Vehicle path prediction method, control system, storage medium and electronic equipment

Abstract

The invention discloses a vehicle path prediction method, a control system, a storage medium and electronic equipment, and the method comprises the steps: obtaining the position information of an own vehicle, determining the vehicle information and lane information of a target vehicle, carrying out the coordinate conversion and fitting according to the vehicle information, and generating a historical driving track of the target vehicle, generating a predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information; generating a plurality of planned trajectories of the target vehicle according to the lane information, and selecting the planned trajectory having the highest similarity with the predicted trajectory from the planned trajectories as a predicted path of the target vehicle; the method comprises the following steps: performing fitting according to vehicle information of a target vehicle to generate a historical driving track, performing prediction to generate a prediction track, fitting a plurality of planned tracks of the target vehicle according to lane information, selecting the planned track with the highest similarity with the prediction track as a prediction path, and fully considering lane change and the historical driving track of the target vehicle. And the accuracy and timeliness of vehicle path prediction are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a vehicle path prediction method, control system, storage medium and electronic equipment. Background Art

[0002] With the widespread application of intelligent driver assistance features, drivers can choose to cruise within a single lane or actively change lanes to respond to slower vehicles ahead, based on lane markings and the target vehicle ahead, to reduce driver fatigue. However, determining the intention of the vehicle ahead to cut in or out is difficult. Existing methods rely on vehicle motion information. Using the vehicle's motion information and the distance from the lane marking, the time of the cut-in or cut-out is calculated. A cut-in or cut-out is considered to occur when a certain threshold is met. Alternatively, the vehicle's motion information is used to predict its trajectory over a certain period of time, and the predicted trajectory is used to determine the cut-in or cut-out. However, in practice, vehicle motion information is distorted, with the predicted trajectory becoming more inaccurate as the prediction time increases. Furthermore, these methods ignore the driver's driving intention, resulting in reduced judgment accuracy and an inability to make timely and accurate decisions. This can lead to late or abrupt braking, ineffective lane changes, and even the risk of accidents caused by failure to brake in time. This reduces the practicality of the feature and poses safety risks. Summary of the Invention

[0003] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a vehicle path prediction method, control system, storage medium and electronic device.

[0004] In order to achieve the above object, the present invention adopts the following technical solution: a vehicle path prediction method, comprising:

[0005] Obtain the position information of the ego vehicle to determine the vehicle information and lane information of the target vehicle;

[0006] generating a historical driving trajectory of the target vehicle according to the vehicle information, and generating a predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information;

[0007] A plurality of planned trajectories of the target vehicle are generated according to the lane information, and a planned trajectory having the highest similarity to the predicted trajectory is selected as the predicted path of the target vehicle.

[0008] As a further description of the above technical solution: the position information of the ego vehicle is used to establish a coordinate system, and the vehicle information and lane information of the target vehicle are determined based on the coordinate system. The vehicle information includes the vehicle's position information, speed information, yaw angular velocity information and timestamp information, and the lane information includes the position information of the lane centerline.

[0009] As a further description of the above technical solution: the method for generating the historical driving trajectory of the target vehicle according to the vehicle information includes:

[0010] Acquiring some position information and speed information of the target vehicle within a preset first time interval according to the timestamp information;

[0011] The position information is transformed into coordinates through a transformation matrix to obtain several position matrices of the target vehicle;

[0012] Linear fitting is performed based on the plurality of position matrices to generate a historical driving trajectory.

[0013] As a further description of the above technical solution: a method for generating a predicted trajectory of a target vehicle based on the historical driving trajectory and the vehicle information includes:

[0014] Polynomial fitting is performed based on the plurality of position matrices and speed information to generate the predicted trajectory of the target vehicle in the second time interval.

[0015] As a further description of the above technical solution: a method for generating multiple planned trajectories of a target vehicle based on the lane information includes:

[0016] According to the current lane of the target vehicle, selecting the current lane or an adjacent lane as the target lane;

[0017] According to the preset third time interval, polynomial fitting is performed between the current lane and the target lane according to the preset time interval to generate multiple planned trajectories.

[0018] As a further description of the above technical solution: the third time interval is the time when the target vehicle reaches the center line of the target lane.

