An accelerated testing method for continuous conflict scenarios involving multiple traffic participants in a closed area

Abstract

The present invention relates to an accelerated testing method for a continuous conflict scenario in a closed field with multiple traffic participants, comprising: generating a global path for the traffic participants based on the global path of the tested vehicle; constructing an interaction area between the tested vehicle and the traffic participants based on the global paths of the tested vehicle and the traffic participants, and assigning test tasks and priorities to the traffic participants in each interaction area; constructing an SL graph and an ST graph for each traffic participant based on known test tasks and priorities, and generating the optimal local trajectory of the traffic participant based on the SL graph and the ST graph; after obtaining the optimal local trajectory for each traffic participant, determining whether to terminate the test based on the execution status of all test tasks and the collision status with the tested vehicle. Compared with the prior art, the present invention implements accelerated testing of continuous conflict scenarios in a closed field while ensuring the driving safety of traffic participants by allocating a local path for each traffic participant based on the collision avoidance between multiple traffic participants and the active test tasks of the tested vehicle.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving vehicle testing technology, and in particular to an accelerated testing method for continuous conflict scenarios involving multiple traffic participants in a closed area. Background Art

[0002] Closed-field testing is a crucial safety verification step before autonomous vehicles are deployed. With the rapid development of autonomous driving technology, low-level, low-intelligence closed-field testing methods are no longer sufficient to assess high-level, highly intelligent autonomous vehicles. Existing closed-field testing methods suffer from numerous issues, specifically: The test equipment (dummy humans and dummy vehicles) used to simulate traffic participants lacks intelligence and lacks coordination during testing, resulting in poor interactivity within the autonomous driving test system and a failure to reflect the intelligence level of the autonomous vehicle. Testing methods for specific autonomous driving functions suffer from fragmented test conditions, simple scenario conditions, and fragmented functional testing. As a result, most test scenarios focus on a single condition, with fragmented and episodic testing predominating. This makes it difficult to reflect the autonomous vehicle's ability to cope with complex traffic environments. Testing methods based on random traffic flow models suffer from high costs, unclear scenario requirements, and random conflicts, resulting in long autonomous vehicle testing times and poor performance in edge scenarios, making it difficult to reflect the vehicle's intelligence level in these scenarios. Taking Chinese patent application CN118092393A as an example, although the cloud-controlled test system for autonomous vehicles established through 5G communication and cloud computing technology realizes the continuous interaction between simulated traffic participant clusters and test vehicles, and completes high-integrity tests that are close to the real traffic environment and cover high-risk, low-probability, and difficult-to-reproduce scenarios, it still has the following shortcomings: 1) The motion control of simulated traffic participants is completely dependent on preset instructions, and it is impossible to achieve active coordination of multiple traffic participants; 2) Taking the roundabout test as an example, the test scenario is to combine discrete scenarios such as roundabout merging, roundabout following, and roundabout merging. The test scenario is fragmented, and it is difficult to reproduce the complex conflicts of continuous interaction in real traffic.

[0003] Therefore, providing a testing method for autonomous driving vehicles that can realize active collaboration among multiple traffic participants and continuous complex scenarios is a technical problem that needs to be solved. Summary of the Invention

[0004] The purpose of the present invention is to provide an accelerated testing method for continuous conflict scenarios in a closed area with multiple traffic participants. A continuous complex test scenario is constructed through the global path of the tested vehicle, and in the continuous complex test scenario, local trajectory optimization is performed for each traffic participant with the goal of collision avoidance between multiple traffic participants and active testing tasks of the tested vehicle, thereby realizing accelerated testing of the tested vehicle in continuous and complex conflict scenarios.

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

[0006] The present invention provides an accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants, the method comprising:

[0007] Obtaining a global path of the vehicle under test, constructing a continuous conflict scenario based on the global path of the vehicle under test, and generating a global path of traffic participants using an optimized A* algorithm based on the continuous conflict scenario;

[0008] Based on the global path of the tested vehicle and the global path of the traffic participants, an interaction area between the tested vehicle and the traffic participants is constructed, and an optimization model based on an auction algorithm is constructed. For each interaction area, the optimization model is used to assign test tasks and priorities to the traffic participants;

[0009] For each traffic participant, the following steps are performed to obtain the optimal local trajectory:

[0010] Obtaining the vehicle states of the tested vehicle and traffic participants with a lower priority than the target traffic participant during the target traffic participant's execution of all assigned test tasks, performing obstacle projection based on the vehicle states of the traffic participants with a lower priority than the target traffic participant, performing task projection based on the vehicle state of the tested vehicle, and constructing an SL diagram and a ST diagram of the target traffic participant;

[0011] Acquiring road information in a test scenario, constructing a kinematic model of a target traffic participant and its constraints based on the road information, and acquiring a real-time state of the target traffic participant based on the kinematic model and its constraints; the road information including road curvature;

[0012] Constructing a local trajectory allocation multi-objective optimization function, sampling target traffic participant trajectory points in the SL map and the ST map based on the real-time state, and combining the trajectory points to generate a trajectory cluster, wherein the trajectory cluster includes multiple trajectories, calculating the local trajectory allocation multi-objective optimization function for each trajectory, and selecting the trajectory corresponding to the optimal value as the optimal local trajectory;

[0013] After obtaining the optimal local trajectory for each traffic participant, the execution status of all test tasks and the collision status between the tested vehicle and the traffic participant are calculated. If all test tasks are executed or the tested vehicle collides with the traffic participant, the test is terminated; otherwise, the interaction area is updated based on the optimal local trajectory, and the test tasks are reallocated.

