Trajectory generation method and device, vehicle and storage medium

Abstract

The application discloses a trajectory generation method and device, a vehicle and a readable storage medium. The method comprises the following steps: determining a target perception lane in which a target obstacle is located from at least one perceived lane perceived by a self vehicle; determining a relative angle difference between a heading angle of the target obstacle and a lane center line direction of the target perception lane; determining a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located; each angle interval corresponds to a respective obstacle intention; and generating a motion trajectory of the target obstacle according to the target obstacle intention. According to the method, the trajectory of the target obstacle is predicted, 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 technical field of vehicles, and more particularly, to a trajectory generation method and device, a vehicle, and a computer readable storage medium. BACKGROUND

[0002] In the face of complex and variable dynamic traffic environment, accurate and reliable estimation of the trajectory of an obstacle is an important basis for decision planning of an unmanned vehicle. Therefore, a means is urgently needed to accurately predict the trajectory of an obstacle. SUMMARY

[0003] The present application provides 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, and the method comprises:

[0005] determining a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle;

[0006] determining a relative angle difference between a heading angle of the target obstacle and a direction of a lane center line of the target perception lane;

[0007] determining a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located; each angle interval corresponds to a respective obstacle intention;

[0008] generating a motion trajectory of the target obstacle according to the target obstacle intention.

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

[0010] a lane determination module configured to determine a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle;

[0011] an angle determination module configured to determine a relative angle difference between a heading angle of the target obstacle and a direction of a lane center line of the target perception lane;

[0012] an intention determination module configured to determine a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located; each angle interval corresponds to a respective obstacle intention;

[0013] a trajectory generation module configured to generate a motion trajectory of the target obstacle according to the target obstacle intention.

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

[0015] In a fourth aspect, the embodiments of the present application further provide a computer-readable storage medium, which stores processor-executable program code, and the program code, when executed by a processor, causes the processor to execute the method described above.

[0016] The trajectory generation method, device, vehicle and computer-readable storage medium provided by the present application, in the present application, first determine the target sensing lane where the target obstacle is located, and then determine the target obstacle intention based on the relative angle difference between the heading angle of the target obstacle and the direction of the lane center line of the target sensing lane, and then generate the motion trajectory of the target obstacle according to the target obstacle intention of the target obstacle. 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, since 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, and the accuracy of the trajectory prediction of the target obstacle is improved.

[0017] Other features and advantages of the embodiments of the present application will be described in the following description, and some will become apparent from the description, or will be understood through implementation of the embodiments of the present application. The purposes and other advantages of the embodiments of the present application can be achieved and obtained through the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

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

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

[0021] Figure 3 A flowchart of step S140 of the corresponding embodiment is shown. Figure 2 A flowchart of step S140 of the corresponding embodiment is shown.

[0022] Figure 4 This diagram illustrates a trajectory generation process according to an embodiment of the present application.

[0023] Figure 5 A structural block diagram of a trajectory generation device according to an embodiment of this application is shown. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. The components of the embodiments of the present application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are within the scope of protection of the present application.

[0025] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] Reference Figure 1 , Figure 1 A schematic diagram of a vehicle hardware environment applicable to an embodiment of this application is shown. The vehicle 100 includes a driving system 110, which can have multiple built-in autonomous driving functions. The driving system 110 can store electronic maps. The driving system 110 can plan driving routes based on the electronic maps it stores, and can also control the vehicle to drive autonomously based on the planned driving routes.

[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 memory 113.

[0028] The data acquisition device 111 is configured to detect driving data of the vehicle and environmental information around the vehicle. The data acquisition device 111 can include an in-vehicle camera, an in-vehicle infrared sensor, an in-vehicle monitoring radar, an out-vehicle camera, an out-vehicle monitoring radar, a door monitoring radar, and the like. The driving data of the vehicle can include speed, acceleration, heading angle, steering wheel angle, and the like of the vehicle itself. The environmental information around the vehicle can include lane information and obstacle information (which can include speed, acceleration, heading angle, and the like of the obstacle), and the like.