[0019] As a further description of the above technical solution: the method of selecting the planned trajectory with the highest similarity to the predicted trajectory includes:

[0020] Acquiring data parameters of the predicted trajectory and each of the planned trajectories, wherein the data parameters include lateral distance, lateral velocity, lateral acceleration, curvature, and heading angle;

[0021] Calculating a first cost function of the planned trajectory based on the data parameters, and

[0022] Calculating a second cost function for similarity between the predicted trajectory and each of the planned trajectories;

[0023] Calculate the sum of the first cost function and each of the second cost functions, and select the minimum cost function as the planning trajectory with the highest similarity.

[0024] As a further description of the above technical solution: the first cost function and the second cost function adjust the selected minimum cost function by adjusting the weights of the data parameters.

[0025] Also included is a vehicle path prediction control system, comprising:

[0026] Acquisition module, obtaining vehicle information and lane information of the target vehicle;

[0027] A processing module, which establishes a coordinate system according to the vehicle information, performs coordinate conversion and fitting processing, and generates a historical driving trajectory;

[0028] a prediction module, generating a predicted trajectory based on the historical driving trajectory and the vehicle information, and generating a planned trajectory based on the lane information;

[0029] A calculation module calculates a minimum cost function according to the predicted trajectory and the planned trajectory.

[0030] Also included is a computer-readable storage medium storing a computer program for running the prediction method, wherein the computer program enables a computer to execute the prediction method as described in any one of the above technical solutions.

[0031] Also included is an electronic device comprising:

[0032] one or more processors; memory; and

[0033] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs include a method for executing the prediction method as described in any one of the above technical solutions.

[0034] The above technical solution has the following advantages or beneficial effects:

[0035] The historical driving trajectory of the target vehicle is fitted based on the vehicle information, and the predicted trajectory is predicted. Multiple planned trajectories of the target vehicle are fitted based on the lane information, and the planned trajectory with the highest similarity to the predicted trajectory is selected as the predicted path. The lane changes and historical driving trajectory of the target vehicle are fully considered to improve the accuracy and timeliness of vehicle path prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 Flowchart of the prediction method proposed by the present invention;

[0038] Figure 2 A flowchart of a method for generating a historical driving trajectory in the present invention;

[0039] Figure 3 A flowchart for generating multiple planned trajectories of a target vehicle in the present invention;

[0040] Figure 4 A flowchart for selecting the planned trajectory with the highest similarity to the predicted trajectory in the present invention;

[0041] Figure 5 This is a schematic diagram of the structure of the control system proposed by the present invention;

[0042] Figure 6 A schematic diagram of the historical driving trajectory and the predicted trajectory generated in the present invention;

[0043] Figure 7 A schematic diagram of the planning trajectory generated in the present invention;

[0044] Figure 8 This is a schematic diagram of the planning trajectory corresponding to the minimum cost function selected in the present invention.

[0045] Legend:

[0046] 1. Acquisition module; 2. Processing module; 3. Prediction module; 4. Calculation module. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] Existing vehicle path prediction methods use vehicle motion information and the distance from lane markings to calculate the time it takes to cut in or out. A vehicle is considered to have cut in or out if a certain threshold is met. Alternatively, they use motion information to predict a trajectory for a certain period of time and use the predicted trajectory to determine whether a vehicle has cut in or out. However, in practice, vehicle motion information is distorted, and the predicted trajectory becomes more inaccurate as the prediction time increases. Furthermore, these methods ignore the driver's driving intention, resulting in reduced accuracy and an inability to make timely and accurate judgments.

[0049] Reference Figure 1 , an embodiment of a vehicle path prediction method provided by the present invention includes:

[0050] S1: Obtain the position information of the ego vehicle to determine the vehicle information and lane information of the target vehicle;

[0051] S2: Generate a historical driving trajectory of the target vehicle based on the vehicle information, and generate a predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information;

[0052] S3: Generate multiple planned trajectories of the target vehicle based on the lane information, and select the planned trajectory with the highest similarity to the predicted trajectory as the predicted path of the target vehicle.