[0014] As a preferred technical solution, the method for generating the global path of the traffic participants is:

[0015] Acquiring key path points based on the global path of the tested vehicle, selecting path points based on the key path points using an optimized A* algorithm, and generating a global path for traffic participants based on the path points;

[0016] The method for selecting path points is to calculate the comprehensive priority of the path points to be selected, and select the path points based on the comprehensive priority, and the expression is:

[0017] ,

[0018] ,

[0019] ,

[0020] ,

[0021] in, Indicates the distance between the path point n to be selected and the starting point; Indicates the estimated distance between the path point n to be selected and the end point; represents the sum of the distances between the path point to be selected n and all the key path points; represents the comprehensive priority of the i-th path point; Indicates the position of the path point n to be selected; Indicates the location of the end point; represents the location of the jth critical path point; Indicates the number of critical path points.

[0022] As a preferred technical solution, the method for allocating test tasks to traffic participants and their priorities when performing the test tasks is as follows:

[0023] Acquire an interaction area between the tested vehicle and the traffic participant based on the global path of the tested vehicle and the global path of the traffic participant, and define a test task set for the interaction area;

[0024] Based on the optimization model of the auction algorithm and the test task set, the test task allocation of traffic participants is performed. The expression of the optimization model is:

[0025] ,

[0026] ,

[0027] in, represents the net benefit of traffic participant i when performing test task j, which is obtained based on the distance between traffic participant i and the corresponding point of test task j; N represents the number of traffic participants; Indicates the number of test tasks; Represents the test task allocation status of traffic participant i, if It means that traffic participant i is assigned to perform test task j. It means that the traffic participant does not perform the test task j; represents the set of traffic participants; Indicates the upper limit of the number of traffic participants assigned to perform the test task;

[0028] Based on the global path of the tested vehicle, the order in which the tested vehicle passes through the road topology structure is obtained, and the priority of executing the test task is determined based on the order.

[0029] As a preferred technical solution, the method for constructing the SL graph and ST graph of the target traffic participant includes:

[0030] The target traffic participant is transformed from the Cartesian coordinate system to the Frenet coordinate system. In the Frenet coordinate system, the trajectory of the target traffic participant is hierarchically decoupled into the SL graph and the ST graph.

[0031] Traversing the trajectories of other traffic participants to obtain a potential interaction range, screening traffic participants with a lower priority than the target traffic participant within the potential interaction range, and projecting them as obstacles into the SL map and the ST map respectively;

[0032] Calculating the task time domain of the target traffic participant when performing each assigned test task based on the vehicle state of the tested vehicle, that is, the time it takes for the tested vehicle to reach the interaction area corresponding to the test task;

[0033] Obtain the physical dimensions of the tested vehicle and traffic participants, and based on the physical dimensions, project the task time domain corresponding to each assigned test task onto the T-axis of the ST diagram, and project the interaction area corresponding to the test task onto the S-axis of the SL diagram and the ST diagram to achieve task projection.

[0034] As a preferred technical solution, the obstacle projection method is:

[0035] The obstacle is subjected to geometric expansion modeling to obtain a geometric expansion model, which is expressed as follows:

[0036] ,

[0037] ,

[0038] ,

[0039] ,

[0040] in, The right front corner of the obstacle; The left front corner of the obstacle; Indicates the left rear corner of the obstacle; Indicates the right rear corner of the obstacle; Indicates the coordinate point of the obstacle center; The heading angle is the direction of the obstacle. Axis angle; Represents half of the longitudinal length of the obstacle; Represents half of the horizontal length of the obstacle; represents the longitudinal expansion coefficient; represents the lateral expansion coefficient;

[0041] The coordinates of the geometric expansion model are converted to the Frenet coordinate system, and the geometric expansion model in the Frenet coordinate system is projected onto the SL diagram and the ST diagram.

[0042] As a preferred technical solution, the projection method of the task time domain is:

[0043] ,

[0044] ,

[0045] in, Indicates the time when the tested vehicle arrives at the interaction area, and , Indicates the distance from the tested vehicle to the interaction area, Indicates the speed of the vehicle being tested; Indicates the T-axis projection range of the task time domain on the ST diagram; Represents half the width of the target traffic participant.

