Autonomous driving car testing method based on collaborative round-up of multiple traffic participants

Abstract

The present invention relates to a method for testing autonomous vehicles based on collaborative round-up and confrontation among multiple traffic participants. The method comprises: dividing a test scenario into multiple spatiotemporal resource grids based on scenario information, constructing a conflict event set based on the trajectories of the tested vehicle and traffic participants; defining a conflict degree index, a safety index, and a travel time; constructing a multi-objective speed optimization search model based on the conflict degree index, safety index, and travel time, utilizing multiple traffic participants to collaboratively round up the tested vehicle, solving the multi-objective speed optimization search model during the collaborative round-up process, and obtaining a spatiotemporal resource occupancy table; performing collision detection, and if there is no conflict, generating the optimal trajectory for all traffic participants; otherwise, returning to perform collaborative round-up; and obtaining comprehensive indicators of the tested vehicle when the traffic participants execute the optimal trajectory for performance evaluation. Compared to existing technologies, the present invention achieves more accurate autonomous driving testing by constructing complex scenarios that take into account the collaboration between traffic participants.

Description

Technical Field

[0001] The present invention relates to the field of testing and evaluation of autonomous vehicle technology, and in particular to a method for testing autonomous vehicles based on collaborative encirclement and confrontation of multiple traffic participants. Background Art

[0002] The development of autonomous vehicles is in full swing, and has now entered the stage of large-scale testing and demonstration applications. Testing is a core technical challenge that urgently needs to be addressed for commercial application. For example, Chinese patent application CN111915888A provides a method for calculating the complexity of traffic participants in autonomous driving test scenarios. This method quantifies the complexity of traffic participants in the test scenario. While this method fills a gap in current methods for quantifying the complexity of autonomous driving test scenarios, it ignores the collaborative nature of traffic participants. Many existing scenario tests are often scripted, with preset trajectories of traffic participants that cannot be dynamically adjusted based on the driving behavior of the vehicle being tested. This makes it difficult to characterize the complex interactions between traffic participants and, therefore, cannot fully test the intelligent decision-making and response capabilities of autonomous driving systems. Furthermore, current testing methods suffer from the problem of simple scenario settings. For example, in automatic emergency braking tests and intersection collision tests, the number of traffic participants that conflict with the vehicle being tested is extremely small.

[0003] In order to meet the testing requirements of high-level autonomous driving and better evaluate the response and decision-making capabilities of autonomous driving systems, there is an urgent need for a testing method that can generate complex dynamic scenarios that take into account the coordination between traffic participants and comprehensively examine the intelligence and risk response capabilities of autonomous driving vehicles. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide an autonomous driving vehicle testing method based on the collaborative capture and confrontation of multiple traffic participants. Through the collaborative capture strategy of multiple traffic participants, complex traffic conflict scenarios are actively generated, and based on the occupancy of the spatiotemporal resource grid of the tested vehicle and the traffic participants, the concept of conflict degree is proposed. The driving stability and passing time of the tested vehicle are further comprehensively considered to quantitatively examine the comprehensive reaction ability of the tested vehicle, thereby accelerating the test.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] The present invention provides a method for testing an autonomous vehicle based on collaborative encirclement and confrontation of multiple traffic participants, the method comprising:

[0007] Predicting the driving trajectory of the tested vehicle and obtaining the driving paths of traffic participants and scenario information of the test scenario, dividing the test scenario into a plurality of spatiotemporal resource grids based on the scenario information, obtaining conflict grids between the tested vehicle and traffic participants based on the driving trajectory and driving path, and constructing a conflict event set for each tested vehicle based on the conflict grids;

[0008] Define the conflict degree index of the tested vehicle, the safety index of traffic participants and the travel time. The conflict degree index includes the occupation index of a single spatiotemporal resource grid and the conflict degree between the tested vehicle and traffic participants; the safety index includes the cumulative conflict time between all traffic participants;

[0009] Based on the conflict event set, multiple traffic participants are used to collaboratively capture the vehicle under test;

