Vehicle obstacle avoidance method, device, vehicle, equipment and storage medium

Abstract

The embodiment of the present application provides a vehicle obstacle avoidance method, device, vehicle, equipment and storage medium, the method comprising: obtaining the associated information of each obstacle, the associated information comprising: the position information, speed, acceleration, angular velocity and three-axis attitude of the obstacle; predicting the trajectory of each obstacle in the future preset time period according to the associated information; predicting collision information according to the current vehicle's own trajectory in the future preset time period and the predicted trajectory of each obstacle, so as to perform the following vehicle or parking avoidance operation. The method provided by the embodiment of the present application can overcome the problem that the prior art cannot accurately achieve vehicle obstacle avoidance.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of autonomous driving technology, and in particular to a vehicle obstacle avoidance method, apparatus, vehicle, equipment and storage medium. Background Art

[0002] Autonomous driving vehicles are equipped with sensors and other sensing devices to identify information about stationary obstacles such as potholes, rocks and other debris directly in front of the vehicle, so as to implement automatic emergency braking, automatic deceleration or lane changing to avoid them.

[0003] However, for closed-scene autonomous driving test sites, when multiple vehicles work together, it is difficult to accurately detect the surrounding environment due to the field of view limitations and cost reasons of single-vehicle equipment (such as a single vehicle), and there may be a risk of collision with other vehicles; for example, when other vehicles change lanes or change their driving status, or when other vehicles follow the original path without making any changes, it is possible for vehicles to collide.

[0004] Therefore, the existing technology cannot accurately achieve vehicle obstacle avoidance. Summary of the invention

[0005] The embodiments of the present application provide a vehicle obstacle avoidance method, apparatus, vehicle, equipment and storage medium to overcome the problem that the prior art cannot accurately achieve vehicle obstacle avoidance.

[0006] In a first aspect, an embodiment of the present application provides a vehicle obstacle avoidance method, comprising:

[0007] Obtaining relevant information of each obstacle, the relevant information including: position information, velocity, acceleration, angular velocity and three-axis attitude of the obstacle;

[0008] Predicting the trajectory of each obstacle at a preset time in the future based on the associated information;

[0009] Based on the current vehicle's own trajectory for a preset time in the future and the predicted trajectory of each obstacle, collision information is predicted to perform following or parking avoidance operations.

[0010] In a possible design, predicting the trajectory of each obstacle at a preset time in the future according to the associated information includes:

[0011] For each obstacle, within the preset time length and according to the preset time interval, the following steps are repeatedly performed to obtain the predicted trajectory of the obstacle:

[0012] According to the initial state of the obstacle, predict the state of the obstacle after the Nth preset time interval, wherein the initial state includes the position information, speed, acceleration, angular velocity and three-axis attitude at the current moment;

[0013] The state of the obstacle after the Nth preset time interval is taken as the starting state, and the state of the obstacle after the N+1th preset time interval is predicted.

[0014] In a possible design, the association information further includes a size of a labeling box, and the size of the labeling box is determined by a target detection algorithm; and the method further includes:

[0015] When performing a following or parking avoidance operation according to the collision information, if it is detected that the obstacle as a leading vehicle has stopped for a time exceeding a preset time threshold in the driving direction of the current vehicle, a lane change and detour operation is triggered;

[0016] In response to a triggering operation of lane change and detour, a detour trajectory is determined according to a high-precision map, a size of a marked box of the preceding vehicle, current position information of the preceding vehicle, a size of a marked box of the current vehicle, and current position information of the current vehicle;

[0017] If the detour trajectory meets the preset detour conditions, the vehicle is driven according to the detour trajectory, and the operation of predicting and obtaining collision information is performed during the detour to avoid collision risks.

[0018] In a possible design, determining the detour trajectory according to the high-precision map, the size of the marked box of the preceding vehicle, the current position information of the preceding vehicle, the size of the marked box of the current vehicle, and the current position information of the current vehicle includes:

[0019] Determine the maximum width of the lane change to bypass the front vehicle according to the high-precision map and the size of the marked box of the front vehicle;

[0020] If the maximum width satisfies a predefined condition, a plurality of track points are determined according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle;

[0021] According to multiple trajectory points, the detour trajectory and the curvature of any point of the detour trajectory are determined by a spline interpolation method, and the curvature is used to support the judgment of whether the detour condition is met.

[0022] In a possible design, the determining of multiple track points according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle includes:

[0023] Determine a first track point, a second track point, a third track point, and a fourth track point according to the high-precision map, a size of the marking box of the preceding vehicle, and current position information of the preceding vehicle;

[0024] Determine a fifth track point, a sixth track point, and a seventh track point according to the high-precision map, a size of a marking box of the current vehicle, current position information of the current vehicle, and current position information of the preceding vehicle;

[0025] The first and second track points are respectively located on the front path of the preceding vehicle where the center point of the marking frame of the preceding vehicle is located, the fourth track point is located on the parallel path to the left of the center point of the marking frame of the preceding vehicle, the third track point is located on the front path where the fourth track point is located, the seventh track point is located at the center point of the marking frame of the current vehicle, the sixth track point is located at the intersection of the front path of the current vehicle where the seventh track point is located and the marking frame of the current vehicle, and the fifth track point is located on the rear path where the fourth track point is located and is located in front of the left of the sixth track point;

[0026] The distance between the first trajectory point and the seventh trajectory point is calibrated according to the current vehicle performance test, the distance between the third trajectory point and the parallel line where the upper border of the marking box of the preceding vehicle is located is determined by the body length of the current vehicle and the transformation performance of the current vehicle, and the distance between the third trajectory point and the parallel line where the upper border of the marking box of the preceding vehicle is located is determined by a high-precision map.

