Track generation method and device, vehicle and storage medium

Abstract

The invention discloses a track generation method and device, a vehicle and a readable storage medium. The method comprises the steps that a target sensing lane where a target obstacle is located is determined from at least one sensing lane sensed by a vehicle; the relative angle difference between the course angle of the target obstacle and the lane center line direction of the target sensing lane is determined; determining a target obstacle intention corresponding to a target angle interval where the relative angle difference is located; each angle interval corresponds to respective obstacle intention; and generating a motion track of the target obstacle according to the target obstacle intention. According to the method provided by the invention, the trajectory prediction of the target obstacle is realized, and the accuracy of the predicted motion trajectory of the target obstacle is relatively high.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and more specifically, to a trajectory generation method, device, vehicle, and computer-readable storage medium. Background Art

[0002] Facing a complex and dynamic traffic environment, accurate and reliable estimation of obstacle trajectories is an important basis for autonomous vehicles to make decisions and plans. Therefore, a method is urgently needed to accurately predict the trajectory of obstacles. Summary of the invention

[0003] The present application proposes a trajectory generation method, device, vehicle and computer-readable storage medium to provide a means for predicting the trajectory of an obstacle.

[0004] In a first aspect, an embodiment of the present application provides a trajectory generation method, the method comprising:

[0005] Determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the ego-vehicle;

[0006] Determine the relative angle difference between the heading angle of the target obstacle and the lane centerline direction of the target sensing lane;

[0007] Determine the target obstacle intention corresponding to the target angle interval where the relative angle difference is located; each angle interval corresponds to its own obstacle intention;

[0008] According to the intention of the target obstacle, the motion trajectory of the target obstacle is generated.

[0009] In a second aspect, an embodiment of the present application further provides a trajectory generation device, the device comprising:

[0010] A lane determination module, used to determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the vehicle;

[0011] An angle determination module, used to determine the relative angle difference between the heading angle of the target obstacle and the lane centerline direction of the target sensing lane;

[0012] An intention determination module is used to determine the intention of a target obstacle corresponding to a target angle interval where the relative angle difference is located; each angle interval corresponds to its own obstacle intention;

[0013] The trajectory generation module is used to generate the motion trajectory of the target obstacle according to the intention of the target obstacle.

[0014] In a third aspect, an embodiment of the present application further provides a vehicle, characterized in that the vehicle comprises: one or more processors; a memory; one or more applications, wherein one or more applications are stored in the memory and configured to be executed by one or more processors, and one or more programs are configured to execute the above method.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a program code executable by a processor, and when the program code is executed by the processor, the processor executes the above method.

[0016] The present application provides a trajectory generation method, device, vehicle and computer-readable storage medium. In the present application, a target perception lane where a target obstacle is located is first determined, and then the target obstacle intention is determined based on the relative angle difference between the heading angle of the target obstacle and the lane centerline direction of the target perception lane. Then, according to the target obstacle intention of the target obstacle, the motion trajectory of the target obstacle is generated, thereby achieving the purpose of generating the trajectory of the target obstacle according to the target obstacle intention of the target obstacle; at the same time, because the target obstacle intention accurately reflects the real intention of the target obstacle, the accuracy of the motion trajectory generated according to the target obstacle intention is higher, thereby improving the accuracy of the trajectory prediction of the target obstacle.

[0017] Other features and advantages of the embodiments of the present application will be described in the subsequent description, and partly become apparent from the description, or can be understood by practicing the embodiments of the present application. The purposes and other advantages of the embodiments of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of a vehicle hardware environment suitable for an embodiment of the present application is shown.

[0020] Figure 2 A flow chart of a trajectory generation method proposed according to an embodiment of the present application is shown.

[0021] Figure 3 Show Figure 2 The corresponding step S140 of the embodiment is a flowchart in an embodiment.

[0022] Figure 4 A schematic diagram showing a trajectory generation process in an embodiment of the present application.

[0023] Figure 5 A structural block diagram of a trajectory generating device proposed in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] In order to make those skilled in the art better understand the present application scheme, the technical scheme in the present application embodiment will be clearly and completely described below in conjunction with the drawings in the present application embodiment. Obviously, the described embodiment is only a part of the present application embodiment, rather than all the embodiments. The components of the present application embodiment usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application for protection, but merely represents the selected embodiment of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0025] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0026] Reference Figure 1 , Figure 1 A schematic diagram of a vehicle hardware environment suitable for an embodiment of the present application is shown, wherein the vehicle 100 includes a driving system 110, which may have multiple built-in autonomous driving functions. The driving system 110 may store an electronic map, and the driving system 110 may plan a driving route based on the electronic map stored in the driving system, and may also control the vehicle's autonomous driving based on the planned driving route.