[0029] The processor 112 can be a micro control unit (MCU). The processor 112 can include a built-in memory 113 in which programs for implementing the embodiments described below are stored. The processor 112 can execute the programs stored in the memory 113.

[0030] The processor 112 can include one or more processors. The processor 112 can be connected to various parts of the vehicle 100 through various interfaces and lines, and can execute various functions of the vehicle 100 and process 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 can include a random access memory (RAM) and a read-only memory (ROM). The memory 113 can be configured to store instructions, programs, codes, code sets or instruction sets. The memory 113 can include a program storage area and a data storage area. The program storage area can store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, and the like), instructions for implementing various method embodiments described below, and the like.

[0032] Please refer to Figure 2 , Figure 2 A flowchart of a trajectory generation method according to an embodiment of the present application is shown. The method is used for a vehicle and includes the following steps.

[0033] In S110, a target perception lane in which a target obstacle is located is determined from at least one perception lane perceived by the ego vehicle.

[0034] The vehicle in the embodiment can be an electric vehicle or a fuel vehicle, and can be a car, a suv, a bus, a truck, or the like. The ego vehicle refers to the vehicle itself.

[0035] The ego vehicle can determine a perception map according to driving data of the ego vehicle and environmental information around the ego vehicle, the perception map can include at least one lane and at least one obstacle, one lane in the perception map is a perceived lane, and the obstacle in the perception map can be a stationary obstacle or a moving obstacle. The obstacle in the perception map can include a pedestrian, a non-motor vehicle, a vehicle, a traffic sign, a temporary construction sign, a roadside object, and the like.

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

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

[0038] In yet some embodiments, an obstacle located outside the perception map of the ego vehicle and within a preset range can also be acquired as a target obstacle, wherein the preset range can refer to a 3s time-to-collision range in front of the ego vehicle and a 1.5 vehicle width range on the left and right of the ego vehicle.

[0039] Generally, the obstacle located within the range of the perception map is an obstacle that has been perceived by the ego vehicle and has a corresponding trajectory, so that the trajectory generation for the obstacle can be omitted, thereby screening the obstacle outside the perception map. However, the obstacle outside the range of the perception map is difficult to estimate in the driving process of the ego vehicle, so it is necessary to determine whether the obstacle outside the range of the perception map is located within the preset range. If the obstacle is located within the preset range, it means that the obstacle has an impact on the driving process of the ego vehicle, and it is necessary to determine the obstacle as a target obstacle and predict a corresponding motion trajectory for the obstacle. If the obstacle is not located within the preset range, it means that the obstacle has no impact on the driving process of the ego vehicle, and the obstacle can be ignored at this time.

[0040] In some other embodiments, an obstacle located outside the perception map of the ego vehicle, within the preset range, and belonging to a target category can also be acquired as a target obstacle, wherein the target category can be a vulnerable road user category (VRU), and the vulnerable road user category can include a pedestrian and a non-motor vehicle.

[0041] That is, after the candidate obstacle located outside the perception map of the ego vehicle and within the preset range is screened out, it is necessary to select the obstacle belonging to the vulnerable road user category as a target obstacle from the candidate obstacle.

[0042] After determining the target obstacle, in an embodiment, a target perception lane where the target obstacle is located is determined as a target perception lane where the target obstacle is located, for a perception lane where the target obstacle is determined to be located or a perception lane closest to the target obstacle.

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

[0044] For each target obstacle, a target perception lane where the target obstacle is located can be determined according to the positional relationship between the target obstacle and the lane center line of each perception lane, and each target perception lane is a candidate perception lane where the target obstacle is located. Among them, the candidate perception lane corresponding to the target obstacle can be determined according to the positional relationship between the target obstacle and the lane center line of each perception lane, and the lane to which the lane center line with a distance not more than half the width of the perception lane from the target obstacle belongs is the candidate perception lane corresponding to the target obstacle.

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

[0046] S120, determine the relative angle difference between the heading angle of the target obstacle and the direction of the lane center line 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 direction of the lane center line of the target perception lane is determined as the relative angle difference.