[0027] In a possible design, the prediction of collision information based on the current vehicle's own trajectory at a preset time in the future and the predicted trajectory of each obstacle, so as to perform a following vehicle or parking avoidance operation, includes:

[0028] For each obstacle, by comparing the own trajectory with the predicted trajectory, obtaining collision information between the current vehicle and the obstacle, the collision position and the collision time of the collision information;

[0029] According to the collision positions with each obstacle, determining the target collision position where the current vehicle has the shortest travel distance on the own trajectory and the target collision time and target obstacle corresponding to the target collision position;

[0030] The following and stopping relationship between the current vehicle and the target obstacle within the target collision time is determined according to the target collision position and the target collision time. The following and stopping relationship includes the distance between the current vehicle and the target obstacle, the speed of the current vehicle, the following time for the target obstacle in the same driving direction, and the stopping time.

[0031] In a possible design, the obtaining of association information of each obstacle includes:

[0032] Acquire sensing information through sensing devices arranged on the roadside, wherein the sensing information includes relevant data of each sensed target detected, wherein the relevant data includes the size of the annotation box, position information, speed, acceleration, angular velocity and three-axis attitude of the sensed target;

[0033] According to the relevant data of each of the perceived targets and the self-positioning information of the current vehicle, the relevant data belonging to the current vehicle among the perceived targets are filtered to obtain the association information of the remaining perceived targets as obstacles.

[0034] In a second aspect, an embodiment of the present application provides a vehicle obstacle avoidance device, comprising:

[0035] An acquisition module, used to acquire the associated information of each obstacle, the associated information including: the position information, velocity, acceleration, angular velocity and three-axis posture of the obstacle;

[0036] A prediction module, used to predict the trajectory of each obstacle within a preset time period in the future based on the associated information;

[0037] The processing module is used to predict collision information based on the current vehicle's own trajectory at a preset time in the future and the predicted trajectory of each obstacle, so as to perform following or parking avoidance operations.

[0038] In a third aspect, an embodiment of the present application provides a vehicle, which is used to execute the vehicle obstacle avoidance method as described in any one of the first aspects above.

[0039] In a fourth aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory;

[0040] The memory stores computer-executable instructions;

[0041] The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method as described in any one of the first aspects above.

[0042] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the method described in any one of the first aspects above is implemented.

[0043] The vehicle obstacle avoidance method, device, vehicle, equipment and storage medium provided in this embodiment first obtains the associated information of each obstacle, and the associated information includes: the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle; then, based on the associated information, predict the trajectory of each obstacle in the future preset time; and then predict the collision information based on the current vehicle's own trajectory in the future preset time and the predicted trajectory of each obstacle, so as to perform the following vehicle or parking avoidance operation. Therefore, based on the acquired information such as the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle, the trajectory of the obstacle is predicted, which can ensure the accuracy of the predicted trajectory, and based on the comparison between the predicted trajectory and the current vehicle's own trajectory, the collision information is obtained, so that the current vehicle can follow the vehicle in front or stop to avoid danger, and the collision problem can be accurately predicted by using the trajectory prediction and the own trajectory, so as to maintain the distance and speed with the vehicle in front, so as to avoid vehicle collision. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 A schematic diagram of a scenario of a vehicle obstacle avoidance method provided in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of a process flow of a vehicle obstacle avoidance method provided in an embodiment of the present application;

[0047] Figure 3 A schematic diagram of a scenario of a vehicle obstacle avoidance method provided in yet another embodiment of the present application;

[0048] Figure 4 A schematic diagram of a scenario of a vehicle obstacle avoidance method provided in yet another embodiment of the present application;

[0049] Figure 5 A schematic diagram of the structure of a vehicle obstacle avoidance device provided in an embodiment of the present application;

[0050] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0053] At present, for closed-scene autonomous driving test sites, when multiple vehicles work together, it is difficult to accurately detect the surrounding environment due to the field of view limitations and cost of a single vehicle device (such as a single vehicle), and there may be a risk of collision with other vehicles; for example, when other vehicles change lanes or change their driving status, or when other vehicles follow the original path without making any changes, it is possible for the vehicle to collide. Therefore, the existing technology cannot accurately achieve vehicle obstacle avoidance.