[0027] The driving system 110 may include a data acquisition device 111 , one or more (only one is shown in the figure) processors 112 , and a memory 113 .

[0028] The data acquisition device 111 is used to detect the driving data of the vehicle and the environmental information around the vehicle. The data acquisition device 111 may include sensors such as an in-vehicle camera, an in-vehicle infrared sensor, an in-vehicle monitoring radar, an outside camera, an outside monitoring radar, and a door monitoring radar. The driving data of the vehicle may include the speed, acceleration, heading angle, and steering wheel angle of the vehicle itself, and the environmental information around the vehicle may include lane information and obstacle information (the obstacle information may include the speed, acceleration, heading angle, etc. of the obstacle).

[0029] The processor 112 may be a microcontroller unit (MCU) having a built-in memory 113 . The memory 113 stores a program that can execute the contents of the following embodiments, and the processor 112 may execute the program stored in the memory 113 .

[0030] The processor 112 may include one or more processors. The processor 112 uses various interfaces and lines to connect various parts of the entire vehicle 100, and executes various functions of the vehicle 100 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 113, and calling data stored in the memory 113.

[0031] The memory 113 may include a random access memory (RAM) or a read-only memory (ROM). The memory 15 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 15 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc.

[0032] See also Figure 2 , Figure 2 A flow chart of a trajectory generation method proposed according to an embodiment of the present application is shown, which is used for a vehicle, and the method includes:

[0033] S110: Determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the vehicle.

[0034] The vehicle in this embodiment can be an electric vehicle or a fuel vehicle, or a sedan, an SUV, a bus, a truck, etc.; the self-vehicle refers to the vehicle itself.

[0035] The vehicle can determine a perception map based on the vehicle's driving data and the environment information around the vehicle. The perception map may include at least one lane and at least one obstacle. A lane in the perception map serves as a perception lane. Obstacles in the perception map may be stationary obstacles or moving obstacles. Obstacles in the perception map may include pedestrians, non-motor vehicles, vehicles, traffic signs, temporary construction signs, roadside objects, and the like.

[0036] The vehicle may be provided with an environment modeling module, which may have a built-in perception algorithm. The environment modeling module determines the perception map based on the built-in perception algorithm according to the driving data of the vehicle and the environmental information around the vehicle. The perception algorithm may be an algorithm for realizing the autonomous driving function of the vehicle (which may be L1 level autonomous driving or L2 level autonomous driving), etc., which will not be described in detail.

[0037] In some implementations, the target obstacle may refer to any obstacle in the perception map.

[0038] In some other embodiments, obstacles outside the perception map of the vehicle and within a preset range may be obtained as target obstacles, wherein the preset range may refer to a 3s time distance range in front of the vehicle and a 1.5 vehicle width range to the left and right of the vehicle.

[0039] Generally speaking, obstacles within the range of the perception map are obstacles that have been perceived by the vehicle and have corresponding trajectories. Therefore, it is no longer necessary to generate trajectories for them, thereby filtering out obstacles outside the perception map. However, for obstacles that are not within the range of the perception map, it is difficult to estimate the impact on the vehicle's driving process. Therefore, it is also necessary to determine whether the obstacles that are not within the range of the perception map are within the preset range. If the obstacle is within the preset range, it means that the obstacle has an impact on the vehicle's driving process, and it is necessary to determine that it is a target obstacle and predict its corresponding motion trajectory; if the obstacle is not within the preset range, it means that the obstacle has no impact on the vehicle's driving process, and the obstacle can be ignored.

[0040] In other embodiments, obstacles that are outside the perception map of the vehicle, within a preset range, and belong to a target category can also be obtained as target obstacles, wherein the target category can be a vulnerable road user category (Vulnerable road users, referred to as VRU), and the vulnerable road user category can include pedestrians, non-motor vehicles, etc.

[0041] That is, after screening out candidate obstacles that are outside the perception map of the vehicle and within a preset range, it is also necessary to select obstacles of the vulnerable road user category from the candidate obstacles as target obstacles.

[0042] After the target obstacle is determined, in one embodiment, a perception lane where the target obstacle can be determined or a perception lane to which the target obstacle is closest is used as the target perception lane where the target obstacle is located.

[0043] In another embodiment, the candidate perception lane corresponding to the target obstacle can be determined from each perception lane based on the positional relationship between the target obstacle and the lane center lines of each perception lane; and the target perception lane where the target obstacle is located can be determined from each candidate perception lane based on the lateral distance between the target obstacle and each candidate perception lane.