[0048] Among them, the direction of the lane center line of the target perception lane refers to the lane center line of the target perception lane and the direction along which the vehicle travels.

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

[0050] Among them, each angle interval corresponds to a respective obstacle intention. In this application, each angle interval can be set based on the demand, and each angle interval is configured with a respective obstacle intention.

[0051] For example, when the obstacle is located on the left side of the ego vehicle, the obstacle intention corresponding to each angle interval is set according to Formula One, and Formula One is as follows:

[0052]

[0053] The eight obstacle intentions shown in Formula One can also be divided into different intention types. For example, cutting into the lane from the left front, cutting into the lane from the left rear, cutting into the lane from the right front, cutting into the lane from the right rear all belong to the lane-cutting intention type, driving along the lane in reverse and driving along the lane straight all belong to the lane-following intention type, and crossing the lane from the left and crossing the lane from the right all belong to the lane-crossing intention type.

[0054] The relative angle difference can be directly determined to be in an angle interval, as a 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 the target obstacle is multiple, a relative angle difference is determined for each target obstacle, so that each target obstacle determines a target angle interval, and each target obstacle determines a target obstacle intention.

[0055] S140, generating a motion trajectory of the target obstacle according to the target obstacle intention.

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

[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 direction of the lane center line of the target perception lane, and then the motion trajectory of the target obstacle is generated according to the target obstacle intention of the target obstacle. Thus, the purpose of generating the trajectory of the target obstacle according to the target obstacle intention of the target obstacle is achieved, and since 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 high, and the accuracy of the trajectory prediction of the target obstacle is improved.

[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 real intention of the target obstacle, so that the accuracy of the motion trajectory generated according to the target obstacle intention is high, and the accuracy of the trajectory prediction of the VRU category obstacle is improved.

[0059] In an embodiment, as shown in FIG. 1B, S140 can include: Figure 3

[0060] S210, if the target obstacle intention belongs to the lane-cutting intention type, determining the aiming point corresponding to the target obstacle according to the position information and the speed of the target obstacle.

[0061] ​That is, when the target obstacle intention of the target obstacle is any one of cutting into the lane in the front left, cutting into the lane in the rear left, cutting into the lane in the front right, and cutting into the lane in the rear right, 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 continue driving along the lane where the ego vehicle is located after entering 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 as the aiming point corresponding to the target obstacle according to the position information and speed of the target obstacle.

[0064] In some embodiments, a target coordinate system can be constructed according to the target boundary line closest to the target obstacle in the target perception lane; a 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; and the aiming point corresponding to the target obstacle is determined based on the position information of the key point and the speed projection of the target obstacle under each coordinate axis in the target coordinate system. The target coordinate system can 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, a target perception lane corresponding to the target obstacle is determined, and 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 ego vehicle position on the S axis can be selected as the origin of the SL coordinate system, and the ego vehicle position can be the midpoint of the rear axle of the ego vehicle or the position of the center of mass of the ego vehicle), and the target coordinate system is obtained. In this way, for each target obstacle, a target coordinate system is constructed.

[0066] Then, a perpendicular line is drawn through the position of the centroid of the target obstacle, which is 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 is 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), wherein S0 is the coordinate of the key point corresponding to the target obstacle in the S-axis direction of the target coordinate system. In this way, a key point is determined for each target obstacle.

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

[0068] In some embodiments, the coordinate of the aiming point corresponding to the target obstacle in the S-axis direction of the target coordinate system can be calculated by Formula Two in combination with the position information of the key point of the target obstacle, the transverse distance l of the target obstacle to the coordinate axis corresponding to the target obstacle (that is, the L-axis direction coordinate of the target obstacle in the target coordinate system), and the velocity projection of the speed of the target obstacle in each coordinate axis of the corresponding target coordinate system, so as to determine the coordinates of the aiming point corresponding to the target obstacle according to the coordinate of the aiming point corresponding to the target obstacle in the S-axis direction of the target coordinate system. Formula Two is as follows:

[0069] S1=S0+S′*l / l′

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

[0071] After obtaining the coordinate of the aiming point corresponding to the target obstacle in the S-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, according to the position information of the target obstacle and the position information of the aiming point corresponding to the target obstacle, performing circular arc trajectory fitting to obtain the first motion trajectory of the target obstacle.