[0053] In this embodiment, vehicle information of the ego vehicle is obtained, and a Cartesian coordinate system is established with the ego vehicle as the origin and the ego vehicle's direction of travel as the positive direction of a coordinate axis. Specifically, the ego vehicle's x- and y-axis position information, speed information, yaw rate information, and timestamp information are recorded, and five sets of information are saved. If there are more than five sets, the latest information is stored and the oldest information is ejected. Each information storage process processes the historical information, establishes a coordinate system, and always ensures that the latest set of x- and y-axis positions and yaw angles are 0. Subsequent data is updated based on the x- and y-axis speeds and yaw rates. The position of the target vehicle relative to the ego vehicle can be represented by coordinate values in this coordinate system, providing an effective reference system for subsequently obtaining the target vehicle's accurate position information.

[0054] By obtaining the vehicle information of the target vehicle, the motion state of the target vehicle is determined, and the position information of the target vehicle is obtained. The position information may be based on different coordinate systems. In actual analysis, it is converted to a unified coordinate system for data processing and analysis. Fitting is performed according to the position information of the target vehicle. Through curve fitting algorithms such as polynomial fitting and spline curve fitting, the historical driving trajectory of the target vehicle is generated. Based on the generated historical driving trajectory, combined with the current speed information and yaw angular velocity information of the target vehicle, a predicted trajectory is generated, so that the target vehicle driving trajectory can be predicted in a short time.

[0055] The planned trajectories of multiple target vehicles are fitted according to the lane information. Based on the position information of the lane centerline, a path planning algorithm is used to generate multiple planned trajectories. The planned trajectory with the highest similarity to the predicted trajectory is selected as the predicted path, fully considering the lane change and historical driving trajectory of the target vehicle to improve the accuracy and timeliness of vehicle path prediction.

[0056] Reference Figure 2 ,The method for generating the historical driving trajectory of the target vehicle according to the vehicle information includes:

[0057] S21: Acquire some position information and speed information of the target vehicle within a preset first time interval according to the timestamp information;

[0058] S22: The position information is transformed into coordinates through a transformation matrix to obtain several position matrices of the target vehicle;

[0059] S23: Perform linear fitting based on several position matrices to generate a historical driving trajectory.

[0060] In this embodiment, based on the target vehicle's location information, several locations and speeds of the target vehicle within a first time interval are selected. By acquiring the location and speed information of the target vehicle at different times within the first time interval, the historical driving trajectory is subsequently generated. The first time interval can be set to 3 seconds, and the vehicle information of the target vehicle within 3 seconds is stored. After acquiring the target vehicle's location information, since the location information may be acquired in different sensor coordinate systems, it must be converted to an established coordinate system for unified analysis. Therefore, a coordinate transformation operation is required to unify it into a Cartesian coordinate system. The original location information is converted into a position matrix using a transformation matrix. The selection and calculation of the transformation matrix generally depends on the relationship between the two coordinate systems, including transformation parameters such as translation and rotation. Optionally, the transformation matrix is a rotation matrix. In actual calculations, the transformation matrix performs coordinate transformation based on the known parameters of the two coordinate systems. dx, dy, and dyaw are obtained by searching the current timestamp and stored historical timestamps. Each location information of the target vehicle is then calculated with the rotation matrix to obtain a transformed position matrix. These position matrices accurately describe the location of the target vehicle within the unified coordinate system.

[0061] The historical driving trajectory of the target vehicle is generated by linear fitting based on the converted position matrix, and the discrete position data is converted into a continuous trajectory curve, so as to intuitively display the historical driving trajectory of the target vehicle (refer to Figure 6 Specifically, several position matrices contain vehicle location information at different times. Within a preset first time interval, the location information is sorted in timestamp order, and the position coordinates corresponding to each moment are extracted as the historical driving trajectory points of the target vehicle. These trajectory points are then connected in sequence to form the vehicle's historical driving trajectory.

[0062] The method for generating a predicted trajectory of a target vehicle based on historical driving trajectories and vehicle information includes: performing polynomial fitting according to a plurality of position matrices and speed information to generate a predicted trajectory of the target vehicle within a second time interval.