[0054] Therefore, in response to the above problems, the technical concept of the present application is to predict the trajectory of the obstacle based on the acquired information such as the position, speed, acceleration, angular velocity and three-axis posture of the obstacle, so as to ensure the accuracy of the predicted trajectory, and to obtain collision information based on the comparison of the predicted trajectory with the current vehicle's own trajectory, so that the current vehicle can follow the vehicle in front or stop to avoid danger. The use of trajectory prediction and the vehicle's own trajectory can accurately predict collision problems, and then maintain the distance and speed with the vehicle in front to avoid vehicle collisions.

[0055] In practical applications, see Figure 1 As shown, Figure 1A schematic diagram of a scenario of a vehicle obstacle avoidance method provided in an embodiment of the present application. The vehicle obstacle avoidance method can be applied to vehicles with an autonomous driving function. In a closed scenario (such as an autonomous driving test scenario or a road scenario with a simple road condition, etc.), in order to improve the perception accuracy of a single vehicle device (such as a vehicle) to the overall scenario, the perception device can be arranged on a road test. When multiple vehicles work together, the perception data can be obtained through the perception device, and the trajectory of each obstacle in the future preset time period (such as the trajectory of the next 5 seconds) can be predicted based on the perception data. Then, the predicted trajectory is compared with the current vehicle's own trajectory to find the collision point with a collision risk in the future preset time period, and then the driving state of the current vehicle in the future preset time period is determined based on the data corresponding to the collision point (i.e., collision information), and then the vehicle in front is followed and stopped and possible collisions are avoided.

[0056] If the vehicle in front of the current vehicle stops for a long time (such as when the vehicle is performing a task), the lane change detour planning is triggered. The specific detour route is obtained based on the high-precision map and the bounding_box (i.e. the size of the annotation box) of the vehicle in front. During the detour, the previous collision detection mechanism is still used to ensure the safety of the vehicle's operation.

[0057] Therefore, after the sensing equipment is arranged on the roadside, the present application improves the vehicle's perception accuracy of the overall scene, and predicts the trajectory based on the acquired obstacle's position information, speed, acceleration, angular velocity and three-axis posture, and then determines the collision information, thereby achieving following the vehicle in front and stopping to avoid possible collisions; by combining with high-precision maps, the function of bypassing and avoiding obstacles is obtained, which improves the safety factor and efficiency of bypassing.

[0058] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0059] Figure 2 A schematic diagram of a process flow of a vehicle obstacle avoidance method provided in an embodiment of the present application, the method may include:

[0060] S201: Obtain relevant information of each obstacle.

[0061] The associated information includes: the location information, speed, acceleration, angular velocity and three-axis posture of the obstacle.

[0062] In this embodiment, the obstacle here may refer to other vehicles different from the current vehicle in the scenario of multiple vehicles working together, and may also refer to static objects, such as potholes, rocks, and other debris. The following takes the obstacle as another vehicle as an example to describe the vehicle obstacle avoidance method in detail.

[0063] In a possible design, obtaining the associated information of each obstacle can be achieved by the following steps:

[0064] Step a1: Acquire sensing information through sensing devices arranged on the roadside, where the sensing information includes relevant data of each sensed target detected, including the size of the annotation box, position information, speed, acceleration, angular velocity and three-axis attitude of the sensed target;

[0065] Step a2: According to the relevant data of each perceived target and the current vehicle's own positioning information, the relevant data of the perceived targets belonging to the current vehicle are filtered to obtain the associated information of the remaining perceived targets as obstacles.

[0066] In this embodiment, for closed scenarios, such as autonomous driving test scenarios, when multiple vehicles work together, due to the limitations of vehicle vision and cost reasons, the single-vehicle equipment can be connected to the perception device through 5G, and then multi-vehicle collaboration can be achieved based on high-precision maps.

[0067] In order to improve the perception accuracy of the overall scene of a single vehicle, the perception device can be arranged on the roadside (i.e., the roadside perception device). The perception device can obtain relevant data of detectable objects, such as the object's annotation box size, position information, speed, acceleration, angular velocity and three-axis posture.

[0068] Specifically, the perception information accessed through 5G includes the length, width, height (here used to indicate the size of the annotation box) and location information of all detectable objects in the field, and combined with the current vehicle positioning and body information, collision detection is performed using the time series prediction method. The current vehicle can use the perception information and its own positioning to filter out the perception information of the vehicle itself, and treat all remaining perception targets after filtering the vehicle as obstacles, and obtain the perception information of the obstacle, where the perception information of the obstacle can refer to the associated information of the obstacle.

[0069] S202: predicting the trajectory of each obstacle within a preset time period in the future based on the associated information.

[0070] In this embodiment, according to the acquired information such as the position, speed, acceleration, angular velocity and three-axis posture of the obstacle, the trajectory of the obstacle is predicted through the time series prediction algorithm, which can ensure the accuracy of the predicted trajectory. Here, the time series prediction algorithm uses different time intervals to predict the vehicle state, and obtains the predicted trajectory of the obstacle through continuous iteration.

[0071] S203: predict collision information based on the current vehicle's own trajectory within a preset time period in the future and the predicted trajectories of various obstacles, so as to perform a following vehicle or parking avoidance operation.

