A trajectory screening processing method and device based on a kinematic model, an electronic device, and a storage medium

Abstract

The application provides a trajectory screening processing method and device based on a kinematic model, electronic equipment and a storage medium, comprising: obtaining driving data of a target object within a preset range of a current vehicle, and determining a plurality of candidate predicted trajectories of the target object based on the driving data and a prediction algorithm; obtaining scene information and type information of the target object, and determining a priori score of the prediction algorithm based on the scene information and the type information; evaluating each candidate predicted trajectory, obtaining a trajectory score of each candidate predicted trajectory, determining a final score of each candidate predicted trajectory based on the trajectory score and the priori score, and outputting a candidate predicted trajectory with the highest score as a predicted trajectory of the target object. It can screen and process a plurality of predicted trajectories output by a plurality of prediction algorithms, output an optimal predicted trajectory of the target object, and thus improve the accuracy of the predicted trajectory.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle detection, in particular to a trajectory screening processing method and device based on a kinematic model, an electronic device and a storage medium. BACKGROUND

[0002] An autonomous vehicle is a motor vehicle that is operated using artificial intelligence, visual computing, radar, monitoring devices, and a global positioning system to cooperate, so that the computer can automatically and safely operate the motor vehicle without any human initiative. In the entire autonomous driving software system, the importance of the prediction system is self-evident. An excellent prediction system can accurately reflect the future running trajectory of the surrounding target and judge the interaction with the future driving trajectory of the ego vehicle, which is an important input to determine the future behavior mode and speed control of the ego vehicle.

[0003] In order to further improve the accuracy of the prediction result and the comprehensiveness of the prediction system, there may be multiple prediction algorithms in the prediction system, such as combining rule algorithms and deep learning algorithms, so that they can play the advantages of their respective algorithms in their respective good prediction scenes or target types. Therefore, the prediction system outputs multiple prediction trajectories for the same target. How to select an optimal prediction trajectory from numerous prediction trajectories is a difficulty in the prediction system. SUMMARY

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a trajectory screening processing method and device based on a kinematic model, an electronic device and a storage medium, which can screen and process several prediction trajectories output by multiple prediction algorithms, output an optimal prediction trajectory of a target object, and thus improve the accuracy of the prediction trajectory.

[0005] The present application provides a trajectory screening processing method based on a kinematic model, which comprises:

[0006] Obtaining driving data of a target object within a preset range of a current vehicle, and determining multiple candidate prediction trajectories of the target object based on the driving data and a prediction algorithm;

[0007] Obtaining scene information of the target object and type information of the target object, and determining a prior score of the prediction algorithm based on the scene information and the type information;

[0008] Evaluating each candidate prediction trajectory, obtaining a trajectory score of each candidate prediction trajectory, determining a final score of each candidate prediction trajectory based on the trajectory score and the prior score, and outputting a candidate prediction trajectory with the highest score as a prediction trajectory of the target object.

[0009] In an example of the trajectory screening method based on the kinematic model, the prediction algorithm has multiple types, and each type of the prediction algorithm outputs one or more candidate prediction trajectories based on the driving data of the target object.

[0010] In an example of the trajectory screening method based on the kinematic model, the target object is assigned with a target ID when the driving data of the target object is acquired.

[0011] The candidate prediction trajectory of the target object carries the target ID information of the target object when the prediction algorithm outputs the candidate prediction trajectory of the target object.

[0012] The candidate prediction trajectories with the same target ID information are summarized.

[0013] In an example of the trajectory screening method based on the kinematic model, multiple prior scores of a prediction algorithm, and the scenario information and type information corresponding to each prior score are preset.

[0014] The scenario information and the type information of the target object are matched with the preset scenario information and type information, so as to obtain the prior score of the prediction algorithm applied to the target object in the current scenario.

[0015] In an example of the trajectory screening method based on the kinematic model, the evaluation of each candidate prediction trajectory includes:

[0016] A plurality of evaluation items and trajectory score initial values are preset.

[0017] Each evaluation item is assigned with a score.

[0018] When an evaluation item of a candidate prediction trajectory exceeds a preset threshold, the score of the evaluation item is deducted from the trajectory score initial value.

