Obstacle avoidance track generation method and device, equipment, medium, program product and vehicle

Abstract

The invention discloses an obstacle avoidance track generation method and device, equipment, a medium, a program product and a vehicle. A target obstacle having a collision risk with the vehicle is determined; according to the surrounding obstacles and the road edge information of the road where the vehicle is located, a driving space of the vehicle is constructed; based on the relative position relationship between the target obstacle and the vehicle, selecting an optimal drivable space; performing path planning in the optimal driving space to obtain an optimal collision avoidance path; and adding speed information to the path points in the optimal collision avoidance path to obtain an optimal collision avoidance track. According to the embodiment of the invention, the optimal drivable space is constructed based on the road edge and the obstacle, the environment information is converted into the constraint information available for trajectory planning, the collision avoidance trajectory planning is performed in the constructed optimal drivable space, and the obstacle information in the surrounding environment is considered in the planning stage. The relative relation between the vehicle and the obstacle can be better processed, and then the emergency collision avoidance track can be generated in a more complex scene.

Description

Technical Field

[0001] This application belongs to the field of intelligent driving technology, and in particular relates to a method, device, equipment, medium, program product and vehicle for generating obstacle avoidance trajectory. Background Technology

[0002] During vehicle operation, road conditions are often complex and obstacles are common. These obstacles can interfere with the vehicle's movement. To ensure safe operation and avoid collisions with obstacles, vehicles need to perform obstacle avoidance maneuvers.

[0003] Currently, with the development of autonomous driving technology, many vehicles are equipped with autonomous driving functions. Vehicles with autonomous driving functions typically perform obstacle avoidance by generating obstacle avoidance trajectories. Existing methods for generating obstacle avoidance trajectories mainly include selecting multiple sets of the vehicle's obstacle avoidance endpoint states, generating multiple sets of trajectories based on the target endpoint states of the vehicle's current state through polynomial curve fitting, performing safety verification and priority ranking on each set of trajectories, and selecting the one with the highest priority as the obstacle avoidance trajectory.

[0004] However, the above method generates a large number of candidate trajectories by sampling a large number of endpoint states, resulting in low overall efficiency and a low success rate when there are many obstacles in the environment. As a result, the above method of generating obstacle avoidance trajectories can only cover simple scenarios with few other traffic participants. Summary of the Invention

[0005] This application provides an obstacle avoidance trajectory generation method, apparatus, device, medium, program product, and vehicle, which can generate collision avoidance trajectories in complex scenarios and can adapt to more scenarios.

[0006] In a first aspect, embodiments of this application provide a collision avoidance trajectory generation method, including:

[0007] Identify the target obstacles that pose a collision risk to the vehicle from among the obstacles around the vehicle;

[0008] Based on the obstacles around the vehicle and the information about the road edge where the vehicle is located, construct the vehicle's driving space;

[0009] Based on the relative positional relationship between the target obstacle and the vehicle, the optimal drivable space is selected from the drivable space.

[0010] Path planning is performed within the optimal drivable space to obtain the optimal collision avoidance path;

[0011] Add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

[0012] Secondly, embodiments of this application provide a collision avoidance trajectory generation device, characterized in that it includes:

[0013] The risk assessment module is used to identify target obstacles that pose a collision risk to the vehicle from among the obstacles around the vehicle.

[0014] The space construction module is used to construct the vehicle's drivable space based on the obstacles around the vehicle and the curb information of the road where the vehicle is located;

[0015] The space filtering module is used to select the optimal drivable space from the drivable space based on the relative positional relationship between the target obstacle and the vehicle.

[0016] The path planning module is used to plan a path within the optimal drivable space to obtain the optimal collision avoidance path.

[0017] The trajectory generation module is used to add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

[0018] Thirdly, embodiments of this application provide an electronic device, the electronic device comprising: a processor and a memory storing computer program instructions;

[0019] When the processor executes the computer program instructions, it implements the steps of the collision avoidance trajectory generation method as described in the first aspect.

[0020] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the steps of the collision avoidance trajectory generation method as described in the first aspect.

[0021] Fifthly, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, enable the electronic device to perform the collision avoidance trajectory generation method as described in the first aspect.

[0022] Sixthly, embodiments of this application provide a vehicle including the electronic equipment described in the third aspect.

[0023] The obstacle avoidance trajectory generation method, apparatus, device, medium, program product, and vehicle of this application embodiment construct an optimal drivable space based on road curb information and obstacle information in the road, transforming environmental information into constraint information usable by the planning module, and performing collision avoidance trajectory planning in the constructed optimal drivable space. By considering obstacle information in the surrounding environment during the planning stage, compared to only considering the endpoint state, it can better handle the relative relationship between the vehicle and obstacles in the environment, thereby achieving the effect of generating emergency collision avoidance trajectories in more complex scenarios. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating the collision avoidance trajectory generation method provided in an embodiment of this application;

[0026] Figure 2 This is a schematic diagram of the natural coordinate system provided in the embodiments of this application;

[0027] Figure 3 This is a schematic diagram of projection space compensation provided in an embodiment of this application;

[0028] Figure 4 This is a schematic diagram of an obstacle avoidance scenario provided in an embodiment of this application;

[0029] Figure 5 This is a schematic diagram of another obstacle avoidance scenario provided in an embodiment of this application;

[0030] Figure 6 This is a schematic diagram of another obstacle avoidance scenario provided in the embodiments of this application;

[0031] Figure 7 This is a schematic diagram of the collision avoidance trajectory generation device provided in the embodiments of this application;

[0032] Figure 8 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0033] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0034] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0036] To address the problems of the prior art, embodiments of this application provide a method, apparatus, device, medium, program product, and vehicle for generating collision avoidance trajectories. The collision avoidance trajectory generation method provided in this application embodiment will be described first below.

[0037] Figure 1 A flowchart illustrating a collision avoidance trajectory generation method according to an embodiment of this application is shown. Figure 1 As shown, the method may include the following steps S11-S15, which will be described below.

[0038] S11. Identify the target obstacle that poses a collision risk to the vehicle from among the obstacles around the vehicle.

[0039] In some embodiments of this application, the vehicle is equipped with an obstacle detection function, which can detect obstacles around the vehicle. The area around the vehicle refers to a preset range around the vehicle, the size of which is set according to actual conditions or manually. Therefore, when the vehicle is traveling on a road and the obstacle detection function detects an obstacle within the preset range around the vehicle, the aforementioned step S11 can be executed.

[0040] In some embodiments of this application, step S11 above, when determining the target obstacle, may include:

[0041] Based on the roadside information of the road where the vehicle is located, select N obstacles that are closest to the vehicle in the direction of travel, located in front of or to the sides of the vehicle, and are located within the roadside as candidate obstacles.

[0042] Each candidate obstacle is assessed for collision risk to determine whether there is a risk of collision between the vehicle and each candidate obstacle.