[0063] In this embodiment, the position matrix contains the vehicle's position information at different times, and the speed information reflects the target vehicle's movement speed and directional trends at each location. A polynomial fit is performed based on the position matrix and speed information. By constructing a polynomial function, optionally using a cubic polynomial, the position information and corresponding speed information in the position matrix are organized in chronological order, ensuring that each trajectory point has corresponding speed data. Calculations based on the trajectory points in the historical trajectory yield coefficients for the cubic polynomial. These coefficients are then used as input for fitting in a selected polynomial fitting algorithm. After fitting, the corresponding timestamps are substituted into the fitted polynomial function for a second time interval, and the corresponding x and y coordinates are calculated. These coordinates constitute the predicted trajectory points. These points are then connected sequentially to generate the predicted trajectory of the vehicle within the second time interval. Specifically, the second time interval can be 0.5 seconds. Several timestamps within the second time interval are selected and substituted into the cubic polynomial function to calculate the predicted x and y positions of the target vehicle at each time stamp. By utilizing the historical driving trajectory of the target vehicle, it can reflect the movement trend of the target vehicle to a certain extent, provide more accurate prediction results, and facilitate implementation in actual systems.

[0064] Reference Figure 3 ,The method of generating multiple planning trajectories of the target vehicle according to lane information includes:

[0065] S311: Selecting the current lane or an adjacent lane as the target lane based on the current lane of the target vehicle;

[0066] S312: According to the preset third time interval, perform polynomial fitting between the current lane and the target lane at a preset time interval to generate multiple planned trajectories.

[0067] In this embodiment, the current lane information of the target vehicle is obtained, and the current lane and the adjacent lane are predicted as the target lane. According to the lane where the target vehicle is located, the path planning for keeping the current lane and cutting in and out is performed. When the ego vehicle and the target vehicle are in the same lane, the prediction for the target vehicle is to keep the current lane state or cut out of the current lane and enter the adjacent lane. When the ego vehicle and the target vehicle are not in the same lane, the prediction for the target vehicle is to keep the current lane state or cut into the lane where the ego vehicle is located (refer to Figure 7 The target lane selection can reflect the driving intention of the target vehicle. The target lane is selected based on the fact that the target vehicle may need to change lanes during driving to adapt to traffic conditions or follow navigation instructions. For example, some lanes may prohibit certain vehicles from passing during certain hours, require the vehicle to pass through a specific lane, or have a high vehicle density in adjacent lanes.

[0068] The preset third time interval determines the future time period covered by the planned trajectory. A number of time intervals are determined based on the third time interval as the time required for the target vehicle to change lanes. The number of time intervals determines the density of the generated planned trajectory. Smaller time intervals generate denser planned trajectories, which can more accurately reflect the possible driving paths of the vehicle at different times, but also increase the computational complexity. Larger time intervals generate relatively sparse trajectories, requiring less computation, but may not accurately capture subtle driving changes of the vehicle. Specifically, the third time interval is 2.5s-5s, and the time interval is 0.5s.

[0069] After determining the current and target lanes, as well as the third time interval, multiple planned trajectories are generated at intervals. Based on the target vehicle's information, the planned trajectories are fitted using a quintic polynomial in the Frenet coordinate system. Frenet coordinates are a commonly used coordinate system for vehicle trajectory planning, describing the vehicle's position and motion relative to the lane centerline. This decouples the vehicle's longitudinal and lateral motion, simplifying the trajectory planning process and making it more suitable for motion planning on roads with complex shapes and curvatures.

[0070] According to the given boundary conditions of the starting and ending points, a linear equation system can be established to solve the coefficients of the fifth-order polynomial, and the lateral distance l, lateral velocity dl / ds, and lateral acceleration d of the starting and ending points of the target vehicle in the third time interval are selected. 2 l / ds 2 These six parameters are calculated to obtain the coefficients of the fifth-order polynomial. After obtaining the coefficients of the fifth-order polynomial, time sampling is performed. Time sampling means selecting several time points at a certain time interval within a given time range, substituting them into the fifth-order polynomial to calculate the planned trajectory points, fitting and connecting them to obtain the planned trajectory, and obtaining the longitudinal distance s, lateral distance l, lateral speed dl / ds, and lateral acceleration d at the sampling time points. 2 l / ds 2 , curvature fk and heading angle fheading in the Frenet coordinate system.