[0046] As a preferred technical solution, the projection method of the interactive area is:

[0047] ,

[0048] ,

[0049] in, Indicates the position of traffic participants arriving at the interaction area; Indicates half of the width of the vehicle being measured; Indicates the projection range of the interaction area onto the S axis.

[0050] As a preferred technical solution, the kinematic model and its constraints are:

[0051] ，

[0052] ,

[0053] ,

[0054] ,

[0055] ,

[0056] in, represents the derivative of the longitudinal distance of the traffic participant in the Frenet coordinate system; Indicates the speed of traffic participants; Indicates the heading angle of the traffic participant; Represents the curvature of the road in the Frenet coordinate system; Indicates the lateral distance traveled by the traffic participant in the Frenet coordinate system.

[0057] As a preferred technical solution, the local trajectory allocation multi-objective optimization function is:

[0058] ,

[0059] ,

[0060] ,

[0061] ,

[0062] ,

[0063] ,

[0064]

[0065] ,

[0066] ,

[0067] ,

[0068] ,

[0069] in, represents the multi-objective cost of the assignment of the i-th trajectory in the trajectory cluster; represents the centripetal acceleration weight; represents the centripetal acceleration cost; represents the lateral travel cost; represents the longitudinal travel cost; represents the lateral offset weight; represents the lateral offset cost; represents the speed at time t; Indicates the road curvature; Indicates the total time required for the driving trajectory; represents the lateral offset weight; represents the lateral offset cost; represents the lateral comfort weight; represents the lateral comfort cost; represents the longitudinal position at time t; represents the lateral offset at the longitudinal position s; represents the lateral error boundary; express Direction factor relative to the initial lateral offset; represents the second derivative of the lateral offset at the longitudinal position s; represents the longitudinal velocity at time t; represents the first derivative of the lateral offset at the longitudinal position s; represents the longitudinal acceleration at time t; represents the weight of reaching the end point of the trajectory; represents the cost of reaching the end point of the trajectory; Represents the weight of avoiding collision; represents the cost of avoiding a collision; represents the smoothing weight; represents the smoothing cost; represents the total number of trajectory points; represents the sampling time of trajectory point j; represents the deviation of trajectory point j from the expected velocity; represents the length of the i-th trajectory in the trajectory cluster; Gaussian function representing the distance cost between trajectory point j and the obstacle projection; represents the longitudinal acceleration of trajectory point j; represents the maximum longitudinal acceleration of trajectory point j; Gaussian function representing the distance cost between trajectory point j and the test task.

[0070] As a preferred technical solution, the method for obtaining the optimal local trajectory further includes: performing collision detection and task detection on the trajectory corresponding to the optimal value; if the requirements are met, the trajectory corresponding to the optimal value is the optimal local trajectory; otherwise, deleting the trajectory corresponding to the optimal value from the trajectory cluster and reselecting the optimal local trajectory;

[0071] The collision detection is as follows: traversing all obstacle projections in the SL graph and the ST graph, and determining whether the trajectory corresponding to the optimal value intersects with any of the obstacle projections. If so, a collision is determined to exist; otherwise, no collision exists.

[0072] The task detection is as follows: traverse all the test task projections in the ST diagram, determine whether the trajectory corresponding to the optimal value intersects with any of the test task projections, and if so, determine that the corresponding traffic participant performs the test task; otherwise, determine that the corresponding traffic participant does not perform the test task.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] 1) The present invention allocates the global path of the traffic participants based on the global path of the tested vehicle, and divides the interaction area according to the global paths of the two. For each interaction area, an optimization model based on an auction algorithm is used to allocate tasks to traffic participants and prioritize traffic participants in the corresponding tasks. This increases the initiative of traffic participants in the interactive decision-making process. The priority competition mechanism is used to dynamically adjust the order of multiple traffic participants in performing the same test task, reducing the absolute dependence on preset instructions and improving the flexibility of multiple traffic participants in performing test tasks.

[0075] 2) This application generates a continuous conflict scenario based on the global path of the vehicle under test, generates the global path of the traffic participants based on the key path points in the global path of the vehicle under test in the continuous conflict scenario, and performs local trajectory allocation in combination with the kinematic model and constraints in the Frenet coordinate system. When screening the optimal local path, the collision avoidance between multiple traffic participants and the active test tasks of the vehicle under test are integrated to achieve quantitative characterization of safe interactions between vehicles and active test tasks, so that the method provided by the present invention can obtain the best test task execution strategy within a safe range, thereby ensuring driving safety and accelerating the acceleration test of closed continuous conflict scenarios.

[0076] 3) The present invention solves the local optimal path for each of the multiple traffic participants participating in the test task separately, greatly reduces the complexity of the optimization problem through distributed computing, reduces computing power, and improves computing speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is a flow chart of the method of the present invention;

[0078] Figure 2 Schematic diagram of a continuous scene with multiple traffic participants at an intersection according to the present invention;

[0079] Figure 3Schematic diagram of calculation of obstacle projection and task projection range of the present invention;

[0080] Figure 4 Schematic diagram of the obstacle projection and task projection results of the present invention;

[0081] Figure 5 Schematic diagram of trajectory sampling of the present invention. DETAILED DESCRIPTION

[0082] 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.