[0072] In this embodiment, the current vehicle can obtain its own trajectory for a preset time in the future according to the driving path and current driving state, and then compare its own trajectory with the predicted trajectory of each obstacle to find the existing intersection as the collision point. Then, based on the information corresponding to the collision point (i.e., collision information), the driving state of the current vehicle for a preset time in the future is determined, and then the vehicle in front is followed and stopped, and a possible collision is avoided by stopping.

[0073] The vehicle obstacle avoidance method provided in this embodiment first obtains the associated information of each obstacle, and the associated information includes: the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle; then, based on the associated information, predict the trajectory of each obstacle in the future preset time; and then predict the collision information based on the current vehicle's own trajectory in the future preset time and the predicted trajectory of each obstacle, so as to perform the following vehicle or parking avoidance operation. Therefore, based on the acquired information such as the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle, the trajectory of the obstacle is predicted, which can ensure the accuracy of the predicted trajectory, and based on the comparison between the predicted trajectory and the current vehicle's own trajectory, the collision information is obtained, so that the current vehicle can follow the vehicle in front or stop to avoid danger. The trajectory prediction and the own trajectory can accurately predict the collision problem, and then the distance and speed can be maintained with the vehicle in front to avoid vehicle collision.

[0074] In a possible design, this embodiment provides a detailed description of S202 based on the above embodiment. That is, based on the associated information, the trajectory of each obstacle in the future for a preset time period can be predicted, which can be achieved by the following steps:

[0075] Step b1: for each obstacle, repeatedly perform the following steps within a preset time length and according to a preset time interval to obtain a predicted trajectory of the obstacle:

[0076] Step b2: predict the state of the obstacle after the Nth preset time interval according to the initial state of the obstacle, where the initial state includes the position information, speed, acceleration, angular velocity and three-axis attitude at the current moment;

[0077] Step b3: taking the state of the obstacle after the Nth preset time interval as the starting state, predicting the state of the obstacle after the N+1th preset time interval.

[0078] In this embodiment, the preset time interval here can be 100 ms. At 100 ms intervals, according to the current vehicle speed, angular velocity, acceleration and posture, the state of the vehicle 100 ms later can be calculated. The state after 100 ms is taken as the starting state, and it is iterated sequentially to estimate the running trajectory in the next few seconds, that is, the predicted trajectory.

[0079] In a possible design, this embodiment provides a detailed description of S203 based on the above embodiment. That is, based on the current vehicle's own trajectory for a preset time in the future and the predicted trajectory of each obstacle, collision information is predicted to perform a following vehicle or parking avoidance operation, which can be achieved through the following steps:

[0080] Step c1: for each obstacle, by comparing the own trajectory with the predicted trajectory, obtain the collision information between the current vehicle and the obstacle, the collision position and the collision time;

[0081] Step c2, according to the collision positions with each obstacle, determining the target collision position with the shortest travel distance of the current vehicle on its own trajectory, as well as the target collision time and target obstacle corresponding to the target collision position;

[0082] Step c3, according to the target collision position and the target collision time, determine the following and stopping relationship between the current vehicle and the target obstacle within the target collision time, the following and stopping relationship includes the distance between the current vehicle and the target obstacle, the speed of the current vehicle, the following time for the target obstacle in the same driving direction, and the stopping time.

[0083] For example, see Figure 3 As shown, Figure 3 A schematic diagram of a scenario of a vehicle obstacle avoidance method provided for another embodiment of the present application. This scenario includes three vehicles working together, such as vehicle 1, vehicle 2, and vehicle 3, wherein vehicle 1 is traveling straight from west to east, and vehicle 2 and vehicle 3 are traveling from south to north on different lanes. With vehicle 3 as the current vehicle and vehicle 1 and vehicle 2 as obstacles, vehicle 3 uses the speed, acceleration, angular velocity, three-axis posture and other information of each obstacle (such as vehicle 1 and vehicle 2) to predict the trajectory of the obstacle in the next 5 seconds, and collision information, such as the collision position and time, can be obtained by comparing its own trajectory with the predicted trajectory. According to Figure 3 The collision position of A and B in .

[0084] By comparing the collision positions of A and B, vehicle 3 and vehicle 2 may collide at point B. Taking point B as an example, the distance between vehicle 3 and vehicle 2, the speed of vehicle 3, the following time and the parking time for the obstacle vehicle 2 in the same driving direction within a preset time period in the future can be determined, so as to follow and stop the vehicle in front (such as vehicle 2) and stop to avoid possible collision.

[0085] In one possible design, see Figure 4 As shown, Figure 4 A schematic diagram of a scene of a vehicle obstacle avoidance method provided by another embodiment of the present application. This embodiment describes the vehicle obstacle avoidance method in detail based on the above embodiment. The associated information also includes the size of the annotation box (such as Figure 4 The size of the annotation box is determined by the target detection algorithm. The vehicle obstacle avoidance method can also be implemented by the following steps:

[0086] Step d1: When performing a following or parking avoidance operation based on collision information, if the obstacle is detected as the parking time of the leading vehicle in the driving direction of the current vehicle and exceeds a preset time threshold, a lane change and detour operation is triggered.