[0019] When another evaluation item of the candidate prediction trajectory is within a preset threshold range, the score of the evaluation item is not deducted from the trajectory score initial value, and finally the trajectory score of the candidate prediction trajectory is obtained.

[0020] In an example of the trajectory screening method based on the kinematic model, when each candidate prediction trajectory is evaluated, the evaluation items include:

[0021] The deviation value a of the average lateral acceleration of the candidate prediction trajectory and the first reference value;

[0022] The deviation value b of the difference between the average speed of the candidate prediction trajectory and the average speed of all non-zero speed targets and the second reference value.

[0023] When the deviation value a or the deviation value b of one of the candidate prediction trajectories is greater than a preset deviation value, the score of the evaluation item is deducted from the initial trajectory score, thereby obtaining the trajectory score of the candidate prediction trajectory.

[0024] In an example of the trajectory screening method based on the kinematic model, the scoring formula of the candidate prediction trajectory is as follows:

[0025] Score i =(Score initial -Offset)*P i

[0026] wherein Score i is the final score, Score initial is the initial trajectory score, Offset is the sum of the scores of all the evaluation items to be deducted, and P i is the prior score.

[0027] The application further provides a trajectory screening and post-processing device based on a kinematic model, comprising:

[0028] a driving data acquisition module configured to acquire driving data of a target object within a preset range of a current vehicle;

[0029] a prediction algorithm module configured to determine a plurality of candidate prediction trajectories of the target object based on the driving data;

[0030] a prior score module configured to acquire scene information and type information of the target object, and determine a prior score of the prediction algorithm based on the scene information and the type information;

[0031] a trajectory scoring module configured to evaluate each of the candidate prediction trajectories and acquire a trajectory score of each of the candidate prediction trajectories;

[0032] a trajectory output module configured to determine a final score of each of the candidate prediction trajectories based on the trajectory score and the prior score, and output a candidate prediction trajectory with the highest score as a prediction trajectory of the target object.

[0033] The application further provides an electronic device, comprising:

[0034] one or more processors;

[0035] a storage device configured to store one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the trajectory screening and post-processing method based on a kinematic model.

[0036] The application further provides a computer readable storage medium, which stores computer readable instructions, and the computer readable instructions, when executed by a processor of a computer, enable the computer to perform the trajectory screening and post-processing method based on the kinematic model.

[0037] The trajectory screening processing method based on the kinematic model outputs multiple candidate predicted trajectories of a target object by using multiple prediction algorithms, determines a priori score of the prediction algorithms according to scene information and type information of the target object, obtains trajectory score by averaging the candidate predicted trajectories, obtains final score of each candidate predicted trajectory by using the priori score and the trajectory score, and outputs a candidate predicted trajectory with the highest score as the predicted trajectory of the target object, so as to ensure that the output predicted trajectory has the highest confidence, and further improve the accuracy and reliability of the predicted trajectory.

[0038] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0039] The drawings incorporated into the specification and forming a part thereof, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application. It is clear that the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those of ordinary skill in the art without creative labor. In the drawings:

[0040] Figure 1 is a step flow chart of the trajectory screening processing method based on the kinematic model in an exemplary embodiment of the application;

[0041] Figure 2 is a step flow chart of obtaining driving data of a target object in an exemplary embodiment of the application;

[0042] Figure 3 is a step flow chart of determining priori score of a prediction algorithm in an exemplary embodiment of the application;

[0043] Figure 4 is a step flow chart of evaluating candidate predicted trajectories in an exemplary embodiment of the application;

[0044] Figure 5 is a schematic diagram of a trajectory screening processing device based on the kinematic model in an exemplary embodiment of the application;

[0045] Figure 6 is a structural schematic diagram of a computer system of an electronic device for implementing embodiments of the application. DETAILED DESCRIPTION

[0046] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0047] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0048] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0049] Before introducing the technical solutions provided in the embodiments of the present invention, the application scenarios of the embodiments of the present invention will be introduced by way of example. Of course, the technical solutions provided in the embodiments of the present invention can also be applied to other scenarios, and the embodiments of the present invention do not limit the application scenarios of the technical solutions. Specifically, the application scenarios of this embodiment include: In complex traffic environments, for a vehicle to drive quickly and safely, it is not only necessary to plan its own route well, but also to dynamically plan the next driving operation based on changes in surrounding targets. How to accurately predict the future driving trajectory of surrounding targets becomes particularly important.