[0071] The third time interval is the time it takes for the target vehicle to reach the center line of the target lane.

[0072] In this embodiment, by setting the range of the third time interval, it represents the time taken by the target vehicle to travel to the target lane. When the third time interval is short, it means that the target vehicle is changing lanes faster, the heading angle is larger, and the curvature of the planned trajectory is larger. When the third time interval is long, it means that the target vehicle is changing lanes slower, the heading angle is smaller, and the curvature of the planned trajectory is smaller.

[0073] Reference Figure 4,The methods for selecting the planned trajectory with the highest similarity to the predicted trajectory include:

[0074] S321: Acquire data parameters of the predicted trajectory and each planned trajectory, wherein the data parameters include lateral distance, lateral velocity, lateral acceleration, curvature, and heading angle;

[0075] S322: Calculating a first cost function of the planned trajectory based on the data parameters, and calculating a second cost function of the similarity between the predicted trajectory and each planned trajectory;

[0076] S323: Calculate the sum of the first cost function and each second cost function, and select the minimum cost function as the planning trajectory with the highest similarity.

[0077] In this embodiment, the data parameters of the predicted trajectory and each planned trajectory are obtained, and the first cost function of the planned trajectory is calculated. The first cost function is used to evaluate the characteristics and quality of the planned trajectory. The algorithm formula is as follows:

[0078] Cost_Part1=∑(W s ·S 2 +W1·L 2 +W t ·T 2 +W fh fheading 2 +W fk ·fk 2 )

[0079] Where T is the sampling time, S is the longitudinal distance, L is the lateral distance, fheading is the heading angle in the Frenet coordinate system, fk is the curvature in the Frenet coordinate system, and Wt, Ws, Wl, Wfh, and Wfk are the corresponding weights respectively.

[0080] The second cost function is the similarity cost function, which represents the similarity between the planned trajectory and the predicted trajectory of the target vehicle. The planned trajectory and the predicted trajectory are calculated in the established Cartesian coordinate system. The algorithm formula is as follows:

[0081] Cost_Part2=∑(W x ·(XX p ) 2 +W y ·(YY p ) 2 +W h ·(heading-heading p ) 2 +W k ·(KK p ) 2 )

[0082] Among them, X is the x-coordinate of the target vehicle corresponding to the sampling time of the planned trajectory, Y is the y-coordinate in the Cartesian coordinate system, heading is the heading angle in the Cartesian coordinate system, K is the curvature in the Cartesian coordinate system, Xp is the x-coordinate corresponding to the predicted trajectory, Yp is the y-coordinate corresponding to the predicted trajectory, heading p It is represented as the heading angle corresponding to the predicted trajectory, Kp is represented as the curvature corresponding to the predicted trajectory, and Wx, Wy, Wh, and Wk are the corresponding weights respectively.

[0083] The first cost function is added to each second cost function to obtain multiple total cost functions. The total cost function values of all planned trajectories are traversed, and the planned trajectory corresponding to the minimum cost function is selected as the trajectory with the highest similarity to the predicted trajectory (refer to Figure 8 ) and used as the predicted path for the target vehicle to travel.

[0084] The first cost function and the second cost function adjust the selected minimum cost function by adjusting the weights of the data parameters.

[0085] In this embodiment, weight coefficients are set in the first cost function and the second cost function, and are adjusted according to actual needs to balance the impact on the cost functions, so that the selection of planned trajectories is more in line with actual driving needs and scenarios. By using a machine learning algorithm, a large amount of driving scene data is learned to automatically optimize the weight values, providing a more reasonable predicted driving path when selecting the minimum cost function.

[0086] Reference Figure 5 The present application also provides an embodiment of a vehicle path prediction control system, comprising:

[0087] Acquisition module 1, obtains vehicle information and lane information of the target vehicle;

[0088] Processing module 2 establishes a coordinate system based on vehicle information, performs coordinate conversion and fitting processing, and generates a historical driving trajectory;

[0089] Prediction module 3 generates a predicted trajectory based on historical driving trajectories and vehicle information, and generates a planned trajectory based on lane information;

[0090] The calculation module 4 calculates the minimum cost function based on the predicted trajectory and the planned trajectory.