[0044] For each target obstacle, the sensing lane where the target obstacle is located can be determined according to the positional relationship between the target obstacle and the lane centerline of each sensing lane, and all of them are used as candidate sensing lanes where the target obstacle is located. Among them, according to the positional relationship between the target obstacle and the lane centerline of each sensing lane, the lane to which the lane centerline whose distance to the target obstacle does not exceed half the width of the sensing lane belongs can be determined as the candidate sensing lane corresponding to the target obstacle.

[0045] Afterwards, the lateral distances between the target obstacle and each candidate perception lane may be determined, and the candidate perception lane with the smallest lateral distance may be selected as the target perception lane where the target obstacle is located.

[0046] S120: Determine a relative angle difference between a heading angle of the target obstacle and a lane centerline direction of the target perception lane.

[0047] The heading angle of the target obstacle can be obtained from the environmental information around the ego-vehicle, and then the angle difference between the heading angle of the target obstacle and the lane centerline direction of the target perception lane is determined as the relative angle difference.

[0048] The lane centerline direction of the target perception lane refers to the lane centerline of the target perception lane and the direction along which the vehicle is traveling.

[0049] S130: Determine the target obstacle intention corresponding to the target angle interval where the relative angle difference is located.

[0050] Each angle interval corresponds to a respective obstacle intention. In the present application, each angle interval can be set based on demand, and a respective obstacle intention can be configured for each angle interval.

[0051] For example, taking the case where the obstacle is located on the left side of the vehicle, the obstacle intention corresponding to each angle interval is set with reference to Formula 1, which is as follows:

[0052]

[0053] Among them, the eight obstacle intentions shown in Formula 1 can also be divided into different intention types. For example, cutting into the lane to the left front, cutting into the lane to the left rear, cutting into the lane to the right front, and cutting into the lane to the right rear all belong to the lane cutting intention type, driving in the opposite direction along the lane and driving straight along the lane both belong to the lane driving intention type, and crossing the lane to the left and crossing the lane to the right both belong to the lane crossing intention type.

[0054] The angle interval where the relative angle difference is located can be directly determined as the target angle interval, and then the obstacle intention corresponding to the target angle interval is obtained as the target obstacle intention of the target obstacle. When there are multiple target obstacles, a relative angle difference is determined for each target obstacle, so that each target obstacle determines a target angle interval, and then each target obstacle determines a target obstacle intention.

[0055] S140: Generate a motion trajectory of the target obstacle according to the intention of the target obstacle.

[0056] After the target obstacle intention of each target obstacle is obtained, the trajectory of each target obstacle is predicted according to the target obstacle intention of each target obstacle to obtain the motion trajectory of each target obstacle.

[0057] First, the target perception lane where the target obstacle is located is determined, and then the target obstacle intention is determined based on the relative angle difference between the heading angle of the target obstacle and the lane centerline direction of the target perception lane. Then, according to the target obstacle intention of the target obstacle, the motion trajectory of the target obstacle is generated. Thus, the purpose of generating the trajectory of the target obstacle according to the target obstacle intention of the target obstacle is achieved. At the same time, because the target obstacle intention accurately reflects the real intention of the target obstacle, the accuracy of the motion trajectory generated according to the target obstacle intention is higher, thereby improving the accuracy of the trajectory prediction of the target obstacle.

[0058] In addition, when the target obstacle is a VRU category obstacle located outside the perception map range, the target obstacle intention accurately reflects the true intention of the target obstacle, so that the accuracy of the motion trajectory generated according to the target obstacle intention is higher, thereby improving the accuracy of the trajectory prediction of the VRU category obstacles.

[0059] In one embodiment, if Figure 3 As shown, S140 may include:

[0060] S210: If the target obstacle intention is of the type of lane-cutting intention, determine the aiming point corresponding to the target obstacle according to the position information and speed of the target obstacle.

[0061] That is, when the target obstacle intends to cut into the lane to the left front, to the left rear, to the right front, or to the right rear, S210 is executed.

[0062] When the target obstacle intention of the target obstacle belongs to the lane cutting intention type, it means that the target obstacle wants to enter the lane where the ego vehicle is located, and after entering the lane where the ego vehicle is located, continue to travel along the lane where the ego vehicle is located. In other words, when the target obstacle intention of the target obstacle belongs to the lane cutting intention type, it means that the target obstacle wants to merge into the lane where the ego vehicle is located.

[0063] The aiming point corresponding to the target obstacle is used to indicate the position of the target obstacle when the target obstacle completes merging into the lane where the ego vehicle is located. The ego vehicle can estimate the position of the target obstacle when the target obstacle completes merging into the lane where the ego vehicle is located based on the position information and speed of the target obstacle, and use it as the aiming point corresponding to the target obstacle.