[0087] Step d2, in response to the triggering operation of lane change and detour, determine the detour trajectory according to the high-precision map, the marking box size of the front vehicle, the current position information of the front vehicle, the marking box size of the current vehicle and the current position information of the current vehicle.

[0088] Step d3: If the detour trajectory meets the preset detour conditions, the vehicle drives according to the detour trajectory and performs operations to predict and obtain collision information during detour to avoid collision risks.

[0089] In this embodiment, when the front vehicle stops for a long time (vehicles perform tasks, etc.), the lane change detour planning is triggered. The specific detour route is obtained according to the high-precision map and the bounding_box of the front vehicle, that is, according to the high-precision map, the size of the annotation box of the front vehicle, the current position information of the front vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle. Since the environment of the closed scene is relatively simple, multiple waypoints can be selected in the free space for curve fitting to form a detour trajectory. The curvature of any point on the trajectory is recorded, and the actual response speed of the current vehicle actuator is used to determine whether the average curvature meets the detour conditions. If it meets the conditions, the detour is performed according to the detour trajectory, and the previous collision detection mechanism is still used during the detour to ensure the safety of vehicle operation.

[0090] In a possible design, the detour trajectory is determined according to the high-precision map, the size of the annotation box of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle, which can be achieved by the following steps:

[0091] Step d21, determining the maximum width of the lane change to bypass the front vehicle based on the high-precision map and the size of the marked box of the front vehicle;

[0092] Step d22: if the maximum width satisfies the predefined condition, multiple track points are determined according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle;

[0093] Step d23: Determine the detour trajectory and the curvature of any point of the detour trajectory based on multiple trajectory points by spline interpolation method, and the curvature is used to support the judgment of whether the detour conditions are met.

[0094] In this embodiment, combined with Figure 4 As shown, according to the size and position information of the annotation boxes of the current vehicle and the preceding vehicle, three points (1, 4, 7) of the detour trajectory, the starting point, the end point, and the detour trajectory point on the left side of the obstacle are selected. In order to make the vehicle turn more smoothly and prevent the influence of large-angle steering on the vehicle posture return control, four more points (2, 3, 5, 6) are selected.

[0095] Specifically, the implementation process of the model of the detour planning algorithm can be: first determine the distances e and f (the distance e is the vertical distance between the center point of the marked box of vehicle A as the front vehicle and the left edge line of the lane change lane, and the distance f is the vertical distance between the left box of the marked box of vehicle A and the left edge line of the lane change lane), determine the limit width of the variable lane on the left and right sides of the variable lane, if the limit width meets a defined value, that is, a predefined condition, such as greater than or equal to M meters, then it is considered that the lane change behavior can be carried out; by adjusting the x, y, z distances and using the spline interpolation method, a smooth detour trajectory is obtained.

[0096] In a possible design, according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle, the following steps can be performed:

[0097] Step d221, determining a first track point, a second track point, a third track point, and a fourth track point according to the high-precision map, the size of the marking box of the preceding vehicle, and the current position information of the preceding vehicle;

[0098] Step d222, determine the fifth track point, the sixth track point and the seventh track point according to the high-precision map, the size of the annotation box of the current vehicle, the current position information of the current vehicle and the current position information of the preceding vehicle.

[0099] Among them, the first trajectory point and the second trajectory point are respectively located on the front path of the front vehicle where the center point of the marking box of the front vehicle is located, the fourth trajectory point is located on the parallel path to the left of the center point of the marking box of the front vehicle, the third trajectory point is located on the front path where the fourth trajectory point is located, the seventh trajectory point is located at the center point of the marking box of the current vehicle, the sixth trajectory point is located at the intersection of the front path of the current vehicle where the seventh trajectory point is located and the marking box of the current vehicle, and the fifth trajectory point is located on the rear path where the fourth trajectory point is located and is located in front of the left of the sixth trajectory point.

[0100] The distance between the first track point and the seventh track point is calibrated according to the current vehicle performance test. The distance between the third track point and the parallel line of the upper border of the marking box of the preceding vehicle is determined by the body length of the current vehicle and the transformation performance of the current vehicle. The distance between the third track point and the parallel line of the upper border of the marking box of the preceding vehicle is determined by a high-precision map.

[0101] In this embodiment, see Figure 4 As shown in the figure, the selection method of multiple trajectory points is:

[0102] Points 2 and 6 are on the path of the original trajectory, and the selection of points 2 and 6 is to make the vehicle's current posture cut in closer to the lane change trajectory, so that the convergence effect of the control algorithm is better. After actual measurement, it can optimize the convergence speed of the control function. The distance between points 1 and 7 needs to be calibrated according to the vehicle performance test. The line connecting points 3, 4, and 5 is parallel to the left boundary of bounding_box_A (i.e., the left side frame of the annotation box of vehicle A); among them, x (i.e., the distance between point 3 and the parallel line of the upper side frame of the annotation box of vehicle A) can be slightly larger than half the length of the body of the current vehicle (such as vehicle B), and can be corrected according to the steering performance of the current vehicle.