[0073] The position information of the target obstacle can be the coordinates of the target obstacle in a target coordinate system corresponding to the target obstacle, and the position information of the aiming point corresponding to the target obstacle can be 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 can be converted into the coordinates (x0, y0) in the xoy plane in the world coordinate system, and the coordinates (S1, 0) of the aiming point corresponding to the target obstacle in the target coordinate system corresponding to the target obstacle can be converted into the coordinates (x1, y1) in the xoy plane in the world coordinate system. Then, the fitting of the circular arc trajectory is performed according to (x0, y0) and (x1, y1), and the fitted circular arc trajectory is obtained as the first motion trajectory of the target obstacle.

[0074] In this embodiment, the fitting of the circular arc trajectory 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 in a polynomial fitting manner (for example, quadratic polynomial fitting), and the first motion trajectory is obtained.

[0075] Alternatively, the position information of the target obstacle can be taken as the starting point coordinates of the first motion trajectory, and the position information of the aiming point corresponding to the target obstacle can be taken as the end point coordinates of the first motion trajectory, and the fitting of the circular arc trajectory is performed to obtain the first motion trajectory.

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

[0077] S230, fitting a straight line trajectory 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 taken as the starting point coordinates of the second motion trajectory, and a straight line trajectory of a specified length is fitted according to the extension direction of the lane where the ego vehicle is located, as the second motion trajectory of the target obstacle. The specified length can be, for example, 10m or 20m, etc.

[0079] In this embodiment, the target obstacle intends to belong to the lane cut-in intention type, 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, the target obstacle will proceed 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 can be the coordinates of the aiming point corresponding to the target obstacle in a target coordinate system corresponding to the target obstacle. Correspondingly, 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 the 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 can be understood that when there are multiple target obstacles, the second motion trajectory of each target obstacle is determined according to the foregoing process of S230.

[0082] S240, based on the first motion trajectory and the second motion trajectory, obtaining the motion trajectory of the target obstacle.

[0083] The first motion trajectory and the second motion trajectory of the target obstacle can be connected as a trajectory, as the motion trajectory of the target obstacle.

[0084] As described above, the first motion trajectory of the target obstacle can be a circular 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, and 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 as a trajectory with the aiming point of the target obstacle as the connection point, to obtain the motion trajectory of the target obstacle.

[0085] It can be understood that when there are multiple target obstacles, the motion trajectory of each target obstacle is determined according to the foregoing process of S240.

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

[0087] In an embodiment, S140 further includes: if the target obstacle intention belongs to the lane-following intention type or the lane-crossing intention type, performing straight line trajectory fitting according to the position information and the speed of the target obstacle to obtain the motion trajectory of the target obstacle.

[0088] As mentioned in the aforementioned obstacle intention, both the reverse driving along the lane and the straight driving along the lane belong to the driving along lane intention type, and both the left crossing lane and the right crossing lane belong to the crossing lane driving intention type.

[0089] In the case where the target obstacle intention belongs to the driving along lane intention type or the crossing lane driving intention type, it is determined that the motion process of the target obstacle will not change the motion direction, and thus the motion direction of the target obstacle can be obtained (the direction of the speed of the target obstacle can be determined as the motion direction of the target obstacle), and then the kinematic deduction of the target obstacle (for example, determining that the target obstacle is in uniform linear motion, uniform accelerated linear motion, or uniform decelerated linear motion) is predicted according to the position information of the target obstacle and the motion state (for example, including the motion direction, the speed, and the acceleration) of the target obstacle, and linear trajectory fitting is performed to obtain the fitted linear trajectory as the motion trajectory of the target obstacle.