[0050] Autonomous vehicles need to accurately plan their next actions based on the future trajectories of surrounding targets. Therefore, autonomous vehicles are equipped with various sensors, such as visual cameras, millimeter-wave radar, and lidar. Through these sensor devices, autonomous vehicles can continuously and comprehensively detect surrounding targets and the environment, thereby acquiring information about the targets and environment around the autonomous vehicle. This information is then input into a prediction algorithm to output a predicted trajectory.

[0051] like Figure 1 As shown, Figure 1 A flowchart illustrating the steps of a trajectory filtering processing method based on a kinematic model in an exemplary embodiment of the present invention.

[0052] This invention provides a trajectory filtering and processing method based on a kinematic model, the method comprising the following steps:

[0053] Step S101: Obtain the driving data of the target object within a preset range of the current vehicle, and determine multiple candidate predicted trajectories of the target object based on the driving data and the prediction algorithm.

[0054] In embodiments of the present invention, there are multiple prediction algorithms. Based on the driving data of the target object, each prediction algorithm outputs one or more candidate prediction trajectories. After outputting the candidate prediction trajectories, it is necessary to correct them, including: removing obviously unreasonable trajectory points and correcting deviated trajectory points.

[0055] like Figure 2 As shown, Figure 2 This is a flowchart illustrating the steps of obtaining driving data of a target object in an exemplary embodiment of the present invention.

[0056] When acquiring the driving data of the current vehicle within a preset range of a target object, it is necessary to collect and organize the driving data of the target object. The steps include:

[0057] Step S201: When acquiring the driving data of the target object, assign a target ID to the target object.

[0058] In step S202, when the prediction algorithm outputs the candidate predicted trajectory of the target object, the candidate predicted trajectory carries the target ID information of the target object.

[0059] Step S203: Summarize and organize the candidate predicted trajectories that have the same target ID information.

[0060] like Figure 1 As shown, in step S102, the scene information of the target object and the type information of the target object are obtained, and the prior score of the prediction algorithm is determined based on the scene information and the type information.

[0061] like Figure 3 As shown, Figure 3 This is a flowchart illustrating the steps of determining the prior score of the prediction algorithm in an exemplary embodiment of the present invention.

[0062] The steps for determining the prior score of the prediction algorithm include:

[0063] Step S301: Preset multiple prior scores of a prediction algorithm, as well as scene information and type information corresponding to each prior score.

[0064] Step S302, match the acquired scene information and type information of the target object with the preset scene information and type information, so as to obtain the prior score of the target object applying the prediction algorithm in the current scene.

[0065] Since the prediction ability of each prediction algorithm for different target types is different in different scenes, the prior score is a score of the prediction ability of a prediction algorithm for the current scene and a determined target type. Therefore, in the method of the present application, a plurality of prior scores are preset for each prediction algorithm, and the scene information and type information corresponding to each prior score are matched with the scene information and type information in the database, so as to obtain the prior score of the target object applying the prediction algorithm in the current scene.

[0066] The scene information can include expressway, urban highway, rural highway, etc., and can be classified as smooth, congested and general, etc. The type information can include heavy truck, bus, SUV, car, etc., and can also be classified in more detail. Of course, other scene information and type information can also be included, which will not be described here.

[0067] As shown in FIG. 1, Figure 1 Step S103, evaluate each candidate prediction trajectory to obtain the trajectory score of each candidate prediction trajectory.

[0068] As shown in FIG. 1, Figure 4 Figure 4 is a step flow chart for evaluating candidate prediction trajectories in an exemplary embodiment of the present application.

[0069] Evaluating each candidate prediction trajectory includes:

[0070] Step S401, preset a plurality of evaluation items and initial values of trajectory scores.

[0071] Step S402, assign a score to each evaluation item.