[0064] In some implementations, a target coordinate system may be constructed based on the target boundary line closest to the target obstacle in the target perception lane; the projection point of the target obstacle to the target coordinate axis of the target coordinate system is determined as the key point corresponding to the target obstacle; based on the position information of the key point and the speed projection of the speed of the target obstacle under each coordinate axis in the target coordinate system, the aiming point corresponding to the target obstacle is determined. The target coordinate system may be an SL coordinate system (also called a Frenet coordinate system), and the target coordinate axis is the S axis of the SL coordinate system.

[0065] For each target obstacle, there is a corresponding target perception lane. The boundary line closest to the target obstacle is determined from the two boundary lines of the target perception lane corresponding to the target obstacle as the target boundary line corresponding to the target obstacle. The SL coordinate system is constructed with the target boundary line of the target obstacle as the S axis of the SL coordinate system (the projection point of the vehicle's position on the S axis can be selected as the origin of the SL coordinate system, where the vehicle's position can be the midpoint of the vehicle's rear axle or the position of the vehicle's center of mass) to obtain the target coordinate system. In this way, a target coordinate system is constructed for each target obstacle.

[0066] Then, a perpendicular line is drawn through the position of the target obstacle's centroid, perpendicular to the S axis of the target coordinate system corresponding to the target obstacle. The foot of the perpendicular line on the S axis of the target coordinate system corresponding to the target obstacle is the projection point of the target obstacle to the target coordinate axis of the target coordinate system. Therefore, the foot of the perpendicular line is used as the key point corresponding to the target obstacle. Obviously, the key point corresponding to the target obstacle is located on the S axis of the target coordinate system corresponding to the target obstacle, that is, the coordinates of the key point corresponding to the target obstacle are (S0, 0), where S0 is the coordinate of the key point corresponding to the target obstacle in the direction of the S axis of the target coordinate system. In this way, a key point is determined for each target obstacle.

[0067] Afterwards, for each target obstacle, the aiming point corresponding to the target obstacle is determined according to the position information of the key points of the target obstacle and the velocity projection of the velocity of the target obstacle under each coordinate axis in the target coordinate system.

[0068] In some implementations, the coordinates of the aiming point corresponding to the target obstacle in the S coordinate direction of the target coordinate system may be calculated by formula 2 in combination with the position information of the key point of the target obstacle, the lateral distance l between the target obstacle and the coordinate axis corresponding to the target obstacle (that is, the coordinate of the target obstacle in the L coordinate axis direction in the corresponding target coordinate system), and the velocity projection of the speed of the target obstacle in each coordinate axis in the corresponding target coordinate system. Thus, the coordinates of the aiming point corresponding to the target obstacle in the S coordinate axis direction of the target coordinate system are determined according to the coordinates of the aiming point corresponding to the target obstacle in the S coordinate axis direction of the target coordinate system. Formula 2 is as follows:

[0069] S1=S0+S′*l / l′

[0070] Among them, S1 is the coordinate of the aiming point corresponding to the target obstacle in the S coordinate direction of the target coordinate system, S0 is the coordinate of the key point corresponding to the target obstacle in the S coordinate direction of the target coordinate system, S ′ is the velocity projection of the target obstacle in the S-axis direction in the corresponding target coordinate system, l ′ It is the velocity projection of the target obstacle in the L-axis direction in the corresponding target coordinate system.

[0071] After obtaining the coordinates of the aiming point corresponding to the target obstacle in the S coordinate axis direction of the target coordinate system, it can be determined that the coordinates of the aiming point corresponding to the target obstacle in the target coordinate system are (S1, 0).

[0072] S220: Perform arc trajectory fitting according to the position information of the target obstacle and the position information of the aiming point corresponding to the target obstacle to obtain a first motion trajectory of the target obstacle.

[0073] The position information of the target obstacle may refer to the coordinates of the target obstacle in the target coordinate system corresponding to the target obstacle, and the position information of the aiming point corresponding to the target obstacle may refer to the coordinates of the aiming point corresponding to the target obstacle in the target coordinate system corresponding to the target obstacle. Accordingly, the coordinates of the target obstacle in the target coordinate system corresponding to the target obstacle may be converted into coordinates (x0, y0) of the xoy plane in the world coordinate system. At the same time, the coordinates (S1, 0) of the aiming point corresponding to the target obstacle in the target coordinate system corresponding to the target obstacle are converted into coordinates (x1, y1) of the xoy plane in the world coordinate system, and then the arc trajectory is fitted according to (x0, y0) and (x1, y1), and the fitted arc trajectory is obtained as the first motion trajectory of the target obstacle.

[0074] In this embodiment, arc trajectory fitting can be performed based on the position information of the target obstacle and the position information of the aiming point corresponding to the target obstacle by means of polynomial fitting (for example, quadratic polynomial fitting) to obtain the first motion trajectory.