[0043] Select alternative obstacles that pose a collision risk to the vehicle as target obstacles.

[0044] Here, N is a positive integer. The value of N can be set according to the actual situation or manually. For example, the value of N can be 3. By setting the number of candidate obstacles, the problem of large data processing volume caused by selecting too many candidate obstacles can be avoided.

[0045] When selecting candidate obstacles, the vehicle's centerline along its direction of travel can be used as the center line. Two expansion spaces of a preset width are then created on either side of this centerline. From the obstacles within these two expansion spaces, N candidate obstacles are selected that are closest to the vehicle, located in front of or to either side of it along its direction of travel. The preset widths on the left and right sides can be the same or different. These preset widths can be set based on road edge information to ensure that the expansion spaces based on the preset widths are located within the road edge, thus ensuring that the selected candidate obstacles are also located within the road edge.

[0046] The above method first selects at least one obstacle located within the curb and surrounding the vehicle as a candidate obstacle based on road environment information. Then, a collision risk assessment is performed on the candidate obstacle to determine the target obstacle. Compared to performing a collision risk assessment on all obstacles around the vehicle, this method can effectively reduce the amount of data processing.

[0047] After obtaining the alternative obstacles, each alternative obstacle can be assessed separately using a preset collision risk assessment method to determine whether there is a collision risk between the alternative obstacle and the vehicle.

[0048] In some embodiments of this application, for each candidate obstacle, the following collision risk assessment method can be used to determine whether there is a collision risk with the vehicle:

[0049] Determine the speed difference between the vehicle and the candidate obstacles;

[0050] The minimum safe distance between the vehicle and the candidate obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter, where the first deceleration parameter is twice the maximum deceleration of the vehicle.

[0051] Compare the actual distance between the vehicle and the candidate obstacle with the minimum safe distance;

[0052] If the actual distance is less than the minimum safe distance, a collision risk is determined between the candidate obstacle and the vehicle; if the actual distance is greater than or equal to the minimum safe distance, no collision risk is determined between the candidate obstacle and the vehicle.

[0053] Here, the speed difference between the vehicle and the candidate obstacle can be obtained by subtracting the speed of the candidate obstacle from the speed of the vehicle. The maximum deceleration of the vehicle is a fixed parameter of the vehicle and can be obtained directly. The minimum safe distance between the vehicle and the candidate obstacle refers to the minimum braking distance required for the vehicle to brake at its maximum deceleration. After obtaining the speed difference and the maximum deceleration, the minimum safe distance between the vehicle and the candidate obstacle can be calculated according to the displacement calculation formula, as follows (1):

[0054]

[0055] In the formula, D safe V represents the minimum safe distance between the vehicle and the alternative obstacle, and V represents the speed of the vehicle. obs The velocity of the candidate obstacle is represented by , and 'a' represents the maximum deceleration of the vehicle.

[0056] For each candidate obstacle, the minimum safe distance between the candidate obstacle and the vehicle is compared with the current actual distance between the candidate obstacle and the vehicle. If the actual distance is greater than or equal to the minimum safe distance, it means that the vehicle can currently avoid a collision with the candidate obstacle by braking, and there is no risk of collision between the candidate obstacle and the vehicle. However, if the actual distance is less than the minimum safe distance, it means that the vehicle cannot currently avoid a collision with the candidate obstacle by braking, and there is a risk of collision between the candidate obstacle and the vehicle.

[0057] Based on the above method, target obstacles that pose a collision risk to the vehicle can be accurately and quickly identified.

[0058] S12. Construct the driving space for the vehicle based on the obstacles around the vehicle and the roadside information of the road where the vehicle is located.

[0059] In this embodiment, the drivable space of the vehicle refers to the space located in the road space defined by the roadside curbs on both sides of the road, and which will not collide with obstacles.

[0060] In some embodiments of this application, the drivable space of the vehicle can be constructed through the following steps S121-S127:

[0061] S121. Discretize the road edges on both sides into point sets to obtain the left road edge point set and the right road edge point set.

[0062] In this embodiment, the road includes two road edges. One side is the road edge located on the left side of the vehicle in the direction of travel, referred to as the left road edge, and the other side is the road edge located on the right side of the vehicle in the direction of travel, referred to as the right road edge. In practical applications, the road edge information of both sides of the road can be obtained through the vehicle's perception module. The road edge information includes, but is not limited to, information such as the road edge position and size. Based on the obtained road edge information, the two road edges are discretized into point sets, thereby obtaining the left road edge point set corresponding to the left road edge and the right road edge point set corresponding to the right road edge.

[0063] S122. Determine the distances between the vehicle and the road edges on both sides to obtain the distances to the left and right road edges.

[0064] Here, the distance between the vehicle's centerline along the driving direction and the road edges on both sides can be determined by the vehicle's ranging module and sensing module, thus obtaining the distance between the vehicle and the left side road edge, and the distance between the vehicle and the right side road edge.

[0065] S123. Determine the target point set of the vehicle by means of linear interpolation, based on the set of points on the left side of the road, the set of points on the right side of the road, the distance on the left side of the road, and the distance on the right side of the road.

[0066] Here, after obtaining the set of points on the left side of the road, the set of points on the right side of the road, the distance between the left side of the road and the distance between the right side of the road, a set of points of the vehicle, i.e. the target point set, can be obtained by linear interpolation.

[0067] For example, see Figure 2 , Figure 2 The vehicles in the text are private cars, such as... Figure 2 As shown, a vehicle travels on a road, which includes a left curb located to the left of the vehicle and a right curb located to the right of the vehicle. Through discretization, the left curb is discretized into a set of left curb points, denoted as L(x,y), and the right curb is discretized into a set of right curb points, denoted as R(x,y). The distance between the vehicle's centerline along its travel direction and the left curb is taken as the left curb distance, denoted as D. l The distance between the centerline of the vehicle along the direction of travel and the right curb is taken as the right curb distance, and denoted as D. r Based on the set of points along the left and right sides of the road, the distance between the left and right sides of the road, and the distance between the left and right sides of the road, linear interpolation is performed according to the following formula (2) to obtain the target point set M(x,y):

[0068]

[0069] Among them, the central axis of the vehicle along the direction of travel is parallel to the road surface and passes through the center of mass of the vehicle. That is, the distance between the central axis and the road surface is the distance between the center of mass of the vehicle and the road surface.

[0070] S124. Use the straight line obtained by fitting the target point set as the reference line to construct a natural coordinate system with the reference line as the horizontal axis and the line perpendicular to the reference line and parallel to the road surface as the vertical axis.

[0071] Here, by fitting the points in the target point set, a fitting line can be obtained. This fitting line is parallel to the road surface, and the distance between the fitting line and the road surface is the distance between the vehicle's center of gravity and the road surface. Using this fitting line as a reference line, a natural coordinate system is constructed with the vehicle's center of gravity as the origin, the reference line as the horizontal axis, and a line perpendicular to the reference line and parallel to the road surface as the vertical axis.