[0010] A multi-objective speed optimization search model is constructed based on the conflict degree index, safety index, and travel time. The multi-objective speed optimization search model is solved during the collaborative roundup process. The spatiotemporal resource occupancy table of the tested vehicle and traffic participants in the test scenario is obtained in combination with the spatiotemporal resource grid, that is, the timetable of the tested vehicle and traffic participants arriving at or leaving the corresponding spatiotemporal resource grid;

[0011] Perform collision detection based on the spatiotemporal resource occupancy table, and generate optimal trajectories for all traffic participants if there is no conflict; otherwise, return to perform collaborative round-up;

[0012] Obtain comprehensive indicators of the tested vehicle when the traffic participant executes the optimal trajectory, and perform performance evaluation based on the comprehensive indicators.

[0013] As a preferred technical solution, the spatiotemporal resource grid occupancy index is:

[0014] , ,

[0015] ,

[0016] ,

[0017] in, represents the occupancy time of the spatiotemporal resource grid k; represents the time of entering the spatiotemporal resource grid k; represents the time of leaving the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the mth passed space-time resource grid; represents the time when the tested vehicle i enters the first spatiotemporal resource grid that needs to be passed; m represents the number of spatiotemporal resource grids that the tested vehicle i passes through; Indicates the width of the spatiotemporal resource grid; represents the speed of the measured vehicle i; Indicates the number of resource blocks that the tested vehicle is expected to pass through; represents the time when the tested vehicle i leaves the mth space-time resource grid; Represents the length of the measured vehicle i.

[0018] As a preferred technical solution, the conflict degree between the tested vehicle and the traffic participant is:

[0019] ,

[0020] ,

[0021] in, represents the conflict time between the tested vehicle i and the traffic participant j on the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; A calculation function representing the conflict time between the tested vehicle and traffic participants; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when traffic participant j leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; represents the conflict degree, i.e., the cumulative resource block conflict time between the tested vehicle i and all traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; Represents the set of all tested vehicles and traffic participants.

[0022] As a preferred technical solution, the passage time is:

[0023] ,

[0024] in, Indicates the travel time; Indicates the time to pass the test scenario; Indicates the time when the test scenario begins.

[0025] As a preferred technical solution, the safety index is:

[0026] ,

[0027] in, represents the cumulative conflict time between traffic participant j and all other traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; represents the set of all traffic participants except traffic participant j; represents traffic participant t; represents the conflict time between traffic participants j and t on the spatiotemporal resource grid k.

[0028] As a preferred technical solution, the multi-objective speed optimization search model includes an objective function and constraints, wherein the objective function is:

[0029] ,

[0030] in, Represents the parameter information of all tested vehicles and traffic participants, and and Indicates the vehicle's location information. Indicates vehicle speed; 、 and represents weight; Indicates the conflict degree between the tested vehicle and traffic participants; Indicates the travel time; Indicates safety indicators.

[0031] As a preferred technical solution, the constraints are:

[0032] ,

[0033] ,

[0034] in, , representing vehicle p and vehicle q, and Represents the set of all tested vehicles and traffic participants; Indicates that vehicle p enters the spatiotemporal resource grid time; Indicates that vehicle q leaves the spatiotemporal resource grid time, , represents the label of the spatiotemporal resource grid; represents positive infinity; Indicates the priority. If vehicle p passes before vehicle q, then ,on the contrary .

[0035] As a preferred technical solution, the method for generating the optimal trajectory is:

[0036] Constructing an optimization constraint model for traffic participants, and solving the optimization constraint model based on the spatiotemporal resource occupancy table to obtain solution parameters; the solution parameters include: a first acceleration change rate and a second acceleration change rate;

[0037] Constructing a trigonometric function model, and solving the trigonometric function model based on the solution parameters to obtain speed values ​​in each time period;

[0038] An optimal trajectory is generated based on the speed values ​​in each time period.