[0083] 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.

[0084] In view of the technical problems existing in the prior art, the present invention provides an accelerated testing method for a continuous conflict scenario in a closed field with multiple traffic participants. The process is as follows: Figure 1 shown.

[0085] Detailed, including:

[0086] S1. Obtain the global path of the vehicle under test, and construct a continuous conflict scenario based on the global path of the vehicle under test. Generate the global path of the traffic participants using the optimized A* algorithm based on the continuous conflict scenario.

[0087] S11. Combine the high-precision map, operational design conditions (ODC), test scenario requirements and closed site conditions to predict the global path of the vehicle under test and express it in the Cartesian coordinate system. The results are as follows: Figure 2 As shown, the path of the tested vehicle is to turn left at the intersection.

[0088] Among them, ODC includes the test operation status of the vehicle under test, which is mainly vehicle speed information; the test scenario requirements include the road topology structure required by the vehicle under test, the complexity of the traffic environment, etc.

[0089] S12. Generate a global path for traffic participants.

[0090] S121. Acquire key path points based on the global path of the vehicle under test to form a continuous conflict scenario.

[0091] S122, based on the key path points in the continuous conflict scenario, the optimized A* algorithm is used to select path points, and the global path of the traffic participants is generated based on the path points. The result is as follows Figure 2 As shown, the global paths of traffic participants 1 and 3 are the closed-loop paths before the tested vehicle turns left, and the global paths of traffic participants 2 and 4 are the closed-loop paths after the tested vehicle turns left.

[0092] The optimized A* algorithm provided by the present invention adds the concept of path points to the A* algorithm to optimize the A* algorithm. Specifically, the concept of path points is quantified by calculating the comprehensive optimization level of the path points to be selected, and the path points are selected based on the comprehensive priority. The expression is:

[0093] ,

[0094] ,

[0095] ,

[0096] ,

[0097] in, Represents the distance between the path point n to be selected and the starting point, that is, the sum of the weights of the path; It represents the estimated distance between the path point n to be selected and the end point, calculated using the Euclidean distance; It represents the sum of the distances between the path point to be selected n and all key path points, calculated using the Euclidean distance; represents the comprehensive priority of the i-th path point; Indicates the position of the path point n to be selected; Indicates the location of the end point; represents the location of the jth critical path point; Indicates the number of critical path points.

[0098] S2. Based on the global path of the tested vehicle and the global path of the traffic participants, an interaction area between the tested vehicle and the traffic participants is constructed. For each interaction area, a test task and a priority of the traffic participants are assigned using an optimization model.

[0099] In the method provided by the present invention, each test task can be jointly performed by multiple traffic participants; each traffic participant can perform multiple test tasks; and cloud-based collaborative control technology is used in the present invention to realize task allocation for multiple traffic participants and control of traffic participants according to the divided priorities.

[0100] The detailed steps are:

[0101] S21. Obtain an interaction area between the tested vehicle and the traffic participant based on the global path of the tested vehicle and the global path of the traffic participant, and define a corresponding test task set for the interaction area.

[0102] The interaction area is a circular area with a radius of 1m that expands from the global path trajectory coincidence point, such as Figure 2 The circled area in .

[0103] S22. Based on the optimization model of the auction algorithm and the test task set, the test tasks of traffic participants are allocated. The optimization allocation process is executed in the cloud. The expression of the optimization model is:

[0104] ,

[0105] ,

[0106] in, represents the net benefit of traffic participant i when performing test task j, which is obtained based on the distance between traffic participant i and the corresponding point of test task j; N represents the number of traffic participants; Indicates the number of test tasks; Represents the test task allocation status of traffic participant i, if It means that traffic participant i is assigned to perform test task j. It means that the traffic participant does not perform the test task j; represents the set of traffic participants; Indicates the upper limit of the number of traffic participants assigned to perform the test task.

[0107] S23. Based on the global path of the tested vehicle, obtain the order in which the tested vehicle passes through the road topology structure, and determine the priority of executing the test task based on the order.

[0108] For each traffic participant, execute S3 to S5 to obtain the optimal local trajectory:

[0109] S3, obtain the vehicle status of the tested vehicle and the traffic participants with lower priority than the target traffic participant during the execution of all assigned test tasks, perform obstacle projection and task projection based on the vehicle status, and construct the SL diagram and ST diagram. The effect is as follows Figure 4 As shown, in Figure 4 Traffic participant 3 is selected as the target traffic participant, and the details are in Figure 4 The driving trajectory of the tested vehicle is to turn right into the fork where traffic participant 4 is located. Traffic participant 3 goes straight, and the interaction area between the tested vehicle and traffic participant 3 is the circle in the figure.