[0072] For example, see Figure 2 , where s represents the positive direction of the horizontal axis of the natural coordinate system, which is the direction in which the vehicle moves along the reference line, and l represents the positive direction of the vertical axis of the natural coordinate system, which is the direction to the left perpendicular to the reference line.

[0073] S125. In the natural coordinate system, construct the complete driving space of the road based on the set of points on the left and right sides of the road.

[0074] Here, after obtaining the natural coordinate system, the two road edges can be projected onto the natural coordinate system based on the set of points on the left and right sides of the road edge. The space formed by the projected road edges in the natural coordinate system is taken as the complete driving space of the road. The complete driving space of the road is denoted as C(s,l), where s represents the distance the vehicle travels along the reference line, and l represents the distance of a point in the space relative to the reference line.

[0075] S126. Project the obstacles around the vehicle onto the natural coordinate system to obtain the projection space of the obstacles in the natural coordinate system.

[0076] The purpose of projecting obstacles onto the natural coordinate system is to determine the non-drivable space in the complete driving space of the road. Since the vehicle cannot drive on the obstacles, this embodiment takes the projected space of the obstacles in the natural coordinate system as the non-drivable space.

[0077] In some embodiments of this application, projecting obstacles around the vehicle onto a natural coordinate system to obtain the projection space of the obstacles in the natural coordinate system may include:

[0078] For static obstacles around the vehicle, based on the perception information of the static obstacles, the static obstacles are projected onto the natural coordinate system to obtain the projection space of the static obstacles in the natural coordinate system. The perception information of the static obstacles includes the position and size of the static obstacles.

[0079] For dynamic obstacles around the vehicle, calculate the speed difference between the dynamic obstacle and the vehicle; determine the minimum safe distance between the vehicle and the dynamic obstacle based on the speed difference and the vehicle's maximum deceleration; acquire the perception information of the dynamic obstacle, including its position and size; add the minimum safe distance to the length of the dynamic obstacle in the vehicle's driving direction to obtain the projected distance; based on the perception information and the projected distance, project the dynamic obstacle onto the natural coordinate system to obtain the projected space of the dynamic obstacle in the natural coordinate system.

[0080] Here, after obtaining the minimum safe distance between the dynamic obstacle and the vehicle, when projecting the dynamic obstacle onto the natural coordinate system, the length of the dynamic obstacle in the vehicle's driving direction is extended by this minimum safe distance in the direction closer to the vehicle. For example, if the vehicle is behind the dynamic obstacle, the length of the dynamic obstacle in the vehicle's driving direction is extended by the minimum safe distance behind the dynamic obstacle; if the vehicle is in front of the dynamic obstacle, the length of the dynamic obstacle in the vehicle's driving direction is extended by the minimum safe distance in front of the dynamic obstacle.

[0081] For example, see Figure 3 In this diagram, the red vehicle represents the driver, and the blue vehicle represents the target obstacle. When projecting the target obstacle, based on the minimum safe distance between the driver and the target obstacle, an additional projection is added behind the target obstacle. Figure 3 The safety compensation space M is shown.

[0082] By using the above method, the motion of dynamic obstacles is taken into account when projecting them. The projection space of dynamic obstacles is compensated based on the minimum safe distance, and the corresponding safety compensation space is increased, making the final determination of the non-drivable area more accurate.

[0083] In some embodiments of this application, when projecting a dynamic obstacle, determining the minimum safe distance between the vehicle and the dynamic obstacle based on the speed difference and the vehicle's maximum deceleration may include:

[0084] If the absolute value of the speed difference is greater than the speed threshold, perform the following steps S21 or S22:

[0085] S21. When the dynamic obstacle is in front of the vehicle and the speed difference is greater than 0, the minimum safe distance between the vehicle and the dynamic obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter.

[0086] S22. When the dynamic obstacle is located behind the vehicle and the speed difference is less than 0, the first safe distance is obtained by dividing the square of the vehicle's speed by the first deceleration parameter, and the second safe distance is obtained by dividing the square of the dynamic obstacle's speed by the second deceleration parameter. The sum of the first and second safe distances is taken as the minimum safe distance between the vehicle and the dynamic obstacle. The second deceleration parameter is twice the maximum deceleration of the dynamic obstacle.

[0087] Here, the speed threshold can be set according to the actual situation; for example, the speed threshold can be 60 km / h. If the absolute value of the speed difference between the vehicle and the dynamic obstacle is greater than the speed threshold, the dynamic obstacle is then projected onto the natural coordinate system through step S21 or S22.

[0088] For example, taking a speed threshold of 60 km / h as an example, when projecting a dynamic obstacle, the speed difference V between the vehicle and the dynamic obstacle can be calculated. d =VV obs At the absolute value of the velocity difference abs(V) d When the speed is greater than 60km / h, perform the following operations:

[0089] When the dynamic obstacle is in front of the vehicle, and V d When the distance is greater than 0, record the minimum safe distance between the vehicle and the dynamic obstacle.

[0090] When the dynamic obstacle is behind the vehicle, and V d When the distance is less than 0, record the minimum safe distance between the vehicle and the dynamic obstacle. In the formula, a obs This represents the maximum deceleration of a dynamic obstacle.

[0091] Furthermore, if the absolute value of the speed difference between the vehicle and the dynamic obstacle is less than or equal to the speed threshold, the dynamic obstacle can be left unprojected, and the collision between the vehicle and the dynamic obstacle can be avoided through subsequent speed planning.

[0092] S127. Subtract the projected space of obstacles from the complete driving space to obtain the driving space of the vehicle.

[0093] Let the complete driving space of the road be denoted as C(s,l), the projected space of static obstacles in the natural coordinate system be denoted as A1(s,l), and the projected space of dynamic obstacles in the natural coordinate system be denoted as A2(s,l). Then, the driving space H(s,l) of the vehicle can be determined by the following formula (3):

[0094] H(s,l)=C(s,l)-A1(s,l)-A2(s,l)

[0095] By determining the drivable space, some obstacles can be initially eliminated.

[0096] S13. Based on the relative positional relationship between the target obstacle and the vehicle, select the optimal drivable space from the drivable space.

[0097] In this embodiment, in order to obtain the optimal obstacle avoidance trajectory, after obtaining the vehicle's drivable space, the space that minimizes the vehicle's adjustment amount can be further selected from the drivable space based on the relative positional relationship between the target obstacle and the vehicle as the optimal drivable space.

[0098] In some embodiments of this application, before performing step S13 above, the relative positional relationship between the target obstacle and the vehicle can be determined first through the following steps S31-S35:

[0099] S31. Based on the actual distance between the vehicle and the target obstacle, the vehicle's minimum turning radius, and the relative speed between the vehicle and the target obstacle, determine the target time for the vehicle to avoid the target obstacle.