[0090] In the case where the target obstacle intention belongs to the driving along lane intention type or the crossing lane driving intention type, it is determined that the motion process of the target obstacle will not change the motion direction, and thus the motion direction of the target obstacle can be obtained (the direction of the speed of the target obstacle can be determined as the motion direction of the target obstacle), and then the kinematic deduction of the target obstacle (for example, determining that the target obstacle is in uniform linear motion, uniform accelerated linear motion, or uniform decelerated linear motion) is predicted according to the position information of the target obstacle and the motion state (for example, including the motion direction, the speed, and the acceleration) of the target obstacle, and linear trajectory fitting is performed to obtain the fitted linear trajectory as the motion trajectory of the target obstacle.

[0091] It can be understood that in the case where the target obstacle intention belongs to the driving along lane intention type or the crossing lane driving intention type, the target obstacle is multiple, and the linear trajectory fitting of the target obstacle can be performed according to the foregoing process to obtain the motion trajectory.

[0092] In the embodiment, the target obstacle intention belongs to the driving along lane intention type or the crossing lane driving intention type, and it is determined that the target obstacle maintains the original motion state, and thus the linear motion deduction of the motion of the target obstacle is directly performed 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 at the future time, thereby realizing the accurate trajectory prediction of the target obstacle.

[0093] In an embodiment, after S140, the method can 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 the embodiment, the motion trajectory of the target obstacle can be the linear trajectory determined in the foregoing embodiment, or the curve trajectory combining the circular arc and the straight line determined in the foregoing embodiment.

[0095] A point can be sampled from the motion trajectory of the target obstacle at an equal time interval (for example, every 0.1 s) as a sampling point, so as to obtain a plurality of sampling points sampled from the motion trajectory of the target obstacle. Then, a plurality of curve trajectory fittings are performed based on the plurality of sampling points, so as to obtain a fitted curve trajectory as the adjusted motion trajectory of the target obstacle. The fitting manner for fitting the plurality of 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, a plurality of sampling points are sampled from the motion trajectory of the target obstacle, and then curve fitting is performed based on the plurality of sampling points, so as to realize the smoothing processing of the motion trajectory of the target obstacle. The accuracy of the adjusted motion trajectory after the smoothing processing is relatively high, and the accuracy of the trajectory prediction of the target obstacle is further improved.

[0097] In an embodiment, after S140, the method further includes: sampling a plurality of sampling points from the motion trajectory; obtaining the speed and acceleration of the target obstacle at the plurality of sampling points; and 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, determining that the motion trajectory is abnormal, and deleting the motion trajectory.

[0098] After the plurality of sampling points are determined according to the method in the foregoing embodiments, the speed and acceleration at the plurality of sampling points can be obtained. For each sampling point, the motion state of the target obstacle at the sampling point can be determined according to the motion state (for example, including the motion direction, the speed, and the acceleration, etc.) of the target obstacle perceived by the ego vehicle and the time difference between the sampling point and the position of the target obstacle (at the initial time of the trajectory prediction of the present application, the position of the target obstacle), 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 at the sampling point.

[0099] In another implementation, the length of the trajectory between any two adjacent sampling points can also be determined, and then the speed and acceleration between the two adjacent sampling points can be determined according to the length of the trajectory between the two adjacent sampling points and the time interval, and the speed and acceleration between the two adjacent sampling points are 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 greater than 0, and the acceleration range is not more than 5 meters per square second.

[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 determined that the motion state of the target obstacle at the sampling point is unreasonable, and thus the determined motion trajectory for the target obstacle is also unreasonable, and the motion trajectory is directly deleted.

[0102] If the speed at all sampling points conforms to the speed range and the acceleration conforms to the acceleration range, it is determined that the motion state of the target obstacle is reasonable, and thus 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, a plurality of adjusted sampling points can be sampled from the adjusted motion trajectory, the speed and acceleration of the target obstacle at the plurality of adjusted sampling points are obtained, and if the speed at at least one of the adjusted sampling points 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 determination manner of the adjusted sampling point refers to the aforementioned manner of determining the sampling point, and is not described herein.

[0104] That is, if the speed at a certain adjusted 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 state of the target obstacle at the adjusted sampling point is unreasonable, and thus the determined adjusted motion trajectory for the target obstacle is also unreasonable, and the adjusted motion trajectory is directly deleted.