[0110] In detail, the SL graph takes the traffic participant’s own position as the starting point, and the ST graph takes the traffic participant’s own position and the current time as the starting point.

[0111] S31. Convert the target traffic participant from the Cartesian coordinate system to the Frenet coordinate system. In the Frenet coordinate system, decouple the trajectory of the target traffic participant into a SL graph and an ST graph.

[0112] S32, traverse the trajectories of other traffic participants, obtain the potential interaction range, filter out traffic participants with lower priority than the target traffic participant in the potential interaction range, and project them into the SL map and ST map as obstacles. Figure 4 Taking the scenario shown in the figure as an example, it is assumed that the priority of traffic participant 2 and traffic participant 4 is lower than that of traffic participant 3. Then for traffic participant 3, traffic participant 2 and traffic participant 4 are both obstacles. After executing S321~S322, the projection result is as follows: Figure 4 As shown in .

[0113] S321. Perform geometric expansion modeling on the obstacle to obtain a geometric expansion model, which is expressed as:

[0114] ,

[0115] ,

[0116] ,

[0117] ,

[0118] in, The right front corner of the obstacle; The left front corner of the obstacle; Indicates the left rear corner of the obstacle; Indicates the right rear corner of the obstacle; Indicates the coordinate point of the obstacle center; The heading angle is the direction of the obstacle. Axis angle; Represents half of the longitudinal length of the obstacle; Represents half of the horizontal length of the obstacle; represents the longitudinal expansion coefficient; represents the lateral expansion coefficient;

[0119] S322. Convert the coordinates of the geometric expansion model to the Frenet coordinate system, and project the geometric expansion model in the Frenet coordinate system to the SL diagram and the ST diagram.

[0120] Specifically, the geometric expansion model is converted from the Cartesian coordinate system to the Frenet coordinate system (where the origin is the position of the target traffic participant), and the coordinates of each vertex of the obstacle are obtained. The coordinate point value depends on the vertices of the obstacle. The coordinate point values ​​project the obstacle into the SL map.

[0121] The relative value of the longitudinal position of the target traffic participant and the target traffic participant is obtained from the geometric expansion model of the obstacle at each moment, and is projected onto the S axis of the ST diagram. The corresponding moment is projected onto the T axis of the ST diagram to realize the projection of the obstacle on the ST diagram.

[0122] S33. Calculate the task time domain of the target traffic participant when performing each assigned test task based on the vehicle state of the tested vehicle, that is, the time it takes for the tested vehicle to reach the interaction area corresponding to the test task.

[0123] S34, obtain the physical dimensions of the tested vehicle and traffic participants, project the task time domain corresponding to each assigned test task onto the T axis of the ST diagram based on the physical dimensions, and project the interaction area corresponding to the test task (i.e. Figure 4 The area shown in the circle in the middle is projected onto the S axis of the SL diagram and the ST diagram to realize task projection. Specifically, when projecting the task time domain and the interaction area, the calculation diagram of the projection range is as follows: Figure 3 As shown, the projection method of the task time domain is:

[0124] ,

[0125] ,

[0126] in, Indicates the time when the tested vehicle arrives at the interaction area, and , Indicates the distance from the tested vehicle to the interaction area, Indicates the speed of the vehicle being tested; Indicates the T-axis projection range of the task time domain on the ST diagram; Indicates half the width of the target traffic participant (i.e., traffic participant 3).

[0127] The projection method of the interaction area is:

[0128] ,

[0129] ,

[0130] in, Indicates the position of traffic participants arriving at the interaction area; Indicates half of the width of the vehicle being measured; Indicates the projection range of the interaction area onto the S axis.

[0131] S4. Obtain road information in the test scene, construct a kinematic model of the target traffic participant and its constraints based on the road information, and obtain the real-time status of the target traffic participant based on the kinematic model and its constraints.

[0132] Considering the road information, the kinematic model and constraints of traffic participants are constructed. The road information includes the road curvature, which is expressed as:

[0133] ，

[0134] ,

[0135] ,

[0136] ,

[0137] ,

[0138] in, represents the derivative of the longitudinal distance of the traffic participant in the Frenet coordinate system; Indicates the speed of traffic participants; Indicates the heading angle of the traffic participant; Represents the curvature of the road in the Frenet coordinate system; Indicates the lateral distance traveled by the traffic participant in the Frenet coordinate system.

[0139] S5. Construct a multi-objective optimization function for local trajectory allocation, sample trajectory points in the SL graph and ST graph based on the real-time state to generate trajectory clusters, and screen the trajectories in the trajectory clusters based on the multi-objective optimization function to obtain the optimal local trajectory.

[0140] In this step, the target traffic participant's trajectory points are sampled in the SL and ST graphs based on the real-time status, and the trajectory points are combined to generate a trajectory cluster. The trajectory cluster includes multiple trajectories. The multi-objective optimization function is used to calculate the local trajectory of each trajectory and assign the multi-objective optimization function. The trajectory corresponding to the optimal value is selected as the optimal local trajectory to achieve the optimal trajectory solution that meets the test task requirements and the safe interaction between traffic participants through local search.