[0100] In this embodiment, the moment when the vehicle collides with the target obstacle is taken as the target moment when the vehicle avoids the target obstacle. The moment when the vehicle collides with the target obstacle is the moment after the collision time between the vehicle and the target obstacle has elapsed from the current moment. The collision time between the vehicle and the target obstacle refers to the time required from the current moment until the collision between the vehicle and the target obstacle occurs.

[0101] In some embodiments of this application, when determining the target time for the vehicle to avoid a target obstacle, the current time can be set to 0. The actual distance L between the vehicle and the target obstacle can be determined by the vehicle's ranging module or sensing module, and the speed V of the vehicle and the speed V of the target obstacle can be obtained by the sensing module. obs The system obtains the vehicle's minimum turning radius R from its own parameter information, then calculates the collision distance by subtracting 1.5 times the minimum turning radius R from the actual distance L using the collision time calculation formula. Finally, it divides the collision distance by the vehicle's speed V and the target obstacle's speed V. obs The speed difference V between themd The target time T for the vehicle to avoid the target obstacle is obtained. In practical applications, the target time T for the vehicle to avoid the target obstacle can be calculated using the following formula (4):

[0102]

[0103] S32. Based on the collision time and the lateral position and lateral velocity of the vehicle in the natural coordinate system, predict the first lateral position of the vehicle at the target time.

[0104] In this embodiment, to avoid a collision between the vehicle and the target obstacle, the vehicle needs to avoid the target obstacle at a target time. After obtaining the target time T for the vehicle to avoid the target obstacle through the above step S31, the first lateral position of the vehicle at the target time can be predicted according to the formula for uniform motion based on the current lateral position and lateral velocity of the vehicle in the natural coordinate system. Based on this, the first lateral position l of the vehicle at the target time can be predicted according to the following formula (5). ego :

[0105]

[0106] In the formula, l0 ego This indicates the current lateral position of the vehicle in the natural coordinate system. This represents the current lateral velocity of the vehicle in the natural coordinate system.

[0107] S33. Based on the lateral position and lateral velocity of the target obstacle in the natural coordinate system, predict the second lateral position of the target obstacle at the target time.

[0108] Similar to calculating the first lateral position of the vehicle at the target time, the second lateral position of the target obstacle at the target time can be predicted based on the current lateral position and lateral velocity of the target obstacle in the natural coordinate system, according to the formula for average velocity motion. Based on this, the second lateral position l of the target obstacle at the target time can be predicted according to the following formula (6). obs :

[0109]

[0110] In the formula, l0 obs This indicates the lateral position of the current target obstacle in the natural coordinate system. This represents the lateral velocity of the current target obstacle in the natural coordinate system.

[0111] S34. If the first lateral position is smaller than the second lateral position, determine that the target obstacle is located on the left side of the vehicle, where the left side of the vehicle refers to the left side of the vehicle in the direction of travel.

[0112] In this embodiment, if lego <l obs This indicates that the target obstacle is on the left side of the vehicle.

[0113] S35. If the first lateral position is greater than the second lateral position, determine that the target obstacle is located on the right side of the vehicle.

[0114] In this embodiment, if l ego >l obs This indicates that the target obstacle is on the right side of the vehicle, where "right side of the vehicle" refers to the right side of the vehicle in the direction of travel.

[0115] In this way, the relative positional relationship between the vehicle and the target obstacle can be determined.

[0116] Furthermore, after determining the relative positional relationship between the target obstacle and the vehicle, the optimal drivable space can be selected from the drivable space based on the relative positional relationship between the target obstacle and the vehicle through the following steps S41-S43:

[0117] S41. Use the reference line as the dividing line of the drivable space. In the direction of travel of the vehicle, the space to the left of the dividing line is the left drivable space, and the space to the right of the dividing line is the right drivable space.

[0118] In this embodiment, the reference line is a line composed of the set of target points corresponding to the vehicle. Therefore, when dividing the drivable space using the reference line as the boundary line, the space to the left of the boundary line is the space to the left of the vehicle, and the space to the right of the boundary line is the space to the right of the vehicle. The drivable space on the left is denoted as HL(s,l), and the drivable space on the right is denoted as HR(s,l).

[0119] S42. When the target obstacle is located on the left side of the vehicle and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the right-side drivable space is taken as the optimal drivable space.

[0120] In this embodiment, when the vehicle is located to the right of the target obstacle, if the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, it indicates that the vehicle is seriously deviated to the right relative to the target obstacle. In this case, in order to reduce the adjustment amount of the vehicle, the left drivable space HL(s,l) can be discarded and the right drivable space HR(s,l) can be taken as the optimal drivable space.

[0121] S43. When the target obstacle is located on the right side of the vehicle and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the drivable space on the left side is taken as the optimal drivable space.

[0122] In this embodiment, when the vehicle is located to the left of the target obstacle, if the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, it indicates that the vehicle is seriously deviated to the left relative to the target obstacle. In this case, in order to reduce the adjustment amount of the vehicle, the right drivable space HR(s,l) can be discarded and the right drivable space HL(s,l) can be taken as the optimal drivable space.

[0123] Furthermore, in some embodiments of this application, if the optimal drivable space cannot be determined based on the absolute value of the difference between the first lateral position and the second lateral position, that is, if it cannot be determined which portion of the drivable space to discard based on the absolute value of the difference between the first lateral position and the second lateral position, the following steps may also be performed:

[0124] Determine the first position in the left-hand drive space that is narrower than the width of the vehicle, and use the distance between this position and the position of the vehicle as the first distance;

[0125] Determine the first position in front of the vehicle whose width is less than the width of the vehicle within the right-hand drive space, and use the distance between this position and the position of the vehicle as the second distance;

[0126] If the first distance is greater than the second distance, the drivable space on the left is taken as the optimal drivable space;

[0127] If the first distance is less than or equal to the second distance, the drivable space on the right is taken as the optimal drivable space.

[0128] Here, when determining the first distance, the longitudinal distances at each lateral position in the left-hand drive space can be iterated to determine the lateral positions where the longitudinal distance is less than the vehicle's width. The lateral position with a longitudinal distance less than the vehicle's width and closest to the vehicle is taken as the first position in the left-hand drive space in front of the vehicle whose width is less than the vehicle's width. The distance between this position and the vehicle is then taken as the first distance. Similarly, when determining the second distance, the longitudinal distances at each lateral position in the right-hand drive space can be iterated to determine the lateral positions where the longitudinal distance is less than the vehicle's width. The lateral position with a longitudinal distance less than the vehicle's width and closest to the vehicle is taken as the first position in the right-hand drive space in front of the vehicle whose width is less than the vehicle's width. The distance between this position and the vehicle is then taken as the second distance. Let the first distance be denoted as SL. max Let the second distance be SR max When it is impossible to abandon the drivable space on one side by passing through S42 and S43, if SL max >SR max If SL, then the left-hand driving space is selected as the optimal driving space. max <SRmax If the available space on the right is selected as the optimal driving space, then the determined optimal driving space is more sufficient, making collision avoidance maneuvers easier.