[0072] Step S403, when an evaluation item of a candidate prediction trajectory exceeds a preset threshold, deduct the score of the evaluation item from the initial value of the trajectory score.

[0073] Step S404, when another evaluation item of the candidate prediction trajectory is within a preset threshold range, do not deduct the score of the evaluation item from the initial value of the trajectory score.

[0074] Step S405, finally obtain the trajectory score of the candidate prediction trajectory.

[0075] ​In an embodiment of the present application, two evaluation items are preset, each evaluation item is assigned a score of 10 points, the initial value of the trajectory score is set to 100 points, when an evaluation item of a candidate prediction trajectory exceeds the preset threshold, the score of 10 points of the evaluation item is deducted from the initial value of the trajectory score of 100 points, when another evaluation item of the candidate prediction trajectory is within the preset threshold range, the score of the evaluation item is not deducted from the initial value of the trajectory score. That is, the trajectory score of the candidate trajectory is 90 points.

[0076] In another embodiment of the present application, the evaluation items include:

[0077] 1. The deviation value a of the average lateral acceleration of the candidate prediction trajectory and the first reference value.

[0078] 2. The deviation value b of the difference between the average speed of the candidate prediction trajectory and the average speed of all current speed non-zero targets and the second reference value.

[0079] When calculating the average lateral acceleration of a candidate prediction trajectory of a target object, the coordinate system needs to be converted to the coordinate system of the target object, and the lateral acceleration value in unit time can be calculated for every three adjacent trajectory points. The average lateral acceleration of the entire trajectory is obtained by averaging the calculated N-2 lateral acceleration values. For example, the preset deviation value is set to be less than or equal to 0.05 m / s 2 When the deviation value a is less than or equal to 0.05 m / s 2 , the score of the evaluation item is not deducted from the initial value of the trajectory score, and when the deviation value a is greater than 0.05 m / s 2 , the score of the evaluation item is deducted from the initial value of the trajectory score.

[0080] To calculate the average speed of all speed non-zero targets in the current scene, it is necessary to traverse the target list output after sensor fusion and filter out targets with zero speed. However, considering that the performance of the sensor is not stable in a moving scene, the target and the target with speed less than 0.5 m / s are treated as stationary in actual judgment. This threshold can also filter out some targets with small starting speed.

[0081] The above two evaluation items combine the physical information of the prediction trajectory itself and the information of the targets in the scene, and can evaluate the comprehensive performance of a trajectory to a certain extent. Some environmental information evaluation items can also be added, such as the comparison of the average curvature of the trajectory and the current scene curvature.

[0082] As shown in Figure 1 , step 104, determining the final score of each candidate prediction trajectory based on the trajectory score and the prior score.

[0083] In one embodiment of the present invention, the scoring formula for the candidate predicted trajectory is as follows:

[0084] Score i =(Score) initial -Offset)*P i

[0085] Among them, Score i For the final score, Score initial P represents the initial value for the trajectory score, Offset is the sum of the scores for all evaluation items to be deducted, and P is the initial value for the trajectory score. i For prior scoring.

[0086] The final score of a candidate predicted trajectory can be calculated using the formula described above.

[0087] Step 105: Output the candidate predicted trajectory with the highest score as the predicted trajectory of the target object.

[0088] In one embodiment of the present invention, after calculating the scores of all trajectories of a target object, the scores of all trajectories are sorted from high to low, and the candidate predicted trajectory with the highest score is output as the predicted trajectory of the target object.

[0089] like Figure 5 As shown, Figure 5 This is a schematic diagram of a trajectory filtering and processing device based on a kinematic model in an exemplary embodiment of the present invention.

[0090] Embodiments of the present invention also provide a trajectory filtering and processing device based on a kinematic model, comprising: a formal data acquisition module 510, a prediction algorithm module 520, a priori scoring module 530, a trajectory scoring module 540, and a trajectory output module 550.

[0091] The driving data acquisition module 510 is used to acquire driving data of target objects within a preset range of the current vehicle. The driving data acquisition module is connected to various sensors configured in the autonomous vehicle, and acquires driving data of target objects around the autonomous vehicle through sensor devices such as vision cameras, millimeter-wave radar, and lidar. When acquiring driving data of a target object, the driving data acquisition module 510 assigns a target ID to the target object.