[0091] In this embodiment, acquisition module 1 acquires vehicle information and lane information of the target vehicle. Vehicle information includes the vehicle's position, speed, yaw rate, and timestamp information. Lane information includes the position of the lane centerline. Position information is used to determine the target vehicle's coordinates in space and provides real-time feedback on the target vehicle's specific position on the lane. Speed information reflects the speed of the target vehicle's movement and can be measured by a speed sensor to obtain the target vehicle's speed. Yaw rate information indicates the speed at which the vehicle rotates about its vertical axis and is acquired by a gyroscope. When the vehicle turns, the gyroscope can promptly sense and output corresponding data, helping to determine changes in the vehicle's direction of travel. Timestamp information adds a time stamp to each data record to ensure the temporal sequence and timeliness of the data. The vehicle's onboard camera captures images of the road ahead in real time, and an image recognition algorithm is used to identify lane lines, thereby determining the position of the lane centerline.

[0092] Processing module 2 receives and establishes a Cartesian coordinate system based on the position information of the vehicle itself, receives the vehicle information transmitted by the acquisition module 1, performs coordinate conversion, and converts it into a unified coordinate system to ensure the accuracy and consistency of the position information in the new coordinate system. After completing the coordinate conversion, processing module 2 uses a fitting algorithm to process the position information and generate a historical driving trajectory.

[0093] The prediction module 3 generates a predicted trajectory based on the historical driving trajectory and vehicle information generated by the processing module 2, and generates a planned trajectory through a polynomial fitting algorithm based on the lane information obtained by the acquisition module 1.

[0094] Calculation module 4 calculates the minimum cost function based on the predicted trajectory and the planned trajectory. It obtains relevant data parameters of the predicted trajectory and the planned trajectory, calculates a first cost function for the planned trajectory and a second cost function for the similarity between the predicted trajectory and each planned trajectory, adds the first and second cost functions together to obtain multiple total cost functions, and selects the planned trajectory corresponding to the minimum cost function as the predicted path for the target vehicle.

[0095] Also included is a computer-readable storage medium storing a computer program for executing the prediction method, wherein the computer program causes a computer to execute the following steps:

[0096] S1: Acquire the position information of the ego vehicle to establish a coordinate system, and determine the vehicle information and lane information of the target vehicle based on the coordinate system. The vehicle information includes the vehicle's position information, speed information, yaw rate information, and timestamp information, and the lane information includes the position information of the lane centerline.

[0097] S2: Perform coordinate transformation and fitting based on the vehicle information to generate the historical driving trajectory of the target vehicle, and generate the predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information;

[0098] S3: Generate multiple planned trajectories of the target vehicle based on the lane information, and select the planned trajectory with the highest similarity to the predicted trajectory as the predicted path of the target vehicle.

[0099] Among them, the computer-readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a computer-readable storage medium is coupled to a processor so that the processor can read information from the computer-readable storage medium and write information to the computer-readable storage medium. Of course, the computer-readable storage medium can also be an integral part of the processor. The processor and the computer-readable storage medium can be located in an application-specific integrated circuit (ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the computer-readable storage medium can also exist in a communication device as discrete components.

[0100] Specifically, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0101] Also included is an electronic device comprising:

[0102] one or more processors; memory; and

[0103] One or more programs, wherein the one or more programs are stored in a memory and configured to be executed by one or more processors, the programs comprising steps for performing the following steps:

[0104] S1: Acquire the position information of the ego vehicle to establish a coordinate system, and determine the vehicle information and lane information of the target vehicle based on the coordinate system. The vehicle information includes the vehicle's position information, speed information, yaw rate information, and timestamp information, and the lane information includes the position information of the lane centerline.

[0105] S2: Perform coordinate transformation and fitting based on the vehicle information to generate the historical driving trajectory of the target vehicle, and generate the predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information;

[0106] S3: Generate multiple planned trajectories of the target vehicle based on the lane information, and select the planned trajectory with the highest similarity to the predicted trajectory as the predicted path of the target vehicle.