[0103] Points 3 and 5 are selected to prevent the vehicle from making large turning changes in a short period of time. The vehicle's chassis control-by-wire response speed error may cause the vehicle to tilt and cause a collision risk. Points 3 and 5 also ensure that the vehicle is far enough away from obstacles to prevent collisions. The specific values ​​are determined based on the vehicle's own performance.

[0104] Therefore, the present application can obtain comprehensive and high-precision obstacle perception information (such as annotation box size, position information, speed, acceleration, angular velocity, and three-axis attitude, etc.) through the method of 5G access perception equipment. On this basis, combined with high-precision maps, the detour route can be determined by the maximum width of the variable lane range on the left and right sides of the front obstacle and the range of the obstacle bounding_box. The arrangement of road test equipment reduces the manufacturing cost of the vehicle and improves the vehicle's global perception efficiency. The perception data is used to predict the obstacle trajectory, and combined with high-precision maps, the model of the entire detour planning algorithm is simplified, and the safety factor and efficiency of the detour are improved. It solves the problem in the prior art that, under inaccurate perception and scene information, lane changes will cause risks, and only basic operations such as following the vehicle can be achieved, and the current vehicle may stop for various reasons, and the subsequent work vehicles may queue up continuously, so that the overall coordination efficiency is greatly reduced.

[0105] In order to implement the vehicle obstacle avoidance method, this embodiment provides a vehicle obstacle avoidance device. Figure 5 , Figure 5 A schematic diagram of the structure of a vehicle obstacle avoidance device provided in an embodiment of the present application; the vehicle obstacle avoidance device 50 comprises: an acquisition module 501, a prediction module 502 and a processing module 503; the acquisition module 501 is used to acquire the associated information of each obstacle, and the associated information includes: the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle; the prediction module 502 is used to predict the trajectory of each obstacle at a preset time in the future based on the associated information; the processing module 503 is used to predict the collision information based on the current vehicle's own trajectory at a preset time in the future and the predicted trajectory of each obstacle, so as to perform following or parking avoidance operations.

[0106] In this embodiment, an acquisition module 501, a prediction module 502 and a processing module 503 are provided to acquire the associated information of each obstacle, and the associated information includes: the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle; then, based on the associated information, the trajectory of each obstacle in the future preset time is predicted; and then, based on the current vehicle's own trajectory in the future preset time and the predicted trajectory of each obstacle, the collision information is predicted to perform the following vehicle or parking avoidance operation. Therefore, based on the acquired information such as the position information, speed, acceleration, angular velocity and three-axis posture of the obstacle, the trajectory of the obstacle is predicted, which can ensure the accuracy of the predicted trajectory, and based on the comparison between the predicted trajectory and the current vehicle's own trajectory, the collision information is obtained, so that the current vehicle can follow the vehicle in front or stop to avoid danger, and the collision problem can be accurately predicted using the trajectory prediction and the own trajectory, so as to maintain the distance and speed with the vehicle in front, so as to avoid vehicle collision.

[0107] The device provided in this embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.

[0108] In a possible design, the prediction module is specifically used to:

[0109] For each obstacle, repeat the following steps within a preset time period and at a preset time interval to obtain the predicted trajectory of the obstacle:

[0110] According to the initial state of the obstacle, the state of the obstacle after the Nth preset time interval is predicted, and the initial state includes the current position information, speed, acceleration, angular velocity and three-axis attitude;

[0111] The state of the obstacle after the Nth preset time interval is taken as the starting state, and the state of the obstacle after the N+1th preset time interval is predicted.

[0112] In a possible design, the associated information further includes a size of the annotation box, and the size of the annotation box is determined by a target detection algorithm; the vehicle obstacle avoidance device may further include: a path planning module; the path planning module includes a first planning unit, a second planning unit, and a third planning unit;

[0113] A first planning unit is configured to trigger a lane change and detour operation when performing a following or parking avoidance operation according to collision information, if the obstacle is detected as a stop time of the leading vehicle in the driving direction of the current vehicle for more than a preset time threshold;

[0114] A second planning unit is used to respond to the triggering operation of lane change and detour, and determine the detour trajectory according to the high-precision map, the size of the marked box of the preceding vehicle, the current position information of the preceding vehicle, the size of the marked box of the current vehicle, and the current position information of the current vehicle;

[0115] The third planning unit is used to drive according to the detour trajectory when the detour trajectory meets the preset detour conditions, and perform operations to predict and obtain collision information during detour to avoid collision risks.

[0116] In a possible design, the second planning unit is specifically used to:

[0117] Determine the maximum width of the lane change to bypass the vehicle in front based on the high-precision map and the size of the annotation box of the vehicle in front;

[0118] If the maximum width meets the predefined conditions, multiple track points are determined according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle;

[0119] Based on multiple trajectory points, the detour trajectory and the curvature of any point of the detour trajectory are determined through the spline interpolation method. The curvature is used to support the judgment of whether the detour conditions are met.