[0092] The prediction algorithm module 520 is used to determine multiple candidate prediction trajectories of the target object based on the driving data. The prediction algorithm module 520 may include multiple prediction algorithms or a module including multiple prediction algorithms. When outputting the candidate prediction trajectories of the target object, the candidate prediction trajectories carry the target ID information of the target object.

[0093] The prior score module 530 is configured to acquire scene information in which the target object is located and type information of the target object, and determine a prior score of the prediction algorithm based on the scene information and the type information. A plurality of prior scores of each prediction algorithm are preset in the prior score module 530, and the scene information and the type information corresponding to each prior score are preset. The acquired scene information in which the target object is located and the type information of the target object are matched with the preset scene information and type information, so as to obtain the prior score of the prediction algorithm applied to the target object in the current scene.

[0094] The trajectory score module 540 is configured to evaluate each candidate prediction trajectory and acquire a trajectory score of each candidate prediction trajectory. The trajectory output module 550 is configured to determine a final score of each candidate prediction trajectory based on the trajectory score and the prior score, and output a candidate prediction trajectory with the highest score as the prediction trajectory of the target object.

[0095] Embodiments of the present application also provide an electronic device, including: one or more processors; a storage device configured to store one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the control method of scene and device linkage based on vehicle-mounted intelligent hardware provided in each of the above embodiments.

[0096] Figure 6 A structural schematic diagram of a computer system of an electronic device suitable for implementing embodiments of the present application is shown. It should be noted that, Figure 6 The computer system 600 of the electronic device shown is only an example, and should not impose any limitation on the functions and use range of embodiments of the present application.

[0097] As Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 602 or programs loaded from a storage portion 608 to a random access memory (RAM) 603, such as performing the methods described in the above embodiments. In the RAM 603, various programs and data required for system operation are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0098] The following components are connected to the I / O interface 605: an input part 606 including a keyboard, a mouse, etc.; an output part 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), etc., and a speaker, etc.; a storage part 608 including a hard disk, etc.; and a communication part 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication part 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as necessary. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 610 as necessary, so that a computer program read out therefrom is installed in the storage part 608 as necessary.

[0099] In particular, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing a computer program for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, various functions defined in the system of the present application are executed.

[0100] It should be noted that the computer readable medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of computer readable storage media can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable signal medium can include a data signal propagating in the baseband or as a carrier wave part of a signal propagating in the carrier wave, in which the computer readable computer program is carried. Such a propagating data signal can take various forms, including but not limited to electromagnetic signals, optical signals or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit programs for use by or in conjunction with an instruction execution system, device or apparatus. The computer program contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination of the above.

[0101] The trajectory screening processing method based on the kinematic model utilizes a plurality of prediction algorithms to output a plurality of candidate prediction trajectories for a target object, determines a priori score of the prediction algorithm through scene information and type information of the target object, obtains trajectory score through the average candidate prediction trajectory, obtains final score of each candidate prediction trajectory through the priori score and the trajectory score, and outputs a candidate prediction trajectory with the highest score as the prediction trajectory of the target object, thereby ensuring that the output prediction trajectory has the highest confidence, and further improving the accuracy and reliability of the prediction trajectory.

[0102] The flow and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flow and block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions ("instructions"). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may be executed in the reverse order, depending on the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0103] The units described in the embodiments of the present application can be implemented by software, or by hardware, or by a combination of software and hardware. The units described can also be implemented in a processor. In some cases, the names of the units do not constitute a limitation on the units themselves.

[0104] The above-described embodiments are merely illustrative of the principles of the present application and the effect thereof, and are not intended to limit the present application. Any person skilled in the art can make modifications or changes to the above-described embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical ideas of the present application shall be covered by the claims of the present application.