[0039] As a preferred technical solution, the optimization constraint model is:

[0040] ,

[0041] ,

[0042] in, The time for the traffic participant to arrive at the target spatiotemporal resource grid obtained based on the spatiotemporal resource occupancy table; Indicates the weight ratio of maximizing speed; Indicates the weight ratio of gentle acceleration; represents the acceleration time variation function; Indicates the rate of change of speed of a traffic participant when the speed is less than or equal to the expected driving acceleration; Indicates the rate of change of speed of a traffic participant when the speed is greater than the expected driving acceleration; Indicates the minimum acceleration of traffic participants; Indicates the maximum acceleration of traffic participants; when hour, Indicates the acceleration when the speed of the traffic participant is less than or equal to the expected driving acceleration. hour, Indicates the acceleration when the speed of the traffic participant is greater than the expected driving acceleration; hour, Indicates the speed of the traffic participant when the speed is less than or equal to the expected driving acceleration stage. hour, Indicates the speed of the traffic participant when the speed is greater than the expected driving acceleration stage; Indicates the time that the traffic participant is in the jerk phase; Indicates the time that the traffic participant is in the deceleration phase.

[0043] As a preferred technical solution, the trigonometric function is:

[0044] ,

[0045] in, Indicates the expected driving speed of traffic participants; Indicates the difference between the current vehicle speed and the turning speed; Indicates the rate of change of speed of a traffic participant when the speed is less than or equal to the expected driving acceleration; Indicates time; Indicates the rate of change of speed of a traffic participant during a period when the speed is greater than the expected driving acceleration.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1) The present invention addresses the problems of existing technologies that ignore the coordination between traffic participants and that the test scenarios are relatively simple and lack conflict between the tested vehicle and traffic participants. The present invention proposes a test method based on collaborative capture and confrontation of multiple traffic participants. Taking into account that when conducting autonomous driving tests, the different driving paths of traffic participants and the tested vehicle and the different behavioral models between vehicles will affect the test results of autonomous driving, the present invention uses multiple traffic participants to collaboratively capture the tested vehicle to adjust the driving routes of traffic participants, and imposes safety constraints on traffic participants during collaborative capture to avoid conflicts between traffic participants. At the same time, it maximizes the conflicts between traffic participants and the tested vehicle as much as possible. By adjusting the driving routes of traffic participants, the test scenario is complicated, and the intelligence and risk response capabilities of autonomous driving vehicles can be tested more effectively and accurately. In the present invention, the form of collaborative capture is used to generate conflicts. Compared with traditional direct conflict tests, the present invention generates conflicts by guiding traffic participants to capture and confront the tested vehicle but avoids direct collisions, which has the effect of accelerating the test.

[0048] 2) The existing technology also has the problem that the conflict risk description dimension is single, which makes it impossible to fully reflect the performance of autonomous driving tests. To address the above problems, the present invention adopts a comprehensive description system to construct a conflict event set based on spatial resource occupancy information, temporal resource occupancy information, and spatiotemporal conflict information between the tested vehicle and traffic participants. It also defines multi-dimensional indicators including the conflict degree index of the tested vehicle, the safety index of traffic participants and the travel time, and provides a more detailed description of the conflicts between the tested vehicle and traffic participants, as well as the conflicts between traffic participants, thereby achieving a more comprehensive and detailed autonomous driving performance evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the method flow of the present invention;

[0050] Figure 2 A schematic diagram of grid division in the spatial dimension when the test scene is an intersection in an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of grid division in the time dimension when the test scenario is an intersection in an embodiment of the present invention;

[0052] Figure 4 Schematic diagram of collaborative roundup when the test scenario is an intersection in an embodiment of the present invention;

[0053] Figure 5 This is the process for generating the optimal trajectory of traffic participants of the present invention. DETAILED DESCRIPTION