[0120] In a possible design, the second planning unit is specifically used to:

[0121] Determine a first track point, a second track point, a third track point, and a fourth track point according to the high-precision map, the size of the marking box of the preceding vehicle, and the current position information of the preceding vehicle;

[0122] Determine a fifth track point, a sixth track point, and a seventh track point according to the high-precision map, a size of a labeling box of the current vehicle, current position information of the current vehicle, and current position information of a preceding vehicle;

[0123] Among them, the first track point and the second track point are respectively located on the front path of the front vehicle where the center point of the annotation box of the front vehicle is located, the fourth track point is located on the parallel path to the left of the center point of the annotation box of the front vehicle, the third track point is located on the front path where the fourth track point is located, the seventh track point is located at the center point of the annotation box of the current vehicle, the sixth track point is located at the intersection of the front path of the current vehicle where the seventh track point is located and the annotation box of the current vehicle, and the fifth track point is located on the rear path where the fourth track point is located and is located in front of the left of the sixth track point;

[0124] The distance between the first track point and the seventh track point is calibrated according to the current vehicle performance test. The distance between the third track point and the parallel line of the upper border of the marking box of the preceding vehicle is determined by the body length of the current vehicle and the transformation performance of the current vehicle. The distance between the third track point and the parallel line of the upper border of the marking box of the preceding vehicle is determined by a high-precision map.

[0125] In a possible design, the processing module is specifically configured to:

[0126] For each obstacle, by comparing the vehicle’s own trajectory with the predicted trajectory, the collision information, collision location and collision time between the vehicle and the obstacle are obtained;

[0127] According to the collision positions with each obstacle, determine the target collision position with the shortest travel distance of the current vehicle on its own trajectory, as well as the target collision time and target obstacle corresponding to the target collision position;

[0128] According to the target collision position and the target collision time, the following and stopping relationship between the current vehicle and the target obstacle within the target collision time is determined. The following and stopping relationship includes the distance between the current vehicle and the target obstacle, the speed of the current vehicle, the following time for the target obstacle in the same driving direction, and the stopping time.

[0129] In a possible design, the acquisition module is specifically used to:

[0130] Acquire sensing information through sensing devices arranged on the roadside, the sensing information includes relevant data of each sensed target detected, including the size of the annotation box, position information, speed, acceleration, angular velocity and three-axis attitude of the sensed target;

[0131] According to the relevant data of each perceived target and the current vehicle's own positioning information, the relevant data belonging to the current vehicle in the perceived targets are filtered to obtain the associated information of the remaining perceived targets as obstacles.

[0132] In order to implement the method of the above embodiment, this embodiment provides a vehicle, which is used to execute the various steps executed in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0133] In order to implement the method of the above embodiment, this embodiment provides an electronic device. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 6 As shown, the electronic device 60 of this embodiment includes: a processor 601 and a memory 602; wherein the memory 602 is used to store computer-executable instructions; the processor 601 is used to execute the computer-executable instructions stored in the memory to implement the various steps performed in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0134] An embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0135] An embodiment of the present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0136] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms. In addition, each functional module in each embodiment of the present application can be integrated into a processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0137] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (English: processor) to perform some steps of the methods of each embodiment of the present application. It should be understood that the above-mentioned processor can be a central processing unit (English: Central Processing Unit, referred to as: CPU), or other general-purpose processors, digital signal processors (English: Digital Signal Processor, referred to as: DSP), application-specific integrated circuits (English: Application Specific Integrated Circuit, referred to as: ASIC), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0138] The memory may include high-speed RAM memory, and may also include non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a disk or an optical disk, etc. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus. The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a disk or an optical disk. The storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0139] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the storage medium can also exist as discrete components in an electronic device or a main control device.

[0140] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle obstacle avoidance method, It is characterized in that include: Obtaining the associated information of each obstacle, the associated information including: the position information, velocity, acceleration, angular velocity, three-axis posture and marking box size of the obstacle; Predicting the trajectory of each obstacle at a preset time in the future based on the associated information; According to the current vehicle's own trajectory for a preset time in the future and the predicted trajectory of each obstacle, collision information is predicted to perform a following vehicle or parking avoidance operation; When performing a following or parking avoidance operation according to the collision information, if it is detected that the obstacle as a leading vehicle has stopped for a time exceeding a preset time threshold in the driving direction of the current vehicle, a lane change and bypass operation is triggered; In response to a triggering operation of lane change and detour, determining a maximum width of lane change to detour the preceding vehicle according to a high-precision map and a size of a marked box of the preceding vehicle; If the maximum width meets the predefined conditions, multiple track points are determined according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle; the multiple track points include: the starting point and the end point of the detour track, the detour track point on the left side of the obstacle, and four track points that are increased to reduce the influence on the vehicle posture return control; According to a plurality of trajectory points, the detour trajectory and the curvature of any point of the detour trajectory are determined by a spline interpolation method, and the curvature is used to support the determination of whether the detour condition is met.

2. The method according to claim 1, It is characterized in that The predicting, based on the associated information, the trajectory of each obstacle within a preset time period in the future includes: For each obstacle, within the preset time length and according to the preset time interval, the following steps are repeatedly performed to obtain the predicted trajectory of the obstacle: According to the initial state of the obstacle, predict the state of the obstacle after the Nth preset time interval, wherein the initial state includes the position information, speed, acceleration, angular velocity and three-axis attitude at the current moment; The state of the obstacle after the Nth preset time interval is taken as the starting state, and the state of the obstacle after the N+1th preset time interval is predicted.