[0141] by Figure 4 The scenario in is an example. Figure 4 The scene in the example performs point sampling on the SL graph and ST graph based on the real-time status. The results are as follows: Figure 5 As shown in the black dots, based on the scattered sampling results, polynomial curve fitting is used to generate longitudinal and transverse polynomial trajectories, and the transverse and longitudinal polynomial trajectories are synthesized and the synthesized trajectories are converted into Cartesian coordinate points, finally forming a complete trajectory. This process is repeated to generate trajectory clusters. The effect is shown in the figure. Figure 5 Indicated by the dotted line containing the arrow.

[0142] The detailed polynomial fitting trajectory method includes:

[0143] Calculate the longitudinal polynomial coefficients based on the target state and the initial state, and calculate the longitudinal trajectory based on the longitudinal polynomial coefficients:

[0144] ,

[0145] ,

[0146] ,

[0147] ,

[0148] ,

[0149] ,

[0150] ,

[0151] ,

[0152] ,

[0153] in, represents the longitudinal distance function; 、 、 、 、 and represents the longitudinal polynomial coefficient; Indicates time; represents the longitudinal velocity function; represents the longitudinal acceleration function; Indicates the initial vertical position; represents the initial longitudinal velocity; represents the initial longitudinal acceleration; express The target's longitudinal position at the moment; Indicates the target longitudinal velocity; Indicates the target longitudinal acceleration.

[0154] The lateral polynomial coefficients are calculated based on the target state and the initial state, and the lateral trajectory is calculated based on the lateral polynomial coefficients. The expression is:

[0155] ,

[0156] ,

[0157] ,

[0158] ,

[0159] ,

[0160] ,

[0161] ,

[0162] ,

[0163] ,

[0164] in, 、 、 、 、 and represents the transverse polynomial coefficient; Indicates the initial horizontal position; represents the initial lateral velocity; represents the initial lateral acceleration; express The target's lateral position at the moment; Indicates the target lateral velocity; represents the target lateral acceleration; Indicates time.

[0165] In the present invention, the local trajectory allocation multi-objective optimization function is:

[0166] ,

[0167] ,

[0168] ,

[0169] ,

[0170] ,

[0171] ,

[0172]

[0173] ,

[0174] ,

[0175] ,

[0176] ,

[0177] in, represents the assigned multi-objective cost of the i-th trajectory in the trajectory cluster; represents the centripetal acceleration weight; represents the centripetal acceleration cost; represents the lateral travel cost; represents the longitudinal travel cost; represents the lateral offset weight; represents the lateral offset cost; represents the speed at time t; Indicates the road curvature; Indicates the total time required for the driving trajectory; represents the lateral offset weight; represents the lateral offset cost; represents the lateral comfort weight; represents the lateral comfort cost; represents the longitudinal position at time t; represents the lateral offset at the longitudinal position s; represents the lateral error boundary; express Directional factor relative to the initial lateral offset; represents the second derivative of the lateral offset at the longitudinal position s; represents the longitudinal velocity at time t; represents the first derivative of the lateral offset at the longitudinal position s; represents the longitudinal acceleration at time t; represents the weight of reaching the end point of the trajectory; represents the cost of reaching the end point of the trajectory; Represents the weight of avoiding collision; represents the cost of avoiding a collision; represents the smoothing weight; represents the smoothing cost; represents the total number of trajectory points; represents the sampling time of trajectory point j; represents the deviation of trajectory point j from the expected velocity; represents the length of the i-th trajectory in the trajectory cluster; The Gaussian function representing the distance cost between trajectory point j and the obstacle projection, and , represents the variance of the Gaussian function, represents the projection distance function between the trajectory point and the obstacle and , Indicates that the trajectory point is within the obstacle projection judgment window. Indicates that the trajectory point is outside the obstacle projection judgment window; represents the longitudinal acceleration of trajectory point j; represents the maximum longitudinal acceleration of trajectory point j; represents the Gaussian function of the distance cost between trajectory point j and the test task, and , represents the projection distance function between the trajectory point and the test task and , Indicates the preset coefficient, Indicates that the trajectory point is within the test task projection judgment window, Indicates that the trajectory point is in the test task projection judgment window.

[0178] In addition, when generating the optimal local trajectory, the trajectory needs to be further screened. Specifically, the trajectory corresponding to the optimal value is subjected to collision detection and task detection. If the requirements are met, the trajectory corresponding to the optimal value is the optimal local trajectory; otherwise, the trajectory corresponding to the optimal value is deleted from the trajectory cluster and the optimal local trajectory is reselected, so that the screened trajectory can meet the test task as much as possible under the premise of safety.

[0179] The collision detection is as follows: traverse all obstacle projections in the SL graph and the ST graph, and determine whether the trajectory corresponding to the optimal value intersects with any obstacle projection. If so, it is determined that there is a collision, otherwise there is no collision.