[0075] Optionally, the position information of the target obstacle may be used as the starting point coordinates of the first motion trajectory, and the position information of the aiming point corresponding to the target obstacle may be used as the ending point coordinates of the first motion trajectory, and arc trajectory fitting may be performed to obtain the first motion trajectory.

[0076] It is understandable that, when there are multiple target obstacles, the first motion trajectory of each target obstacle is determined according to the aforementioned process of S220.

[0077] S230: Perform straight line trajectory fitting according to the position information of the aiming point of the target obstacle to obtain a second motion trajectory of the target obstacle.

[0078] The position information of the aiming point of the target obstacle can be used as the coordinates of the starting point of the second motion trajectory, and a straight line trajectory of a specified length can be fitted according to the extension direction of the lane where the vehicle is located as the second motion trajectory of the target obstacle. The specified length can be, for example, 10m or 20m.

[0079] In this embodiment, the target obstacle intends to cut into the lane, and the target obstacle wants to merge into the lane where the ego vehicle is located. Therefore, after the target obstacle merges into the lane where the ego vehicle is located, it will move along the extension direction of the lane where the ego vehicle is located.

[0080] The position information of the aiming point corresponding to the target obstacle may refer to the coordinates of the aiming point corresponding to the target obstacle in the target coordinate system corresponding to the target obstacle. Accordingly, the coordinates (S1, 0) of the aiming point corresponding to the target obstacle in the target coordinate system corresponding to the target obstacle are converted into coordinates (x1, y1) of the xoy plane in the world coordinate system, and then a straight line trajectory is fitted according to (x1, y1) to obtain the fitted straight line trajectory as the second motion trajectory of the target obstacle.

[0081] It is understandable that, when there are multiple target obstacles, the second motion trajectory is determined for each target obstacle according to the aforementioned process of S230.

[0082] S240: Obtain a motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory.

[0083] The first motion trajectory and the second motion trajectory of the target obstacle may be connected into one trajectory as the motion trajectory of the target obstacle.

[0084] As mentioned above, the first motion trajectory of the target obstacle can be an arc trajectory with the position of the target obstacle as the starting point and the aiming point of the target obstacle as the end point. The second motion trajectory of the target obstacle can be a straight line trajectory with the aiming point of the target obstacle as the starting point. The first motion trajectory and the second motion trajectory of the target obstacle can be connected into one trajectory with the aiming point of the target obstacle as the connecting point to obtain the motion trajectory of the target obstacle.

[0085] It is understandable that, when there are multiple target obstacles, the respective motion trajectories of each target obstacle are determined according to the aforementioned process of S240.

[0086] In this embodiment, the target obstacle intention belongs to the type of lane cutting intention, and it is determined that the target obstacle wants to merge into the lane where the own vehicle is located. Therefore, the motion of the target obstacle is segmented and predicted to predict a first motion trajectory before merging into the lane where the own vehicle is located and a second motion trajectory after merging into the lane where the own vehicle is located. The first motion trajectory and the second motion trajectory accurately indicate the motion of the target obstacle, so that the motion trajectory of the target obstacle obtained by combining the first motion trajectory and the second motion trajectory has a higher accuracy, thereby achieving accurate prediction of the trajectory of the target obstacle.

[0087] In one embodiment, S140 further includes: if the target obstacle intends to be a driving along the lane type or a driving across the lane type, performing straight line trajectory fitting according to the position information and speed of the target obstacle to obtain the motion trajectory of the target obstacle.

[0088] As mentioned in the above examples, in the obstacle intention, driving in the opposite direction along the lane and driving straight along the lane both belong to the driving intention type of driving along the lane, and crossing the lane to the left and crossing the lane to the right both belong to the driving intention type of crossing the lane.

[0089] When the target obstacle intends to move along the lane or to cross the lane, it is determined that the movement process of the target obstacle will not change the movement direction. Therefore, the movement direction of the target obstacle can be obtained (the direction of the speed of the target obstacle can be determined as the movement direction of the target obstacle). Then, according to the position information of the target obstacle and the movement state of the target obstacle (for example, including the movement direction, speed, and acceleration, etc.), the kinematic deduction of the target obstacle is predicted (for example, determining that the target obstacle is uniform linear motion, uniformly accelerated linear motion, or uniformly decelerated linear motion, etc.), and a straight line trajectory is fitted to obtain the fitted straight line trajectory as the movement trajectory of the target obstacle.

[0090] Among them, the position information of the target obstacle may refer to the coordinates of the target obstacle in the target coordinate system corresponding to the target obstacle, and the coordinates of the target obstacle in the target coordinate system corresponding to the target obstacle may be converted into coordinates (x0, y0) of the xoy plane in the world coordinate system, and then (x0, y0) is used as the starting point, based on the motion state of the target obstacle, the kinematic deduction of the target obstacle is predicted, and a straight line trajectory is fitted to obtain the fitted straight line trajectory as the motion trajectory of the target obstacle.