[0054] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0055] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0056] In response to the problems existing in the prior art, the present application proposes a self-driving car testing method based on the collaborative capture and confrontation of multiple traffic participants. The driving paths of multiple traffic participants are adjusted based on the multi-target speed optimization search strategy of the collaborative capture, and different conflict test cases for self-driving cars are actively generated to improve the test effect of the self-driving car under test. In addition, the present invention fills the gap in the comprehensive intelligence index of multiple traffic participant vehicles in the conflict scenarios in the current autonomous driving field, proposes the concept of conflict degree of the maximum overlapping area of ​​multiple traffic participants based on the proportion of space-time resource occupancy, and further defines the intelligence of the tested vehicle in combination with the driving stability and passing time of the tested vehicle. The detailed process of the present invention is as follows: Figure 1 Shown, including:

[0057] S1. Divide the test scene into multiple spatiotemporal resource grids based on the scene information, and construct a conflict event set according to the trajectories of the tested vehicles and traffic participants.

[0058] S11. Predict the driving trajectory of the vehicle under test and obtain the driving path of the traffic participants and the scene information of the test scene. In detail, the scene information includes the scene type, the number of lanes, and the number of traffic participants.

[0059] S12. Divide the test scene into multiple spatiotemporal resource grids based on the scene information.

[0060] In order to accurately control the space-time resource occupancy of the tested vehicle and traffic participants in the test scene, the spatial resources of the test scene are divided into grids on the plane. The division results are as follows: Figure 2 As shown, squares 1 to 16 represent the spatial resource grids 1 to 16 after the test scene is divided in the spatial dimension. At the same time, each spatial resource grid is divided in the time dimension. The results after division are shown in Figure 3 , and in Figure 3 The middle rectangle "1" represents traffic participant one, the rectangle "2" represents traffic participant two, and the rectangle "3" represents traffic participant three; and in each spatiotemporal resource grid there is a time when a vehicle enters the grid and a time when it leaves the grid.

[0061] According to the spatiotemporal resource grid and the time of entering the grid and the time of leaving the grid stored therein, a spatiotemporal resource occupancy table of the tested vehicle is generated.

[0062] S13. Obtain a conflict grid between the tested vehicle and traffic participants based on the driving trajectory and the driving path, and construct a conflict event set for each tested vehicle based on the conflict grid.

[0063] Specifically, based on the driving trajectory of the tested vehicle, the grid set occupied by the tested vehicle in the spatial dimension is determined, which is recorded as Based on the driving path of traffic participant i, determine the grid set occupied by traffic participant i in the spatial dimension, ; If the set of spatiotemporal resource grids occupied by the tested vehicle is , the set of spatiotemporal resource grids occupied by traffic participant i is , solving the intersection of the two can obtain the conflict event set, namely: .

[0064] S2. Define the conflict degree index of the tested vehicle, the safety index of traffic participants and the travel time.

[0065] The conflict degree index includes the single spatiotemporal resource grid occupancy index and the conflict degree between the measured vehicle and traffic participants. The spatiotemporal resource grid occupancy index is:

[0066] , ,

[0067] ,

[0068] ,

[0069] in, represents the occupancy time of the spatiotemporal resource grid k; represents the time of entering the spatiotemporal resource grid k; represents the time of leaving the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the mth passed space-time resource grid; represents the time when the tested vehicle i enters the first spatiotemporal resource grid that needs to be passed; m represents the number of spatiotemporal resource grids that the tested vehicle i passes through; Indicates the width of the spatiotemporal resource grid; represents the speed of the measured vehicle i; Indicates the number of resource blocks that the tested vehicle is expected to pass through; represents the time when the tested vehicle i leaves the mth space-time resource grid; Represents the length of the measured vehicle i.

[0070] The conflict degree between the tested vehicle and traffic participants is:

[0071] ,

[0072] ,

[0073] in, represents the conflict time between the tested vehicle i and the traffic participant j on the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; A calculation function representing the conflict time between the tested vehicle and traffic participants; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when traffic participant j leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; represents the conflict degree, i.e., the cumulative resource block conflict time between the tested vehicle i and all traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; Represents the set of all traffic participants.

[0074] The safety index includes the cumulative conflict time between all traffic participants, and its expression is:

[0075] ,

[0076] in, represents the cumulative conflict time between traffic participant j and all other traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; represents the set of all traffic participants except traffic participant j; represents traffic participant t; represents the conflict time between traffic participants j and t on the spatiotemporal resource grid k.