[0129] S14. Perform path planning within the optimal drivable space to obtain the optimal collision avoidance path.

[0130] In this embodiment, after determining the optimal drivable space, the optimal collision avoidance path that can avoid collision with the target obstacle can be obtained by performing path planning within the optimal drivable space.

[0131] In some embodiments of this application, each path point p is defined. i Let U be a set of path points (s, l), where s represents the distance the vehicle travels along the reference line and l represents the distance of a path point relative to the reference line; then the complete collision avoidance path can be represented as a set of path points, denoted as U{p0, p1, ..., p...}. l}, where the s values ​​between adjacent path points differ by a specified interval. Both the specified interval and k can be set according to actual conditions. Taking a specified interval of 0.2 and k = 500 as an example, then when s = 0 for the first path point, p0 = (0, l0), p1 = (0.2, l0), ... p k =(0.2*500, l 500 Therefore, solving for the optimal obstacle avoidance path can be transformed into solving for a set of optimal l sequences. Based on this, the optimal collision avoidance path can be obtained by performing path planning within the optimal drivable space through the following steps S141-S143:

[0132] S141. Construct an objective function with the goal of ensuring that the vehicle can maintain a safe distance from obstacles and minimizing the adjustment of the vehicle's operating parameters.

[0133] In this embodiment, an objective function can be constructed based on the goal of maintaining a safe distance between the vehicle and obstacles and minimizing the adjustment of the vehicle's operating parameters. These operating parameters include, but are not limited to, speed, steering, and acceleration. The objective function can be a polynomial containing multidimensional indicators, which can be related to the distance between the vehicle and obstacles, as well as the adjustment of the vehicle's operating parameters.

[0134] In some embodiments of this application, the objective function can be used to indicate the minimum cost function value of the obstacle avoidance path;

[0135] The cost function can be represented by the following equation (7):

[0136]

[0137] In the formula, J represents the cost function value of the obstacle avoidance path, w1, w2, w3, and w4 are four weighting coefficients, l represents the distance of a path point in the obstacle avoidance path relative to the reference line, and biasobs represents the distance of the boundaries of each obstacle around the vehicle relative to the reference line. The values ​​of the four weighting coefficients, w1, w2, w3, and w4, can be determined through data analysis and expert experience. For example, through data analysis, w1 = 10, w2 = 25, w3 = 80, and w4 = 10 can be set.

[0138] S142. Construct the constraint conditions for the objective function based on the kinematic constraints of the vehicle in the natural coordinate system and the constraint that the path points cannot exceed the optimal drivable space.

[0139] In some embodiments of this application, constraints may include:

[0140] l∈H(s,l)

[0141]

[0142]

[0143] Where H(s,l) represents the optimal drivable space, Δs represents the difference in s between two adjacent path points, and l i+1 and l i The l value represents the value of two adjacent path points.

[0144] S143. Solve the objective function based on the constraints to obtain the optimal collision avoidance path.

[0145] By describing the problem as a standard quadratic programming problem, the optimal set of l is obtained, which is the optimal set of path points U, or the optimal collision avoidance path under the current environment.

[0146] S15. Add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

[0147] In this embodiment, after obtaining the optimal collision avoidance path, the optimal collision avoidance trajectory can be obtained by adding the corresponding speed information to the path points in the optimal collision avoidance path.

[0148] The obstacle avoidance trajectory generation method provided in this embodiment constructs an optimal drivable space based on road edge information and obstacle information in the road, transforms environmental information into constraint information usable by the planning module, and performs collision avoidance trajectory planning in the constructed optimal drivable space. The obstacle information in the surrounding environment is considered in the planning stage. Compared with only considering the end state, it can better handle the relative relationship between the vehicle and obstacles in the environment, thereby achieving the effect of generating emergency collision avoidance trajectories in more complex scenarios.

[0149] In some embodiments of this application, in step S15 above, speed information can be added to the path points in the optimal collision avoidance path by performing speed planning on the vehicle. Based on this, speed information can be added to the path points in the optimal collision avoidance path through the following steps S151-S154 to obtain the optimal collision avoidance trajectory:

[0150] S151. Perform deceleration sampling according to the preset deceleration interval to obtain a deceleration set composed of multiple deceleration sample values.

[0151] In this embodiment, a deceleration sampling start point can be set, and then multiple deceleration sampling values ​​can be obtained by sampling at a preset deceleration interval as the sampling step size according to the deceleration sampling start point. The set of multiple deceleration sampling values ​​is used as the deceleration set.

[0152] For example, 0 can be used as the starting point for deceleration sampling, with a speed of 0.5 m / s². 2 By sampling the deceleration at intervals, we obtain the deceleration set Acc{0,0.5,…,5.0}.

[0153] S152. Based on the vehicle's current speed and the planned trajectory time length, generate the speed curve corresponding to each deceleration sample value in the deceleration set.

[0154] In this embodiment, the speed curve refers to the curve of the vehicle speed changing with time. For each deceleration sample value in the deceleration set, a speed curve of the vehicle with the deceleration sample value can be generated according to the speed equation, and the speed curve is used as the speed curve corresponding to the deceleration sample value.

[0155] The velocity equation is shown in equation (8) below:

[0156] V(t) = V0 - at

[0157] In the formula, V0 represents the current speed of the vehicle, which is also the initial speed of the vehicle, a represents the deceleration, a∈Acc, and t represents the time length of the planned trajectory.

[0158] S153. Based on the velocity curve corresponding to each deceleration sample value in the deceleration set, add velocity information to the path points in the optimal collision avoidance path to obtain multiple collision avoidance trajectories.

[0159] In this embodiment, for each deceleration sample value corresponding to a velocity curve, velocity information is attached to each path point obtained by path planning based on the velocity curve, so as to obtain the collision avoidance trajectory corresponding to each deceleration sample value, that is, to obtain multiple collision avoidance trajectories.

[0160] S154. Among multiple collision avoidance trajectories, the trajectory with the smallest corresponding deceleration sample value and no collision is taken as the optimal collision avoidance trajectory.

[0161] In this embodiment, the collision-free trajectory with the smallest 'a' is selected as the final optimal collision avoidance trajectory. The smaller 'a' is, the smaller the vehicle's adjustment amount, and the safer it is.