[0091] It can be understood that, when there are multiple target obstacles of the type of along-lane driving intention or the type of cross-lane driving intention, the target obstacles can be fitted with a straight line trajectory according to the above process to obtain a motion trajectory.

[0092] In this embodiment, the target obstacle intends to be of the type of driving along the lane or the type of driving across the lane, and it is determined that the target obstacle maintains the original motion state. Therefore, the motion of the target obstacle is directly deduced in linear motion to obtain the motion trajectory of the target obstacle. The motion trajectory of the target obstacle can accurately indicate the motion state of the target obstacle in the future, thereby realizing accurate prediction of the trajectory of the target obstacle.

[0093] In one embodiment, after S140, the method may further include: sampling a plurality of sampling points from the motion trajectory; and performing curve trajectory fitting based on the plurality of sampling points to obtain an adjusted motion trajectory of the target obstacle.

[0094] In this embodiment, the motion trajectory of the target obstacle may be the straight line trajectory determined in the above embodiment, or may be the curved trajectory combining the circular arc and the straight line determined in the above embodiment.

[0095] A point can be sampled from the motion trajectory of the target obstacle at equal time intervals (e.g., every 0.1s) as a sampling point, thereby obtaining multiple sampling points sampled from the motion trajectory of the target obstacle, and then, multiple curve trajectory fitting is performed based on the multiple sampling points to obtain the fitted curve trajectory as the adjusted motion trajectory of the target obstacle. The fitting method for fitting the multiple sampling points can be quadratic polynomial fitting, cubic polynomial fitting, and quartic polynomial fitting, etc.

[0096] In this embodiment, after obtaining the motion trajectory of the target obstacle, multiple sampling points are sampled for the motion trajectory of the target obstacle, and then curve fitting is performed based on the multiple sampling points to achieve smoothing of the motion trajectory of the target obstacle, so that the accuracy of the adjusted motion trajectory after smoothing is higher, further improving the accuracy of the trajectory prediction of the target obstacle.

[0097] In one embodiment, after S140, the method further includes: sampling multiple sampling points from the motion trajectory; obtaining the speed and acceleration of the target obstacle at the multiple sampling points; if there is at least one sampling point where the speed does not conform to the speed range or the acceleration does not conform to the acceleration range, determining that the motion trajectory is abnormal, and deleting the motion trajectory.

[0098] After determining multiple sampling points according to the method in the above embodiment, the speed and acceleration at the multiple sampling points can be obtained. Among them, for each sampling point, the motion state of the target obstacle when it is located at the sampling point can be determined according to the motion state of the target obstacle perceived by the vehicle (for example, including motion direction, speed and acceleration, etc., the motion state refers to the motion state of the target obstacle at the initial moment of the trajectory prediction of this application) and the time difference between the sampling point and the location of the target obstacle (the location of the target obstacle at the initial moment of the trajectory prediction of this application), and the speed and acceleration of the target obstacle at the sampling point can be determined according to the motion state of the target obstacle when it is located at the sampling point.

[0099] In another embodiment, the trajectory length between any two adjacent sampling points may be determined, and then the speed and acceleration between the two adjacent sampling points may be determined based on the trajectory length and the time interval between the two adjacent sampling points, and the speed and acceleration between the two adjacent sampling points may be determined as the speed and acceleration of the target obstacle at the two adjacent sampling points.

[0100] The speed range and the acceleration range can be set based on requirements, for example, the speed range is a value greater than 0, and the acceleration range is no more than 5 meters per second squared.

[0101] If the speed at a certain sampling point does not conform to the speed range or the acceleration does not conform to the acceleration range, it is unreasonable to determine the motion state of the target obstacle at the sampling point. Therefore, it is also unreasonable to determine the motion trajectory determined for the target obstacle, and the motion trajectory is directly deleted.

[0102] If the speeds of all sampling points meet the speed range and the accelerations meet the acceleration range, it is reasonable to determine the motion state of the target obstacle, and therefore, the motion trajectory can be retained.

[0103] It is worth mentioning that in some embodiments, after obtaining the adjusted motion trajectory of the target obstacle, multiple adjustment sampling points can be sampled from the adjusted motion trajectory; the speed and acceleration of the target obstacle at multiple adjustment sampling points can be obtained; if there is at least one adjustment sampling point where the speed does not conform to the speed range or the acceleration does not conform to the acceleration range, it is determined that the adjusted motion trajectory is abnormal and the adjusted motion trajectory is deleted. The method for determining the adjustment sampling points refers to the method for determining the sampling points from the motion trajectory described above, and will not be described in detail.