[0077] The travel time is: ,in, Indicates the travel time; Indicates the time to pass the test scenario; Indicates the time when the test scenario begins.

[0078] S3. Based on the conflict event set, multiple traffic participants are used to collaboratively capture the vehicle under test.

[0079] In order to construct a complex test scenario as much as possible, the present invention uses multiple traffic participants to surround the vehicle under test, that is, by adjusting the driving paths of all traffic participants that have a conflict event set with the target vehicle under test, so that they conflict with the target vehicle under test, while no conflict occurs between traffic participants. Figure 2 and Figure 3 Taking the intersection test scenario shown in as an example, it is assumed that the vehicle under test drives along a road to the intersection and passes through spatial resource grid 8, spatial resource grid 7, spatial resource grid 11, spatial resource grid 10 and spatial resource grid 14 in sequence. At this time, traffic participant 1 passes through spatial resource grid 15, spatial resource grid 11, spatial resource grid 7 and spatial resource grid 3 in sequence, traffic participant 2 passes through spatial resource grid 2, spatial resource grid 6, spatial resource grid 10 and spatial resource grid 14 in sequence, and traffic participant 3 passes through spatial resource grid 9, spatial resource grid 10, spatial resource grid 11 and spatial resource grid 12 in sequence, realizing the coordinated capture of the vehicle under test by traffic participants 1, 2 and 1, and the capture result is shown as follows. Figure 4 As shown, and in Figure 4 The middle squares 1 to 16 and Figure 2 The squares 1 to 16 in the figure represent the spatial resource grids 1 to 16 after the test scene is divided in the spatial dimension.

[0080] More specifically, the intensity of the encirclement of the vehicle under test can be dynamically adjusted according to the test requirements. Through the collaborative encirclement of the vehicle under test by multiple traffic participants, a dynamic and complex test environment that supports the collaborative interaction of multiple traffic participants can be built. It can evolve more specifically for different driving paths and different behavior patterns between vehicles, providing a good test environment for testing the intelligence of autonomous driving and its ability to deal with dangers.

[0081] S4. Construct a multi-objective speed optimization search model, solve the multi-objective speed optimization search model during the collaborative roundup process, and obtain the spatiotemporal resource occupancy table of the tested vehicles and traffic participants in the test scenario.

[0082] S41. With the goal of minimizing the conflict degree index, safety index, and travel time, while ensuring active conflict generation, accelerating the test and minimizing conflicts between traffic participants, a multi-objective speed optimization search model is constructed, which is expressed as:

[0083] ,

[0084] ,

[0085] ,

[0086] in, Represents the parameter information of all tested vehicles and traffic participants, and and Indicates the vehicle's location information. Indicates vehicle speed; 、 and represents weight; Indicates the conflict degree between the tested vehicle and traffic participants; Indicates the travel time; Indicates safety indicators; , representing vehicle p and vehicle q, and Represents the set of all tested vehicles and traffic participants; Indicates that vehicle p enters the spatiotemporal resource grid time; Indicates that vehicle q leaves the spatiotemporal resource grid time, , represents the label of the spatiotemporal resource grid; represents positive infinity; Indicates the priority. If vehicle p passes before vehicle q, then ,on the contrary .

[0087] S42. During the collaborative roundup process, a multi-objective speed optimization search model is solved based on genetic algorithm and mixed integer programming to obtain a time-space resource occupancy table of the tested vehicles and traffic participants in the test scenario.

[0088] S421. For each coordinated roundup, the number of spatiotemporal resource grids that the traffic participants are expected to pass through is uniformly encoded as an individual. For example, in a coordinated roundup, assume that the traffic participants are numbered 1 and 2, and that traffic participants 1 and 2 need to pass through 4 and 5 resource grids, respectively. In this case, the individual codes are [1, 1, 1, 1, 2, 2, 2, 2, 2]. This code is used for mutation and crossover in the subsequent genetic algorithm.