[0162] The collision avoidance trajectory generation method provided in this application can handle more complex obstacle scenarios, especially continuous collision avoidance scenarios. For example, the collision avoidance trajectory generation method provided in this embodiment can handle scenarios such as... Figures 4-6 The scene shown depicts a car in red, with other colored vehicles representing obstacles. Figure 4 In the scenario shown, the vehicle cannot avoid a collision with the vehicle in front through braking. In this case, the solution provided in this application embodiment can be used to generate an optimal collision avoidance trajectory. The vehicle then executes the generated optimal collision avoidance trajectory, thus avoiding a collision. Figure 5 In the scenario shown, the vehicle cannot avoid a collision with the vehicle in front through braking, and the environment ahead is complex with many obstacles and vehicles. Compared to existing polynomial sampling schemes, the collision avoidance trajectory planned based on the embodiments of this application can better handle the relative relationship between the vehicle and obstacles in the environment, and can ensure a high success rate in more complex scenarios. Figure 6 In the scenario shown, the vehicle cannot avoid a collision with the vehicle in front by braking, and there is still a nearby obstacle in the avoidance direction. In this scenario, the solution provided by the embodiments of this application can plan a continuous emergency collision avoidance trajectory.

[0163] Based on the collision avoidance trajectory generation method provided in the above embodiments, this application also provides specific implementation methods of the collision avoidance trajectory generation device. Please refer to the following embodiments.

[0164] See Figure 7 The collision avoidance trajectory generation device 700 provided in this application embodiment may include the following modules:

[0165] Risk assessment module 701 is used to identify target obstacles that pose a collision risk to the vehicle from among the obstacles around the vehicle;

[0166] The space construction module 702 is used to construct the driving space of the vehicle based on the obstacles around the vehicle and the curb information of the road where the vehicle is located.

[0167] The space filtering module 703 is used to select the optimal drivable space from the drivable space based on the relative positional relationship between the target obstacle and the vehicle.

[0168] The path planning module 704 is used to perform path planning within the optimal drivable space to obtain the optimal collision avoidance path.

[0169] The trajectory generation module 705 is used to add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

[0170] The obstacle avoidance trajectory generation device of this application constructs an optimal drivable space based on the roadside information and obstacle information in the road, transforms the environmental information into constraint information usable by the planning module, and performs collision avoidance trajectory planning in the constructed optimal drivable space. The obstacle information in the surrounding environment is considered in the planning stage. Compared with only considering the end state, it can better handle the relative relationship between the vehicle and obstacles in the environment, thereby achieving the effect of generating emergency collision avoidance trajectories in more complex scenarios.

[0171] In some embodiments, the risk assessment module 701 includes:

[0172] The filtering unit is used to select N obstacles that are located in front of or to the sides of the vehicle and closest to the vehicle in the direction of the vehicle's travel, based on the roadside information of the road where the vehicle is located, as candidate obstacles.

[0173] The assessment unit is used to conduct a collision risk assessment for each candidate obstacle to determine whether there is a collision risk between each candidate obstacle and the vehicle.

[0174] The determination unit is used to identify alternative obstacles that pose a collision risk to the vehicle as target obstacles.

[0175] In some embodiments, the evaluation unit is used for:

[0176] For each candidate obstacle, perform the following steps:

[0177] Determine the speed difference between the vehicle and the candidate obstacles;

[0178] The minimum safe distance between the vehicle and the candidate obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter, where the first deceleration parameter is twice the maximum deceleration of the vehicle.

[0179] Compare the actual distance between the vehicle and the candidate obstacle with the minimum safe distance;

[0180] If the actual distance is determined to be less than the minimum safe distance, it is determined that there is a risk of collision between the alternative obstacle and the vehicle.

[0181] If the actual distance is greater than or equal to the minimum safe distance, it is determined that there is no risk of collision between the alternative obstacle and the vehicle.

[0182] In some embodiments, the space construction module 702 is used for:

[0183] Discretize the road edges on both sides into point sets to obtain the left road edge point set and the right road edge point set;

[0184] Determine the distances between the vehicle and the roadside curbs on both sides to obtain the distances to the left and right roadside curbs;

[0185] The target point set of the vehicle is determined by linear interpolation based on the set of points on the left side of the road, the set of points on the right side of the road, the distance on the left side of the road, and the distance on the right side of the road.

[0186] The line obtained by fitting the target point set is used as the reference line to construct a natural coordinate system with the vehicle's center of mass as the origin, the reference line as the horizontal axis, and a line perpendicular to the reference line and parallel to the road surface as the vertical axis.

[0187] In the natural coordinate system, the complete driving space of the road is constructed based on the set of points on the left and right sides of the road.

[0188] By projecting the obstacles around the vehicle onto the natural coordinate system, the projection space of the obstacles in the natural coordinate system is obtained.

[0189] Subtract the projected space from the total driving space to obtain the vehicle's usable driving space.

[0190] In some embodiments, the space construction module 702 is used for:

[0191] For static obstacles around the vehicle, based on the perception information of the static obstacles, the static obstacles are projected onto the natural coordinate system to obtain the projection space of the static obstacles in the natural coordinate system. The perception information of the static obstacles includes the position and size of the static obstacles.

[0192] In some embodiments, the space construction module 702 is used for:

[0193] For dynamic obstacles around the vehicle, calculate the speed difference between the dynamic obstacles and the vehicle;

[0194] Determine the minimum safe distance between the vehicle and the dynamic obstacle based on the speed difference and the vehicle's maximum deceleration;

[0195] Acquire perception information of dynamic obstacles, including the position and size of the dynamic obstacles;

[0196] The projected distance is obtained by adding the minimum safe distance to the length of the dynamic obstacle in the direction of the vehicle's travel towards the vehicle.

[0197] Based on the perceived information and projection distance, dynamic obstacles are projected onto the natural coordinate system to obtain the projection space of dynamic obstacles in the natural coordinate system.

[0198] In some embodiments, the space construction module 702 is used for:

[0199] If the absolute value of the speed difference is greater than the speed threshold, perform the following steps:

[0200] When the dynamic obstacle is in front of the vehicle and the speed difference is greater than 0, the minimum safe distance between the vehicle and the dynamic obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter, where the first deceleration parameter is twice the maximum deceleration of the vehicle.

[0201] When the dynamic obstacle is located behind the vehicle and the speed difference is less than 0, the first safe distance is obtained by dividing the square of the vehicle's speed by the first deceleration parameter, and the second safe distance is obtained by dividing the square of the dynamic obstacle's speed by the second deceleration parameter. The sum of the first and second safe distances is taken as the minimum safe distance between the vehicle and the dynamic obstacle. The second deceleration parameter is twice the maximum deceleration of the dynamic obstacle.

[0202] In some embodiments, the apparatus 700 may further include: a relative position determination module, configured to:

[0203] Based on the actual distance between the vehicle and the target obstacle, the vehicle's minimum turning radius, and the relative speed between the vehicle and the target obstacle, the target time for the vehicle to avoid the target obstacle is determined.

[0204] Based on the vehicle's lateral position and lateral velocity in the natural coordinate system, predict the vehicle's first lateral position at the target time;

[0205] Based on the lateral position and lateral velocity of the target obstacle in the natural coordinate system, predict the second lateral position of the target obstacle at the target time;

[0206] If the first lateral position is smaller than the second lateral position, the target obstacle is determined to be located on the left side of the vehicle;

[0207] If the first lateral position is greater than the second lateral position, the target obstacle is determined to be located on the right side of the vehicle.