[0105] If the speed at all adjusted sampling points conforms to the speed range and the acceleration conforms to the acceleration range, it is determined that the adjusted motion state of the target obstacle is reasonable, and thus the adjusted motion trajectory can be retained.

[0106] For example, as shown in FIG. 6, the ego vehicle obtains a perception map, filters out target obstacles according to the perception map, then performs intention prediction on the target obstacles, determines the intentions of the target obstacles, and then predicts the motion trajectories of the target obstacles according to the means in the foregoing embodiments, and then performs post-processing on the motion trajectories of the target obstacles (the post-processing can include the aforementioned smoothing processing and the determination of the speed and acceleration at the sampling points (or adjusted sampling points)), traverses all the target obstacles, and finally obtains the effective target obstacles retaining the motion trajectories (or the adjusted motion trajectories), and outputs the trajectories predicted by the effective target obstacles as the prediction results. Figure 4

[0107] ​In this embodiment, the motion trajectory is also smoothed, and the motion state (i.e., 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, so that the retained motion trajectory is more accurate, and the accuracy of the trajectory prediction of the target obstacle is further improved.

[0108] Referring to the accompanying Figure 5 , Figure 5 A structure block diagram of a trajectory generation device according to an embodiment of the present application is shown. The device 800 for a vehicle comprises:

[0109] A lane determination module 810 is configured to determine a target perception lane in which the target obstacle is located from at least one perception lane perceived by the ego vehicle.

[0110] An angle determination module 820 is configured to determine a relative angle difference between a heading angle of the target obstacle and a direction of a lane center line of the target perception lane.

[0111] An intention determination module 830 is configured to determine a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located; each angle interval corresponds to a respective obstacle intention.

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

[0113] Optionally, the lane determination module 810 is further configured to determine 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; and determine the target perception lane in which the target obstacle is located from the candidate perception lanes according to a lateral distance between the target obstacle and each candidate perception lane.

[0114] Optionally, the trajectory generation module 840 is further configured to, if the target obstacle intention belongs to a cut-in lane intention type, determine a target point corresponding to the target obstacle according to position information of the target obstacle and a speed of the target obstacle; perform circular arc trajectory fitting according to the position information of the target obstacle and position information of the target 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 target 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 configured to construct a target coordinate system according to a target boundary line closest to the target obstacle in the target perceived lane; determine a projection point of the target obstacle to a target coordinate axis of the target coordinate system as a key point corresponding to the target obstacle; and determine a target aiming point corresponding to the target obstacle based on position information of the key point and a speed projection of the target obstacle under each coordinate axis in the target coordinate system.

[0116] Optionally, the trajectory generation module 840 is further configured to, if the target obstacle belongs to the lane-following intention type or the lane-crossing intention type, perform linear trajectory fitting according to position information and a speed of the target obstacle to obtain a motion trajectory of the target obstacle.

[0117] Optionally, the device further comprises a post-processing module configured 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 further configured to sample a plurality of sampling points from the motion trajectory; obtain a speed and an acceleration of the target obstacle at the plurality of sampling points; and if the speed at at least one of the sampling points does not conform to a speed range or the acceleration does not conform to an acceleration range, determine that the motion trajectory is abnormal and delete the motion trajectory.

[0119] Optionally, the device further comprises an obstacle determination module configured to obtain an obstacle located outside a perceived map of the ego vehicle and within a preset range as the target obstacle; and the perceived map is determined based on driving data of the ego vehicle and environmental information around the ego vehicle.

[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device and the module described above can refer to the corresponding process in the foregoing method embodiments, which will not be described herein.

[0121] In addition, each function in each embodiment of the present application can be integrated in one processing module, or each module can exist physically independently, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0122] In addition, each function in each embodiment of the present application can be integrated in one processing module, or each module can exist physically independently, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0123] In addition, each function in each embodiment of the present application can be integrated in one processing module, or each module can be physically present alone, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module.

[0124] In another aspect, the present application also provides a computer readable storage medium, which stores program codes, and the program codes can be invoked by a processor to execute the method described in the above method embodiments.