[0089] S422. Obtain the location information, status information, and occupied space-time resource grids of all vehicles, including the measured vehicle and traffic participants, and construct the above data into two matrices, J and P, where each row of the J matrix represents the grid index and grid order passed by the vehicle, and each row of the P matrix represents the estimated time when the vehicle passes through the corresponding grid. The J and P matrices are used to decode individual codes in the population.

[0090] S423. Based on the J and P matrices and the multi-objective speed optimization search model, a genetic algorithm is used to obtain the initial solution.

[0091] S424. Based on the initial solution, a mixed integer programming method is used to further optimize the solution to obtain a timetable for traffic participants to arrive at or leave the corresponding space-time resource grid.

[0092] S425 , combining the timetable of traffic participants arriving at or leaving the corresponding space-time resource grid with the space-time resource occupancy table of the tested vehicle to obtain a total space-time resource occupancy table as input to S5 .

[0093] S5: Perform collision detection based on the spatiotemporal resource occupancy table. If there is no conflict, generate the optimal trajectory for all traffic participants and then execute S6. Otherwise, execute S3. The process of generating the optimal trajectory is as follows: Figure 5 shown.

[0094] S51. Based on the acquired status information, estimated arrival time and distance to the target space-time resource grid of the traffic participant, determine whether the traffic participant can reach the target space-time resource grid at the target arrival time corresponding to the space-time resource occupancy table. If so, execute S52; otherwise, execute S3 to change the current capture strategy and change the target space-time resource grid.

[0095] S52, build the optimization constraint model of traffic participants, and solve the optimization constraint model using the SLSQP algorithm based on the spatiotemporal resource occupancy table to obtain the solution parameters. The detailed solution parameters include: the first acceleration change rate and the second acceleration rate The solution process of the SLSQP algorithm is common knowledge to those skilled in the art and will not be described in detail in this step.

[0096] In detail, the optimization constraint model is:

[0097] ,

[0098] ,

[0099] in, The time for traffic participants to arrive at the target spatiotemporal resource grid obtained based on the spatiotemporal resource occupancy table; Indicates the weight ratio of maximizing speed; Indicates the weight ratio of gentle acceleration; represents the acceleration time variation function; Indicates the rate of change of speed of a traffic participant when the speed is less than or equal to the expected driving acceleration; Indicates the rate of change of speed of a traffic participant when the speed is greater than the expected driving acceleration; Indicates the minimum acceleration of traffic participants; Indicates the maximum acceleration of traffic participants; when hour, Indicates the acceleration when the speed of the traffic participant is less than or equal to the expected driving acceleration. hour, Indicates the acceleration when the speed of the traffic participant is greater than the expected driving acceleration; hour, Indicates the speed of the traffic participant when the speed is less than or equal to the expected driving acceleration stage. hour, Indicates the speed of the traffic participant when the speed is greater than the expected driving acceleration stage; Indicates the time that the traffic participant is in the jerk phase; Indicates the time that the traffic participant is in the deceleration phase.

[0100] S53: Construct a trigonometric function model, and solve the trigonometric function model based on the solution parameters to obtain the speed value of each time period.

[0101] The trigonometric function model is:

[0102] ,

[0103] in, Indicates the expected driving speed of traffic participants; Indicates the difference between the current vehicle speed and the turning speed; Indicates the rate of change of speed of a traffic participant when the speed is less than or equal to the expected driving acceleration; Indicates time; Indicates the rate of change of speed of a traffic participant during a period when the speed is greater than the expected driving acceleration.

[0104] Obtain the current speed of the traffic participant, the estimated time to reach the target space-time resource grid, and the distance of the intersection, and combine them with the solution parameters solved by S52. Substitute the above parameters into the trigonometric function model to obtain the speed of the traffic participant at each moment in the target period.

[0105] S54: Generate an optimal trajectory based on the speed value in each time period.

[0106] S6. Obtain comprehensive indicators of the tested vehicle when the traffic participants execute the optimal trajectory, and perform performance evaluation based on the comprehensive indicators.