[0107] Memory is used to store computer programs. This memory may include high-speed random access memory (RAM) and non-volatile memory (NVM), such as at least one disk storage device. It can also be a USB flash drive, a mobile hard drive, a read-only memory, a magnetic disk, or an optical disk.

[0108] A processor is used to execute a computer program stored in a memory. The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), or application-specific integrated circuits (ASICs). A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented by a hardware processor or implemented by a combination of hardware and software modules in the processor.

[0109] Optionally, the memory can be independent or integrated with the processor.

[0110] When the memory is a device independent of the processor, the electronic device may further include a bus. The bus is used to connect the memory and the processor. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc.

[0111] It should be noted that, through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment. In this article, relational terms such as first and second are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Without further constraints, an element defined by the phrase "comprises a..." does not preclude the existence of additional identical elements in the process, method, article or apparatus that includes the element.

[0112] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A vehicle path prediction method, characterized in that: include: Obtain the position information of the ego vehicle to determine the vehicle information and lane information of the target vehicle; generating a historical driving trajectory of the target vehicle according to the vehicle information, and generating a predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information; A plurality of planned trajectories of the target vehicle are generated according to the lane information, and a planned trajectory having the highest similarity to the predicted trajectory is selected as the predicted path of the target vehicle.

2. The prediction method according to claim 1, wherein: The position information of the ego vehicle is used to establish a coordinate system, and the vehicle information and lane information of the target vehicle are determined based on the coordinate system. The vehicle information includes the vehicle's position information, speed information, yaw angular velocity information and timestamp information, and the lane information includes the position information of the lane centerline.

3. The prediction method according to claim 1, wherein: The method for generating the historical driving trajectory of the target vehicle according to the vehicle information includes: Acquiring some position information and speed information of the target vehicle within a preset first time interval according to the timestamp information; The position information is transformed into coordinates through a transformation matrix to obtain several position matrices of the target vehicle; Linear fitting is performed based on the plurality of position matrices to generate a historical driving trajectory.

4. The prediction method according to claim 3, wherein: The method for generating the predicted trajectory of the target vehicle based on the historical driving trajectory and the vehicle information includes: Polynomial fitting is performed based on the plurality of position matrices and speed information to generate the predicted trajectory of the target vehicle in the second time interval.

5. The prediction method according to claim 1, wherein: The method for generating multiple planned trajectories of a target vehicle according to the lane information includes: According to the current lane of the target vehicle, selecting the current lane or an adjacent lane as the target lane; According to the preset third time interval, polynomial fitting is performed between the current lane and the target lane according to the preset time interval to generate multiple planned trajectories.

6. The prediction method according to claim 5, characterized in that: The third time interval is the time it takes for the target vehicle to reach the center line of the target lane.

7. The prediction method according to claim 1, wherein: Methods for selecting the planned trajectory with the highest similarity to the predicted trajectory include: Acquiring data parameters of the predicted trajectory and each of the planned trajectories, wherein the data parameters include lateral distance, lateral velocity, lateral acceleration, curvature, and heading angle; Calculating a first cost function of the planned trajectory based on the data parameters, and Calculating a second cost function for similarity between the predicted trajectory and each of the planned trajectories; Calculate the sum of the first cost function and each of the second cost functions, and select the minimum cost function as the planning trajectory with the highest similarity.

8. The prediction method according to claim 7, characterized in that: The first cost function and the second cost function adjust the selected minimum cost function by adjusting the weights of the data parameters.

9. A vehicle path prediction control system, characterized in that: include: Acquisition module, obtaining vehicle information and lane information of the target vehicle; A processing module, which establishes a coordinate system according to the vehicle information, performs coordinate conversion and fitting processing, and generates a historical driving trajectory; a prediction module, generating a predicted trajectory based on the historical driving trajectory and the vehicle information, and generating a planned trajectory based on the lane information; A calculation module calculates a minimum cost function according to the predicted trajectory and the planned trajectory.

10. A computer-readable storage medium, characterized in that The computer stores a computer program for running the prediction method, wherein the computer program enables a computer to execute the prediction method according to any one of claims 1 to 8.

11. An electronic device, characterized in that: include: one or more processors; Memory; as well as One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising instructions for executing the prediction method according to any one of claims 1 to 8.

Images