[0208] In some embodiments, the spatial filtering module 703 is used for:

[0209] The reference line is used as the dividing line of the drivable space. The space to the left of the dividing line in the direction of the vehicle's travel is the left drivable space, and the space to the right of the dividing line is the right drivable space.

[0210] When the target obstacle is located on the left side of the vehicle, and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the right-side driving space is taken as the optimal driving space.

[0211] When the target obstacle is located on the right side of the vehicle, and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the drivable space on the left side is taken as the optimal drivable space.

[0212] In some embodiments, the spatial filtering module 703 is further configured to:

[0213] If the optimal drivable space cannot be selected from the drivable space based on the relative positional relationship between the target obstacle and the vehicle, the first position in front of the vehicle with a width smaller than the width of the vehicle is determined in the drivable space on the left, and the distance between this position and the position of the vehicle is taken as the first distance.

[0214] Determine the first position in front of the vehicle whose width is less than the width of the vehicle within the right-hand drive space, and use the distance between this position and the position of the vehicle as the second distance;

[0215] If the first distance is greater than the second distance, the drivable space on the left is taken as the optimal drivable space;

[0216] If the first distance is less than or equal to the second distance, the drivable space on the right is taken as the optimal drivable space.

[0217] In some embodiments, the route planning module 704 is configured to:

[0218] The objective function is constructed with the goal of ensuring that the vehicle maintains a safe distance from obstacles and minimizing the adjustment of the vehicle's operating parameters.

[0219] The constraints of the objective function are constructed based on the kinematic constraints of the vehicle in the natural coordinate system and the constraint that the path points cannot exceed the optimal drivable space.

[0220] The objective function is solved based on the constraints to obtain the optimal collision avoidance path.

[0221] In some embodiments, the objective function is used to indicate the minimum cost function value of the obstacle avoidance path;

[0222] The cost function includes:

[0223]

[0224] In the formula, J represents the cost function value of the obstacle avoidance path, w1, w2, w3, and w4 are four weight coefficients, l represents the distance of the path point in the obstacle avoidance path relative to the reference line, and biasobs represents the distance of the boundary of each obstacle around the vehicle relative to the reference line.

[0225] In some embodiments, the trajectory generation module 705 is configured to:

[0226] Deceleration is sampled at preset deceleration intervals to obtain a deceleration set consisting of multiple deceleration sample values;

[0227] Based on the vehicle's current speed and the planned trajectory time, generate a speed curve corresponding to each deceleration sample value in the deceleration set;

[0228] Based on the velocity curve corresponding to each deceleration sample value in the deceleration set, velocity information is added to the path points in the optimal collision avoidance path to obtain multiple collision avoidance trajectories.

[0229] The trajectory with the smallest corresponding deceleration sample value and no collision among multiple collision avoidance trajectories is selected as the optimal collision avoidance trajectory.

[0230] The collision avoidance trajectory generation device provided in this application embodiment can achieve Figures 1 to 6 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0231] Figure 8 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0232] Electronic devices may include a processor 801 and a memory 802 storing computer program instructions.

[0233] Specifically, the processor 801 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0234] Memory 802 may include mass storage for data or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 802 may include removable or non-removable (or fixed) media. Where appropriate, memory 802 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 802 is non-volatile solid-state memory.

[0235] Memory 802 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory 802 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it can perform the operations described in any of the collision avoidance trajectory generation methods in the above embodiments.

[0236] The processor 801 reads and executes computer program instructions stored in the memory 802 to implement any of the collision avoidance trajectory generation methods in the above embodiments.

[0237] In one example, the electronic device may also include a communication interface 803 and a bus 810. For example, Figure 8 As shown, the processor 801, memory 802, and communication interface 803 are connected through bus 810 and complete communication with each other.

[0238] The communication interface 803 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0239] Bus 810 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 810 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0240] Furthermore, in conjunction with the collision avoidance trajectory generation method in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the collision avoidance trajectory generation methods in the above embodiments.

[0241] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0242] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0243] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0244] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0245] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for generating collision avoidance trajectories, characterized in that, include: Identify target obstacles that pose a collision risk to the vehicle from among the obstacles surrounding the vehicle; The driving space of the vehicle is constructed based on the obstacles around the vehicle and the road edge information of the road where the vehicle is located; Based on the relative positional relationship between the target obstacle and the vehicle, the optimal drivable space is selected from the drivable space; Path planning is performed within the optimal drivable space to obtain the optimal collision avoidance path; Add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

2. The method according to claim 1, characterized in that, The step of identifying target obstacles from among the obstacles surrounding the vehicle that pose a collision risk to the vehicle includes: Based on the roadside information of the road where the vehicle is located, select N obstacles from the obstacles located within the roadside, which are located in front of or to the sides of the vehicle in the direction of travel of the vehicle and are closest to the vehicle, as candidate obstacles. Each candidate obstacle is assessed for collision risk to determine whether there is a collision risk between each candidate obstacle and the vehicle. The alternative obstacles that pose a collision risk with the vehicle are selected as the target obstacles.

3. The method according to claim 2, characterized in that, Each candidate obstacle is assessed for collision risk to determine whether there is a collision risk between the vehicle and each candidate obstacle, including: For each candidate obstacle, perform the following steps: Determine the speed difference between the vehicle and the candidate obstacle; The minimum safe distance between the vehicle and the candidate obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter, wherein the first deceleration parameter is twice the maximum deceleration of the vehicle. Compare the actual distance between the vehicle and the candidate obstacle with the minimum safe distance; If the actual distance is determined to be less than the minimum safe distance, it is determined that there is a risk of collision between the alternative obstacle and the vehicle. If the actual distance is determined to be greater than or equal to the minimum safe distance, it is determined that there is no risk of collision between the alternative obstacle and the vehicle.

4. The method according to claim 1, characterized in that, The step of constructing the drivable space of the vehicle based on the obstacles around the vehicle and the curb information of the road where the vehicle is located includes: Discretize the road edges on both sides into point sets to obtain the left road edge point set and the right road edge point set; The distances between the vehicle and the two road edges are determined respectively to obtain the distances to the left and right road edges; The target point set of the vehicle is determined by linear interpolation based on the set of left roadside points, the set of right roadside points, the distance of the left roadside, and the distance of the right roadside. Using the line obtained by fitting the target point set as the reference line, a natural coordinate system is constructed with the centroid of the vehicle as the origin, the reference line as the horizontal axis, and a line perpendicular to the reference line and parallel to the road surface as the vertical axis. In the natural coordinate system, the complete driving space of the road is constructed based on the set of left curb points and the set of right curb points; The obstacles around the vehicle are projected onto the natural coordinate system to obtain the projection space of the obstacles in the natural coordinate system. The drivable space of the vehicle is obtained by subtracting the projected space from the complete driving space.