[0107] Specifically, the conflict degree index and conflict completion time of the tested vehicle are calculated when the traffic participants execute the optimal path, and the comprehensive indicators such as the driving trajectory and acceleration of the tested vehicle are obtained to conduct a comprehensive evaluation of the tested vehicle.

[0108] In addition, the present invention also provides an autonomous driving vehicle testing system based on collaborative roundup confrontation of multiple traffic participants, including a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or computer program instructions loaded from a storage unit into a random access memory (RAM). Various programs and data required for device operation can also be stored in the RAM. The CPU, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0109] Many components in a device are connected to the I / O interface, including: input units, such as a keyboard and mouse; output units, such as various types of displays and speakers; storage units, such as magnetic disks and optical disks; and communication units, such as network cards, modems, and wireless communication transceivers. The communication unit allows the device to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0110] The processing unit performs the various methods and processes described above, such as methods S1 to S6. For example, in some embodiments, methods S1 to S6 may be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed onto the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S6 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S6 by any other appropriate means (e.g., by means of firmware).

[0111] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0112] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0113] In the context of the present invention, machine-readable medium can be a tangible medium that can contain or store a program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0114] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A self-driving car testing method based on collaborative round-up confrontation of multiple traffic participants, characterized by: The method includes: Predicting the driving trajectory of the tested vehicle and obtaining the driving paths of traffic participants and scenario information of the test scenario, dividing the test scenario into a plurality of spatiotemporal resource grids based on the scenario information, obtaining conflict grids between the tested vehicle and traffic participants based on the driving trajectory and driving path, and constructing a conflict event set for each tested vehicle based on the conflict grids; Defining a conflict degree index of the tested vehicle, a safety index of traffic participants, and a travel time, wherein the conflict degree index includes a single spatiotemporal resource grid occupancy index and a conflict degree between the tested vehicle and traffic participants; the safety index includes the cumulative conflict time between all traffic participants; Based on the conflict event set, multiple traffic participants are used to collaboratively capture the vehicle under test; A multi-objective speed optimization search model is constructed based on the conflict degree index, safety index, and travel time. The multi-objective speed optimization search model is solved during the collaborative roundup process. The spatiotemporal resource occupancy table of the tested vehicle and traffic participants in the test scenario is obtained in combination with the spatiotemporal resource grid, that is, the timetable of the tested vehicle and traffic participants arriving at or leaving the corresponding spatiotemporal resource grid; Perform collision detection based on the spatiotemporal resource occupancy table, and generate optimal trajectories for all traffic participants if there is no conflict; otherwise, return to perform collaborative round-up; Obtaining comprehensive indicators of the tested vehicle when the traffic participant executes the optimal trajectory, and performing performance evaluation based on the comprehensive indicators; The space-time resource grid occupancy index is: ， ， ， ， in, represents the occupancy time of the spatiotemporal resource grid k; represents the time of entering the spatiotemporal resource grid k; represents the time of leaving the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the mth passed space-time resource grid; represents the time when the tested vehicle i enters the first spatiotemporal resource grid that needs to be passed; m represents the number of spatiotemporal resource grids that the tested vehicle i passes through; Indicates the width of the spatiotemporal resource grid; represents the speed of the tested vehicle i; Indicates the number of resource blocks that the tested vehicle is expected to pass through; represents the time when the tested vehicle i leaves the mth space-time resource grid; represents the length of the measured vehicle i; The conflict degree between the tested vehicle and the traffic participants is: ， ， in, represents the conflict time between the tested vehicle i and the traffic participant j on the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; A calculation function representing the conflict time between the tested vehicle and traffic participants; represents the time when the tested vehicle i leaves the spatiotemporal resource grid k; represents the time when traffic participant j leaves the spatiotemporal resource grid k; represents the time when the tested vehicle i enters the spatiotemporal resource grid k; represents the time when traffic participant j enters the spatiotemporal resource grid k; represents the conflict degree, i.e., the cumulative resource block conflict time between the tested vehicle i and all traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; Represents the set of all tested vehicles and traffic participants.