3. The method according to claim 1 or 2, It is characterized in that The size of the annotation box is determined by a target detection algorithm; the method further comprises: If the detour trajectory meets the preset detour conditions, the vehicle is driven according to the detour trajectory, and the operation of predicting and obtaining collision information is performed during the detour to avoid collision risks.

4. The method according to claim 1, It is characterized in that The method of determining a plurality of track points according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle includes: Determine a first track point, a second track point, a third track point, and a fourth track point according to the high-precision map, a size of the marking box of the preceding vehicle, and current position information of the preceding vehicle; Determine a fifth track point, a sixth track point, and a seventh track point according to the high-precision map, a size of a marking box of the current vehicle, current position information of the current vehicle, and current position information of the preceding vehicle; The first and second track points are respectively located on the front path of the preceding vehicle where the center point of the marking frame of the preceding vehicle is located, the fourth track point is located on the parallel path to the left of the center point of the marking frame of the preceding vehicle, the third track point is located on the front path where the fourth track point is located, the seventh track point is located at the center point of the marking frame of the current vehicle, the sixth track point is located at the intersection of the front path of the current vehicle where the seventh track point is located and the marking frame of the current vehicle, and the fifth track point is located on the rear path where the fourth track point is located and is located in front of the left of the sixth track point; The distance between the first trajectory point and the seventh trajectory point is calibrated according to the current vehicle performance test, and the distance between the third trajectory point and the parallel line of the upper border of the marking box of the front vehicle is determined by the body length of the current vehicle and the transformation performance of the current vehicle.

5. The method according to claim 1 or 2, It is characterized in that The method predicts collision information based on the current vehicle's own trajectory for a preset time period in the future and the predicted trajectory of each obstacle, so as to perform a following vehicle or parking avoidance operation, including: For each obstacle, by comparing the own trajectory with the predicted trajectory, obtaining collision information between the current vehicle and the obstacle, the collision position and the collision time of the collision information; According to the collision positions with each obstacle, determining the target collision position where the current vehicle has the shortest travel distance on the own trajectory and the target collision time and target obstacle corresponding to the target collision position; The following and stopping relationship between the current vehicle and the target obstacle within the target collision time is determined according to the target collision position and the target collision time. The following and stopping relationship includes the distance between the current vehicle and the target obstacle, the speed of the current vehicle, the following time for the target obstacle in the same driving direction, and the stopping time.

6. The method according to claim 1 or 2, It is characterized in that The obtaining of the associated information of each obstacle includes: Acquire sensing information through sensing devices arranged on the roadside, wherein the sensing information includes relevant data of each sensed target detected, wherein the relevant data includes the size of the annotation box, position information, speed, acceleration, angular velocity and three-axis attitude of the sensed target; According to the relevant data of each of the perceived targets and the self-positioning information of the current vehicle, the relevant data belonging to the current vehicle among the perceived targets are filtered to obtain the association information of the remaining perceived targets as obstacles.

7. A vehicle obstacle avoidance device, It is characterized in that include: An acquisition module is used to acquire the associated information of each obstacle, wherein the associated information includes: the position information, velocity, acceleration, angular velocity, three-axis posture and marking box size of the obstacle; A prediction module, used to predict the trajectory of each obstacle within a preset time period in the future based on the associated information; A processing module, used to predict collision information based on the current vehicle's own trajectory for a preset time in the future and the predicted trajectory of each obstacle, so as to perform a following vehicle or parking avoidance operation; The vehicle obstacle avoidance device further includes: a path planning module; the path planning module includes a first planning unit and a second planning unit; a first planning unit, configured to trigger a lane change and detour operation when performing a following or parking avoidance operation according to the collision information, if it is detected that the obstacle as a leading vehicle has stopped for a time exceeding a preset time threshold in the driving direction of the current vehicle; The second planning unit is specifically used for: In response to a triggering operation of lane change and detour, determining a maximum width of lane change to detour the preceding vehicle according to a high-precision map and a size of a marked box of the preceding vehicle; If the maximum width meets the predefined conditions, multiple track points are determined according to the high-precision map, the size of the annotation box in the associated information of the preceding vehicle, the current position information of the preceding vehicle, the size of the annotation box of the current vehicle, and the current position information of the current vehicle; the multiple track points include: the starting point and the end point of the detour track, the detour track point on the left side of the obstacle, and four track points that are increased to reduce the influence on the vehicle posture return control; According to multiple trajectory points, the detour trajectory and the curvature of any point of the detour trajectory are determined by a spline interpolation method, and the curvature is used to support the judgment of whether the detour condition is met.

8. A vehicle, It is characterized in that The vehicle is used to execute the vehicle obstacle avoidance method as described in any one of claims 1-6.

9. An electronic device, It is characterized in that include: Processor and memory; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the vehicle obstacle avoidance method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer-executable instructions, and when the processor executes the computer-executable instructions, the vehicle obstacle avoidance method according to any one of claims 1 to 6 is implemented.

Images