[0125] The computer readable storage medium can be an electronic storage such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk or a ROM. Alternatively, the computer readable storage medium includes a non-transitory computer readable storage medium. The computer readable storage medium has a storage space for program codes to execute any method steps in the above methods. The program codes can be read from or written into one or more computer program products. The program codes can be compressed in a suitable form, for example.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit the same; even though the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent ones; and 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 by, The method comprises: determining a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle; determining a relative angle difference between a heading angle of the target obstacle and a lane center line direction of the target perception lane; determining a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located; each angle interval corresponds to a respective obstacle intention; if the target obstacle intention belongs to a cut-in lane intention type, determining a target point corresponding to the target obstacle according to position information and speed of the target obstacle; performing circular arc trajectory fitting according to the position information of the target obstacle and position information of the target point corresponding to the target obstacle to obtain a first motion trajectory of the target obstacle; performing straight line trajectory fitting according to position information of the target point of the target obstacle to obtain a second motion trajectory of the target obstacle; obtaining a motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory.

2. The method of claim 1, wherein, The method comprises: determining a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle; determining a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle; 3. The method of claim 1, wherein, determining a target perception lane in which a target obstacle is located from at least one perception lane perceived by a host vehicle; The method comprises: constructing a target coordinate system according to a target boundary line closest to the target obstacle in the target perception lane; determining a projection point of the target obstacle to a target coordinate axis of the target coordinate system as a key point corresponding to the target obstacle; 4. The method of claim 1, wherein, determining a target point corresponding to the target obstacle based on position information of the key point and speed projection of the target obstacle on each coordinate axis in the target coordinate system. After determining a target obstacle intention corresponding to a target angle interval in which the relative angle difference is located, the method further comprises:

5. The method according to any one of claims 1 to 4, characterized in that, if the target obstacle intention belongs to a lane following intention type or a lane crossing intention type, performing straight line trajectory fitting according to position information and speed of the target obstacle to obtain a motion trajectory of the target obstacle. After obtaining a motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory, the method comprises: sampling a plurality of sampling points from the motion trajectory; 6. The method according to any one of claims 1 to 4, characterized in that, performing curve trajectory fitting based on the plurality of sampling points to obtain an adjusted motion trajectory of the target obstacle. After obtaining a motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory, the method comprises: sampling a plurality of sampling points from the motion trajectory; obtaining speed and acceleration of the target obstacle at the plurality of sampling points; If there is at least one sampling point where the speed does not meet the speed range or the acceleration does not meet the acceleration range, it is determined that the motion trajectory is abnormal, and the motion trajectory is deleted.

7. The method of claim 1, wherein, Before determining the target perception lane where the target obstacle is located from the at least one perception lane perceived by the ego vehicle, the method further comprises: Obtaining an obstacle located outside the perception map of the ego vehicle and within a preset range as a target obstacle; the perception map is determined based on driving data of the ego vehicle and environmental information around the ego vehicle.

8. A trajectory generation device characterized by comprising: The device comprises: a lane determination module configured to determine a target perception lane where a target obstacle is located from at least one perception lane perceived by an ego vehicle; an angle determination module configured to determine a relative angle difference between a heading angle of the target obstacle and a direction of a lane center line of the target perception lane; an intention determination module configured to determine a target obstacle intention corresponding to a target angle interval where the relative angle difference is located; each angle interval corresponds to a respective obstacle intention; a trajectory generation module configured to, if the target obstacle intention belongs to a lane-cutting intention type, determine a target point corresponding to the target obstacle according to position information of the target obstacle and a speed; perform circular arc trajectory fitting according to the position information of the target obstacle and position information of the target point corresponding to the target obstacle to obtain a first motion trajectory of the target obstacle; perform straight line trajectory fitting according to position information of the target point of the target obstacle to obtain a second motion trajectory of the target obstacle; and obtain a motion trajectory of the target obstacle based on the first motion trajectory and the second motion trajectory.

9. A vehicle characterized by comprising: comprise: one or more processors; a memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores processor-executable program code, and the program code, when executed by the processor, causes the processor to perform the method of any one of claims 1-7.

Images