2. The autonomous driving vehicle testing method based on collaborative round-up confrontation of multiple traffic participants according to claim 1 is characterized in that: The transit times are: ， in, Indicates the travel time; Indicates the time to pass the test scenario; Indicates the time when the test scenario begins.

3. The autonomous driving vehicle testing method based on multi-traffic participant collaborative encirclement and confrontation according to claim 1 is characterized in that: The safety indicators are: ， in, represents the cumulative conflict time between traffic participant j and all other traffic participants; represents the kth spatiotemporal resource grid; Represents a collection of spatiotemporal resource grids; represents the set of all traffic participants except traffic participant j; represents traffic participant t; represents the conflict time between traffic participants j and t on the spatiotemporal resource grid k.

4. The autonomous driving vehicle testing method based on collaborative round-up and confrontation of multiple traffic participants according to claim 1 is characterized in that: The multi-objective speed optimization search model includes an objective function and constraints, wherein the objective function is: ， in, Represents the parameter information of all tested vehicles and traffic participants, and and Indicates the vehicle's location information. Indicates vehicle speed; 、 and represents weight; Indicates the conflict degree between the tested vehicle and traffic participants; Indicates the travel time; Indicates safety indicators.

5. The autonomous driving vehicle testing method based on collaborative round-up and confrontation of multiple traffic participants according to claim 4 is characterized in that: The constraints are: ， ， in, , representing vehicle p and vehicle q, and Represents the set of all tested vehicles and traffic participants; Indicates that vehicle p enters the spatiotemporal resource grid time; Indicates that vehicle q leaves the spatiotemporal resource grid time, , represents the label of the spatiotemporal resource grid; represents positive infinity; Indicates the priority. If vehicle p passes before vehicle q, then ,on the contrary .

6. The autonomous driving vehicle testing method based on collaborative round-up and confrontation of multiple traffic participants according to claim 1 is characterized in that: The method for generating the optimal trajectory is: Constructing an optimization constraint model for traffic participants, and solving the optimization constraint model based on the spatiotemporal resource occupancy table to obtain solution parameters; the solution parameters include: a first acceleration change rate and a second acceleration change rate; Constructing a trigonometric function model, and solving the trigonometric function model based on the solution parameters to obtain speed values ​​in each time period; An optimal trajectory is generated based on the speed values ​​in each time period.

7. The autonomous driving vehicle testing method based on multi-traffic participant collaborative encirclement and confrontation according to claim 6 is characterized in that: The optimization constraint model is: ， ， in, The time for the traffic participant to arrive at the target spatiotemporal resource grid obtained based on the spatiotemporal resource occupancy table; Indicates the weight ratio of maximizing speed; Indicates the weight ratio of gentle acceleration; represents the acceleration time variation function; Indicates the rate of change of speed of a traffic participant when the speed is less than or equal to the expected driving acceleration; Indicates the rate of change of speed of a traffic participant when the speed is greater than the expected driving acceleration; Indicates the minimum acceleration of traffic participants; Indicates the maximum acceleration of traffic participants; when hour, It indicates the acceleration when the speed of the traffic participant is less than or equal to the expected driving acceleration. hour, Indicates the acceleration when the speed of the traffic participant is greater than the expected driving acceleration; hour, Indicates the speed of the traffic participant when the speed is less than or equal to the expected driving acceleration stage. hour, Indicates the speed of the traffic participant when the speed is greater than the expected driving acceleration stage; Indicates the time that the traffic participant is in the jerk phase; Indicates the time that the traffic participant is in the deceleration phase.

8. The autonomous driving vehicle testing method based on multi-traffic participant collaborative encirclement and confrontation according to claim 7 is characterized in that: The trigonometric functions are: ， in, Indicates the expected driving speed of traffic participants; Indicates the difference between the current vehicle speed and the turning speed; represents the first acceleration rate of change; Indicates time; Indicates the second acceleration rate of change.

Images