5. The method according to claim 4, characterized in that, The step of projecting obstacles around the vehicle onto the natural coordinate system to obtain the projection space of the obstacles in the natural coordinate system includes: For static obstacles around the vehicle, based on the perception information of the static obstacles, the static obstacles are projected onto the natural coordinate system to obtain the projection space of the static obstacles in the natural coordinate system. The perception information of the static obstacles includes the position and size of the static obstacles.

6. The method according to claim 4 or 5, characterized in that, The step of projecting obstacles around the vehicle onto the natural coordinate system to obtain the projection space of the obstacles in the natural coordinate system includes: For dynamic obstacles around the vehicle, calculate the speed difference between the dynamic obstacle and the vehicle; Based on the speed difference and the maximum deceleration of the vehicle, determine the minimum safe distance between the vehicle and the dynamic obstacle; Acquire the perception information of the dynamic obstacle, wherein the perception information of the dynamic obstacle includes the position and size of the dynamic obstacle; The length of the dynamic obstacle in the direction of travel of the vehicle is increased by the minimum safe distance in the direction closer to the vehicle to obtain the projected distance; Based on the perceived information and the projection distance, the dynamic obstacle is projected onto the natural coordinate system to obtain the projection space of the dynamic obstacle in the natural coordinate system.

7. The method according to claim 6, characterized in that, Determining the minimum safe distance between the vehicle and the dynamic obstacle based on the speed difference and the vehicle's maximum deceleration includes: If the absolute value of the speed difference is greater than the speed threshold, the following steps are performed: When the dynamic obstacle is in front of the vehicle and the speed difference is greater than 0, the minimum safe distance between the vehicle and the dynamic obstacle is obtained by dividing the square of the speed difference by the first deceleration parameter, wherein the first deceleration parameter is twice the maximum deceleration of the vehicle. When the dynamic obstacle is located behind the vehicle and the speed difference is less than 0, a first safe distance is obtained by dividing the square of the vehicle's speed by the first deceleration parameter, and a second safe distance is obtained by dividing the square of the dynamic obstacle's speed by the second deceleration parameter. The sum of the first safe distance and the second safe distance is taken as the minimum safe distance between the vehicle and the dynamic obstacle, wherein the second deceleration parameter is twice the maximum deceleration of the dynamic obstacle.

8. The method according to claim 4, characterized in that, Before selecting the optimal drivable space from the drivable space based on the relative positional relationship between the target obstacle and the vehicle, the method further includes: Based on the actual distance between the vehicle and the target obstacle, the minimum turning radius of the vehicle, and the relative speed between the vehicle and the target obstacle, the target time for the vehicle to avoid the target obstacle is determined; Based on the lateral position and lateral velocity of the vehicle in the natural coordinate system, predict the first lateral position of the vehicle at the target time; Based on the lateral position and lateral velocity of the target obstacle in the natural coordinate system, predict the second lateral position of the target obstacle at the target time; If the first lateral position is smaller than the second lateral position, the target obstacle is determined to be located on the left side of the vehicle; If the first lateral position is greater than the second lateral position, the target obstacle is determined to be located on the right side of the vehicle.

9. The method according to claim 8, characterized in that, The step of selecting the optimal drivable space from the drivable space based on the relative positional relationship between the target obstacle and the vehicle includes: The reference line is used as the dividing line of the drivable space. The space on the left side of the dividing line in the direction of travel of the vehicle is called the left drivable space, and the space on the right side of the dividing line is called the right drivable space. When the target obstacle is located on the left side of the vehicle, and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the right-side drivable space is taken as the optimal drivable space. When the target obstacle is located on the right side of the vehicle, and the absolute value of the difference between the first lateral position and the second lateral position is greater than half the width of the vehicle, the drivable space on the left side is taken as the optimal drivable space.

10. The method according to claim 9, characterized in that, When the optimal drivable space cannot be selected from the drivable space based on the relative positional relationship between the target obstacle and the vehicle, the method further includes: Determine the first position in the left-hand drivable space that is narrower than the width of the vehicle in front of it, and use the distance between this position and the position of the vehicle as the first distance; In the right-side drivable space, determine the first position in front of the vehicle whose width is less than the width of the vehicle, and use the distance between this position and the position of the vehicle as the second distance; If the first distance is greater than the second distance, the drivable space on the left is taken as the optimal drivable space; If the first distance is less than or equal to the second distance, the right-side drivable space is taken as the optimal drivable space.

11. The method according to claim 4, characterized in that, The step of performing path planning within the optimal drivable space to obtain the optimal collision avoidance path includes: An objective function is constructed with the goal of ensuring that the vehicle maintains a safe distance from obstacles and minimizing the adjustment of the vehicle's operating parameters. The constraints of the objective function are constructed based on the kinematic constraints of the vehicle in the natural coordinate system and the constraint that the path points cannot exceed the optimal drivable space. The objective function is solved based on the constraints to obtain the optimal collision avoidance path.

12. The method according to claim 11, characterized in that, The objective function is used to indicate the minimum cost function value of the obstacle avoidance path; The cost function includes: In the formula, J represents the cost function value of the obstacle avoidance path, w1, w2, w3, and w4 are four weighting coefficients, l represents the distance of the path point in the obstacle avoidance path relative to the reference line, and biasobs represents the distance of the boundary of each obstacle around the vehicle relative to the reference line.

13. The method according to claim 1, characterized in that, The step of adding velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory includes: Deceleration is sampled at preset deceleration intervals to obtain a deceleration set consisting of multiple deceleration sample values; Based on the vehicle's current speed and the planned trajectory time length, a speed curve is generated corresponding to each deceleration sample value in the deceleration set; Based on the velocity curve corresponding to each deceleration sample value in the deceleration set, velocity information is added to the path points in the optimal collision avoidance path to obtain multiple collision avoidance trajectories. The trajectory with the smallest corresponding deceleration sample value and no collision among the multiple collision avoidance trajectories is taken as the optimal collision avoidance trajectory.

14. A collision avoidance trajectory generation device, characterized in that, include: The risk assessment module is used to identify target obstacles that pose a collision risk to the vehicle from among the obstacles around the vehicle. The space construction module is used to construct the drivable space of the vehicle based on the obstacles around the vehicle and the road edge information of the road where the vehicle is located; The space filtering module is used to select the optimal drivable space from the drivable space based on the relative positional relationship between the target obstacle and the vehicle. The path planning module is used to plan a path within the optimal drivable space to obtain the optimal collision avoidance path. The trajectory generation module is used to add velocity information to the path points in the optimal collision avoidance path to obtain the optimal collision avoidance trajectory.

15. An electronic device, characterized in that, The electronic device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the steps of the collision avoidance trajectory generation method as described in any one of claims 1-13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the steps of the collision avoidance trajectory generation method as described in any one of claims 1-13.

17. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device is able to perform the collision avoidance trajectory generation method as described in any one of claims 1-13.

18. A vehicle, characterized in that, Including the electronic device as described in claim 15.

Images