Claims

1. A kinematic model based trajectory screening process method, characterized in that, The method comprises: obtaining driving data of a target object within a preset range of a current vehicle, and determining a plurality of candidate predicted trajectories of the target object based on the driving data and a prediction algorithm, wherein the prediction algorithm has a plurality of types, and each type of the prediction algorithm outputs one or more candidate predicted trajectories based on the driving data of the target object; obtaining scene information in which the target object is located and type information of the target object, and determining a priori score of the prediction algorithm based on the scene information and the type information; evaluating each of the candidate predicted trajectories to obtain a trajectory score of each of the candidate predicted trajectories, determining a final score of each of the candidate predicted trajectories based on the trajectory score and the priori score, and outputting a candidate predicted trajectory with the highest score as a predicted trajectory of the target object; a score formula of the candidate predicted trajectory is as follows: Score i = (Score initial - Offset) * P i where Score i is the final score, Score initial is the initial value of the trajectory score, Offset is the sum of the scores of all evaluation items to be deducted, and P i is the prior score.

2. The trajectory screening process based on a kinematic model according to claim 1, wherein, when obtaining the driving data of the target object, a target ID is assigned to the target object; when the prediction algorithm outputs the candidate predicted trajectory of the target object, the candidate predicted trajectory carries target ID information of the target object; the candidate predicted trajectories with the same target ID information are summarized and arranged.

3. The trajectory screening process based on kinematic model according to claim 1, wherein, a plurality of priori scores of a prediction algorithm are preset, and scene information and type information corresponding to each of the priori scores are preset; the obtained scene information in which the target object is located and the type information of the target object are matched with the preset scene information and the preset type information, so that the priori score of the prediction algorithm applied to the target object in the current scene is obtained.

4. The trajectory screening process based on kinematic model according to claim 1, wherein, evaluating each of the candidate predicted trajectories comprises: a plurality of evaluation items and a trajectory score initial value are preset; a score value is assigned to each of the evaluation items; when an evaluation item of a candidate predicted trajectory exceeds a preset threshold value, the score value of the evaluation item is deducted from the trajectory score initial value; when another evaluation item of the candidate predicted trajectory is within a preset threshold range, the score value of the evaluation item is not deducted from the trajectory score initial value, and finally the trajectory score of the candidate predicted trajectory is obtained.

5. The trajectory screening and post-processing method based on kinematic model according to claim 4, characterized in that, when each of the candidate predicted trajectories is evaluated, the evaluation items comprise: a deviation value a of an average lateral acceleration of the candidate predicted trajectory and a first reference value; a deviation value b of a difference between an average speed of the candidate predicted trajectory and an average speed of all non-zero speed targets and a second reference value; when the deviation value a or the deviation value b of a candidate predicted trajectory is greater than a preset deviation value, the score value of the evaluation item is deducted from the trajectory score initial value, so that the trajectory score of the candidate predicted trajectory is obtained.

6. A trajectory screening and post-processing device based on a kinematic model, characterized in that The method comprises: a driving data obtaining module, configured to obtain driving data of a target object within a preset range of a current vehicle; a prediction algorithm module, configured to determine a plurality of candidate predicted trajectories of the target object based on the driving data; An a priori scoring module is configured to acquire scene information of the target object and type information of the target object, and determine an a priori score of the prediction algorithm based on the scene information and the type information. The prediction algorithm has multiple types, and each type of the prediction algorithm outputs one or more candidate prediction trajectories based on driving data of the target object. A trajectory scoring module is configured to evaluate each of the candidate prediction trajectories and acquire a trajectory score of each of the candidate prediction trajectories. The trajectory output module is configured to determine a final score of each candidate predicted trajectory based on the trajectory score and the prior score, and output a candidate predicted trajectory with the highest score as the predicted trajectory of the target object; the score formula of the candidate predicted trajectory is as follows: Score i = (Score initial - Offset) * P i , wherein Score i is the final score, Score initial is the initial value of the trajectory score, Offset is the sum of the scores of all evaluation items to be deducted, and P i is the prior score.

7. An electronic device, comprising: The method comprises: one or more processors; a storage device configured to store one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the trajectory screening and post-processing method based on the kinematic model according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A computer readable instruction is stored thereon, which, when executed by a processor of a computer, causes the computer to execute the trajectory screening and post-processing method based on the kinematic model according to any one of claims 1 to 5.

Images