to determine at least two left boundary points of a first obstacle in a target coordinate system if a first obstacle marked with a left detour is detected; and to determine at least two right boundary points of the second obstacle in the target coordinate system if a second obstacle marked with a right detour is detected; and to determine the detour constraint boundary of the unmanned vehicle according to the maximum constraint boundary of the road, the minimum constraint boundary of the road, and the left boundary of the obstacle defined by the at least two left boundary points and / or the right boundary of the obstacle defined by the at least two right boundary points, including: if any determined left boundary point is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, then the corresponding maximum constraint boundary point of the road and the left boundary point are used as the detour constraint boundary points of the unmanned vehicle; if the determined If any right boundary point of the obstacle is between the maximum constraint boundary of the road and the minimum constraint boundary of the road, the corresponding minimum constraint boundary point of the road and the right boundary point are used as the detour constraint boundary points of the unmanned vehicle; if the left boundary point of the obstacle marked with left detour and the right boundary point of the obstacle marked with right detour are obtained at the same time, and the right boundary point is greater than the left boundary point and less than the corresponding maximum constraint boundary point of the road, and the left boundary point is greater than the corresponding minimum constraint boundary point of the road, the right boundary point and the left boundary point are used as the detour constraint boundary points of the unmanned vehicle; the vehicle body is controlled to travel within the detour constraint boundary to detour the current obstacle, wherein the target coordinate system includes a first coordinate axis perpendicular to the center line of the road and a second coordinate axis distributed along the center line of the road and in the direction of distance accumulation.

9. A storage medium containing computer executable instructions, It is characterized in that The computer executable instructions, when executed by a computer processor, are used to execute the unmanned vehicle obstacle bypassing method as described in any one of claims 1-6.

Images