[0180] Task detection is as follows: traverse all test task projections in the ST graph, determine whether the trajectory corresponding to the optimal value intersects with any test task projection, and if so, determine that the corresponding traffic participant performs the test task; otherwise, determine that the corresponding traffic participant does not perform the test task.

[0181] S6. In a continuous collision scenario, the cloud sends the planned optimal local trajectory to the corresponding traffic participants. After each traffic participant executes the optimal local trajectory, the execution status of all test tasks and the collision status between the tested vehicle and the traffic participant are obtained. If all test tasks are executed or the tested vehicle collides with the traffic participant, the test is terminated, and the actual driving data of the tested vehicle and the traffic participant are integrated to evaluate the test results of the tested vehicle. Otherwise, return to S2, update the interaction area based on the optimal local trajectory, and reallocate test tasks.

[0182] In addition, the present invention also provides an electronic device, 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 loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. 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.

[0183] 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.

[0184] 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).

[0185] 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.

[0186] 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.

[0187] 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.

[0188] 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. An accelerated testing method for a continuous conflict scenario involving multiple traffic participants in a closed area, characterized in that: The method includes: Obtaining a global path of the vehicle under test, constructing a continuous conflict scenario based on the global path of the vehicle under test, and generating a global path of traffic participants using an optimized A* algorithm based on the continuous conflict scenario; Based on the global path of the tested vehicle and the global path of the traffic participants, an interaction area between the tested vehicle and the traffic participants is constructed, and an optimization model based on an auction algorithm is constructed. For each interaction area, the optimization model is used to assign test tasks and priorities to the traffic participants; For each traffic participant, the following steps are performed to obtain the optimal local trajectory: Obtaining the vehicle states of the tested vehicle and traffic participants with a lower priority than the target traffic participant during the target traffic participant's execution of all assigned test tasks, performing obstacle projection based on the vehicle states of the traffic participants with a lower priority than the target traffic participant, performing task projection based on the vehicle state of the tested vehicle, and constructing an SL diagram and a ST diagram of the target traffic participant; Acquiring road information in a test scenario, constructing a kinematic model of a target traffic participant and its constraints based on the road information, and acquiring a real-time state of the target traffic participant based on the kinematic model and its constraints; the road information including road curvature; Constructing a local trajectory allocation multi-objective optimization function, sampling target traffic participant trajectory points in the SL map and the ST map based on the real-time state, and combining the trajectory points to generate a trajectory cluster, wherein the trajectory cluster includes multiple trajectories, calculating the local trajectory allocation multi-objective optimization function for each trajectory, and selecting the trajectory corresponding to the optimal value as the optimal local trajectory; After obtaining the optimal local trajectory for each traffic participant, the execution status of all test tasks and the collision status between the tested vehicle and the traffic participant are calculated. If all test tasks are executed or the tested vehicle collides with the traffic participant, the test is terminated; otherwise, the interaction area is updated based on the optimal local trajectory, and the test tasks are reallocated.

2. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The method for generating the global path of the traffic participants is: Acquiring key path points based on the global path of the tested vehicle, selecting path points based on the key path points using an optimized A* algorithm, and generating a global path for traffic participants based on the path points; The method for selecting path points is to calculate the comprehensive priority of the path points to be selected, and select the path points based on the comprehensive priority, and the expression is: ， ， ， ， in, Indicates the distance between the path point n to be selected and the starting point; Indicates the estimated distance between the path point n to be selected and the end point; represents the sum of the distances between the path point to be selected n and all the key path points; represents the comprehensive priority of the i-th path point; Indicates the position of the path point n to be selected; Indicates the location of the end point; represents the location of the jth critical path point; Indicates the number of critical path points.

3. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The method for allocating test tasks to traffic participants and their priorities when performing the test tasks is as follows: Acquire an interaction area between the tested vehicle and the traffic participant based on the global path of the tested vehicle and the global path of the traffic participant, and define a test task set for the interaction area; Based on the optimization model of the auction algorithm and the test task set, the test task allocation of traffic participants is performed. The expression of the optimization model is: ， ， in, represents the net benefit of traffic participant i when performing test task j, which is obtained based on the distance between traffic participant i and the corresponding point of test task j; N represents the number of traffic participants; Indicates the number of test tasks; Represents the test task allocation status of traffic participant i, if It means that traffic participant i is assigned to perform test task j. It means that the traffic participant does not perform the test task j; represents the set of traffic participants; Indicates the upper limit of the number of traffic participants assigned to perform the test task; Based on the global path of the tested vehicle, the order in which the tested vehicle passes through the road topology structure is obtained, and the priority of executing the test task is determined based on the order.

4. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The method for constructing the SL graph and ST graph of the target traffic participant includes: The target traffic participant is transformed from the Cartesian coordinate system to the Frenet coordinate system. In the Frenet coordinate system, the trajectory of the target traffic participant is hierarchically decoupled into the SL graph and the ST graph. Traversing the trajectories of other traffic participants to obtain a potential interaction range, screening traffic participants with a lower priority than the target traffic participant within the potential interaction range, and projecting them as obstacles into the SL map and the ST map respectively; Calculating the task time domain of the target traffic participant when performing each assigned test task based on the vehicle state of the tested vehicle, that is, the time it takes for the tested vehicle to reach the interaction area corresponding to the test task; Obtain the physical dimensions of the tested vehicle and traffic participants, and based on the physical dimensions, project the task time domain corresponding to each assigned test task onto the T-axis of the ST diagram, and project the interaction area corresponding to the test task onto the S-axis of the SL diagram and the ST diagram to achieve task projection.

5. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 4, characterized in that: The projection method of the obstacle is: The obstacle is subjected to geometric expansion modeling to obtain a geometric expansion model, which is expressed as follows: ， ， ， ， in, The right front corner of the obstacle; The left front corner of the obstacle; Indicates the left rear corner of the obstacle; Indicates the right rear corner of the obstacle; Indicates the coordinate point of the obstacle center; The heading angle is the direction of the obstacle. Axis angle; Represents half of the longitudinal length of the obstacle; Represents half of the horizontal length of the obstacle; represents the longitudinal expansion coefficient; represents the lateral expansion coefficient; The coordinates of the geometric expansion model are converted to the Frenet coordinate system, and the geometric expansion model in the Frenet coordinate system is projected onto the SL diagram and the ST diagram.

6. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 4, characterized in that: The projection method of the task time domain is: ， ， in, Indicates the time when the tested vehicle arrives at the interaction area, and , Indicates the distance from the tested vehicle to the interaction area, Indicates the speed of the vehicle being tested; Indicates the T-axis projection range of the task time domain on the ST diagram; Represents half the width of the target traffic participant.

7. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 4, characterized in that: The projection method of the interactive area is: ， ， in, Indicates the position of traffic participants arriving at the interaction area; Indicates half of the width of the vehicle being measured; Indicates the projection range of the interaction area onto the S axis.

8. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The kinematic model and its constraints are: ， ， ， ， ， in, represents the derivative of the longitudinal distance of the traffic participant in the Frenet coordinate system; Indicates the speed of traffic participants; Indicates the heading angle of the traffic participant; Represents the curvature of the road in the Frenet coordinate system; Indicates the lateral distance traveled by the traffic participant in the Frenet coordinate system.

9. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The local trajectory allocation multi-objective optimization function is: ， ， ， ， ， ， ， ， ， ， ， in, represents the multi-objective cost of the assignment of the i-th trajectory in the trajectory cluster; represents the centripetal acceleration weight; represents the centripetal acceleration cost; represents the lateral travel cost; represents the longitudinal travel cost; represents the lateral offset weight; represents the lateral offset cost; represents the speed at time t; Indicates the road curvature; Indicates the total time required for the driving trajectory; represents the lateral offset weight; represents the lateral offset cost; represents the lateral comfort weight; represents the lateral comfort cost; represents the longitudinal position at time t; represents the lateral offset at the longitudinal position s; represents the lateral error boundary; express Directional factor relative to the initial lateral offset; represents the second derivative of the lateral offset at the longitudinal position s; represents the longitudinal velocity at time t; represents the first derivative of the lateral offset at the longitudinal position s; represents the longitudinal acceleration at time t; represents the weight of reaching the end point of the trajectory; represents the cost of reaching the end point of the trajectory; Represents the weight of avoiding collision; represents the cost of avoiding a collision; represents the smoothing weight; represents the smoothing cost; represents the total number of trajectory points; represents the sampling time of trajectory point j; represents the deviation of trajectory point j from the expected velocity; represents the length of the i-th trajectory in the trajectory cluster; Gaussian function representing the distance cost between trajectory point j and the obstacle projection; represents the longitudinal acceleration of trajectory point j; represents the maximum longitudinal acceleration of trajectory point j; Gaussian function representing the distance cost between trajectory point j and the test task.

10. The accelerated testing method for a continuous conflict scenario in a closed area with multiple traffic participants according to claim 1, characterized in that: The method for obtaining the optimal local trajectory further includes: performing collision detection and task detection on the trajectory corresponding to the optimal value; if the requirements are met, the trajectory corresponding to the optimal value is the optimal local trajectory; otherwise, deleting the trajectory corresponding to the optimal value from the trajectory cluster and reselecting the optimal local trajectory; The collision detection is as follows: traversing all obstacle projections in the SL graph and the ST graph, and determining whether the trajectory corresponding to the optimal value intersects with any of the obstacle projections. If so, a collision is determined to exist; otherwise, no collision exists. The task detection is as follows: traverse all the test task projections in the ST diagram, determine whether the trajectory corresponding to the optimal value intersects with any of the test task projections, and if so, determine that the corresponding traffic participant performs the test task; otherwise, determine that the corresponding traffic participant does not perform the test task.

Images