[0104] That is to say, if the speed at a certain adjustment sampling point does not conform to the speed range or the acceleration does not conform to the acceleration range, it is unreasonable to determine the motion state of the target obstacle at the adjustment sampling point. Therefore, it is also unreasonable to determine the adjustment motion trajectory determined for the target obstacle, and the adjustment motion trajectory is directly deleted.

[0105] If the speeds of all the adjustment sampling points meet the speed range and the accelerations meet the acceleration range, it is reasonable to determine that the target obstacle is in the adjusted motion state, and therefore, the adjusted motion trajectory can be retained.

[0106] For example, the trajectory generation process is as follows: Figure 4 As shown, the vehicle obtains a perception map, selects target obstacles according to the perception map, and then predicts the intention of the target obstacle to determine the intention of the target obstacle. Then, the motion trajectory of the target obstacle is predicted according to the means in the aforementioned embodiment, and then the motion trajectory of the target obstacle is post-processed (the post-processing may include the aforementioned smoothing processing, and the determination of the speed and acceleration at the sampling point (or the adjustment sampling point)), traverses all target obstacles, and finally obtains a valid target obstacle with a retained motion trajectory (or retained adjusted motion trajectory), and outputs the trajectory predicted by the valid target obstacle as a prediction result.

[0107] In this embodiment, the motion trajectory is also smoothed and the motion state (that is, the aforementioned speed and acceleration) of each sampling point in the motion trajectory (or each adjusted sampling point in the adjusted motion trajectory) is determined, so that the obtained trajectory is smooth and accurate and conforms to the motion state of the target obstacle, making the retained motion trajectory more accurate, further improving the accuracy of the trajectory prediction of the target obstacle.

[0108] See attached Figure 5 , Figure 5 The structure block diagram of a trajectory generation device proposed in one embodiment of the present application is shown. For a vehicle, the device 800 includes:

[0109] A lane determination module 810 is used to determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the vehicle;

[0110] An angle determination module 820, for determining a relative angle difference between a heading angle of a target obstacle and a lane centerline direction of a target sensing lane;

[0111] The intention determination module 830 is used to determine the target obstacle intention corresponding to the target angle interval where the relative angle difference is located; each angle interval corresponds to its own obstacle intention;

[0112] The trajectory generation module 840 is used to generate a motion trajectory of the target obstacle according to the intention of the target obstacle.

[0113] Optionally, the lane determination module 810 is further used to determine a candidate perception lane corresponding to the target obstacle from each perception lane based on a positional relationship between the target obstacle and the lane center lines of each perception lane; and to determine a target perception lane where the target obstacle is located from each candidate perception lane based on a lateral distance between the target obstacle and each candidate perception lane.

[0114] Optionally, the trajectory generation module 840 is further used to determine the aiming point corresponding to the target obstacle according to the position information and speed of the target obstacle if the target obstacle intends to cut into the lane; perform arc trajectory fitting according to the position information of the target obstacle and the position information of the aiming point corresponding to the target obstacle to obtain a first motion trajectory of the target obstacle; perform straight line trajectory fitting according to the position information of the aiming point of the target obstacle to obtain a second motion trajectory of the target obstacle; and obtain the motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory.

[0115] Optionally, the trajectory generation module 840 is further used to construct a target coordinate system according to the target boundary line closest to the target obstacle in the target perception lane; determine the projection point of the target obstacle to the target coordinate axis of the target coordinate system as the key point corresponding to the target obstacle; and determine the aiming point corresponding to the target obstacle based on the position information of the key point and the speed projection of the speed of the target obstacle under each coordinate axis in the target coordinate system.

[0116] Optionally, the trajectory generation module 840 is further used to perform straight line trajectory fitting according to the position information and speed of the target obstacle to obtain the motion trajectory of the target obstacle if the target obstacle intends to move along the lane or cross the lane.

[0117] Optionally, the device further includes a post-processing module, which is used to sample a plurality of sampling points from the motion trajectory; and perform curve trajectory fitting based on the plurality of sampling points to obtain an adjusted motion trajectory of the target obstacle.

[0118] Optionally, the post-processing module is also used to sample multiple sampling points from the motion trajectory; obtain the speed and acceleration of the target obstacle at the multiple sampling points; if there is at least one sampling point where the speed does not conform to the speed range or the acceleration does not conform to the acceleration range, determine that the motion trajectory is abnormal, and delete the motion trajectory.

[0119] Optionally, the device also includes an obstacle determination module, which is used to obtain obstacles that are outside the perception map of the vehicle and within a preset range as target obstacles; the perception map is determined based on the driving data of the vehicle and the environmental information around the vehicle.

[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0121] In addition, each function in each embodiment of the present application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of software function module.

[0122] In addition, each function in each embodiment of the present application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of software function module.

[0123] In addition, each function in each embodiment of the present application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of software function module.

[0124] On the other hand, the present application also provides a computer-readable storage medium, in which program code is stored. The program code can be called by a processor to execute the method described in the above method embodiment.

[0125] The computer readable storage medium may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a cluster of ROMs. Optionally, the computer readable storage medium comprises a non-transitory computer-readable storage medium. The computer readable storage medium has storage space for program codes that execute any of the method steps in the above method. These program codes can be read from or written to one or more computer program products. The program code can be compressed, for example, in an appropriate form.

[0126] 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 make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A trajectory generation method, characterized in that: The method comprises: Determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the ego-vehicle; Determining a relative angle difference between a heading angle of the target obstacle and a lane centerline direction of the target sensing lane; Determine the target obstacle intention corresponding to the target angle interval where the relative angle difference is located; each angle interval corresponds to its own obstacle intention; A motion trajectory of the target obstacle is generated according to the target obstacle intention.

2. The method according to claim 1, characterized in that The step of determining a target perception lane where a target obstacle is located from at least one perception lane perceived by the vehicle includes: Determining a candidate perception lane corresponding to the target obstacle from the perception lanes according to a positional relationship between the target obstacle and the lane center lines of the perception lanes; According to the lateral distance between the target obstacle and each of the candidate perception lanes, a target perception lane where the target obstacle is located is determined from each of the candidate perception lanes.

3. The method according to claim 1, characterized in that The step of generating a motion trajectory of the target obstacle according to the intention of the target obstacle includes: If the target obstacle is intended to cut into the lane, determine the aiming point corresponding to the target obstacle according to the position information and speed of the target obstacle; Perform arc trajectory fitting according to the position information of the target obstacle and the position information of the aiming point corresponding to the target obstacle to obtain a first motion trajectory of the target obstacle; Performing linear trajectory fitting according to the position information of the aiming point of the target obstacle to obtain a second motion trajectory of the target obstacle; Based on the first motion trajectory and the second motion trajectory, a motion trajectory of the target obstacle is obtained.

4. The method according to claim 3, characterized in that The determining, according to the position information and speed of the target obstacle, the aiming point corresponding to the target obstacle includes: Constructing a target coordinate system according to a target boundary line in the target sensing lane that is closest to the target obstacle; Determine a projection point of the target obstacle onto a target coordinate axis of the target coordinate system as a key point corresponding to the target obstacle; Based on the position information of the key point and the velocity projection of the velocity of the target obstacle under each coordinate axis in the target coordinate system, an aiming point corresponding to the target obstacle is determined.

5. The method according to claim 1, characterized in that The step of generating a motion trajectory of the target obstacle according to the intention of the target obstacle includes: If the target obstacle intends to move along the lane or to cross the lane, a straight line trajectory is fitted according to the position information and speed of the target obstacle to obtain the motion trajectory of the target obstacle.

6. The method according to any one of claims 1 to 5, characterized in that: After obtaining the motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory, the method includes: Sampling a plurality of sampling points from the motion trajectory; Curve trajectory fitting is performed based on the multiple sampling points to obtain an adjusted motion trajectory of the target obstacle.

7. The method according to any one of claims 1 to 5, characterized in that: After obtaining the motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory, the method includes: Sampling a plurality of sampling points from the motion trajectory; Obtaining the velocity and acceleration of the target obstacle at the plurality of sampling points; If the speed at at least one sampling point does not conform to the speed range or the acceleration does not conform to the acceleration range, it is determined that the motion trajectory is abnormal and the motion trajectory is deleted.

8. The method according to claim 1, characterized in that Before determining the target perception lane where the target obstacle is located from at least one perception lane perceived by the vehicle, the method further includes: Obstacles located outside the perception map of the vehicle and within a preset range are obtained as target obstacles; the perception map is determined based on the driving data of the vehicle and the environmental information around the vehicle.

9. A trajectory generating device, characterized in that: The device comprises: A lane determination module, used to determine a target perception lane where a target obstacle is located from at least one perception lane perceived by the vehicle; An angle determination module, used to determine the relative angle difference between the heading angle of the target obstacle and the lane centerline direction of the target sensing lane; An intention determination module, used to determine the target obstacle intention corresponding to the target angle interval where the relative angle difference is located; each angle interval corresponds to its own obstacle intention; The trajectory generation module is used to generate the motion trajectory of the target obstacle according to the intention of the target obstacle.

10. A vehicle, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program code executable by a processor, and when the program code is executed by the processor, the processor executes the method according to any one of claims 1 to 8.

Images