Vehicle maneuverability evaluation method, device, electronic device and storage medium

Abstract

The present disclosure relates to a vehicle maneuverability smoothness evaluation method, device, electronic device and storage medium, and relates to the field of autonomous driving technology. The method comprises: obtaining a high-precision map and driving data of a road section passed by a vehicle to be evaluated, and the vehicle to be evaluated drives based on an autonomous driving scheme to be tested; determining at least one lateral maneuverability evaluation parameter and at least one longitudinal maneuverability evaluation parameter based on the high-precision map and driving data; calculating lateral maneuverability smoothness based on at least one lateral maneuverability evaluation parameter; calculating longitudinal maneuverability smoothness based on at least one longitudinal maneuverability evaluation parameter; determining the maneuverability smoothness of the vehicle to be evaluated when executing the autonomous driving scheme to be tested based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness. The present disclosure can evaluate the stability of a vehicle when executing an autonomous driving scheme more accurately to ensure the validity of the evaluation results.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a vehicle maneuverability assessment method, device, electronic device and storage medium. Background Art

[0002] Autonomous driving uses advanced communication, computer, network and control technologies to achieve real-time and continuous control of vehicles. With the development of technology, autonomous driving technology is being applied to more and more scenarios, such as automatic vehicle wake-up, start-up and sleep, automatic parking, automatic cleaning, automatic driving, automatic parking, automatic door opening and closing, and automatic fault recovery.

[0003] The autonomous driving technology is based on a pre-set autonomous driving scheme. When the user selects the autonomous driving mode, the vehicle drives according to the preset autonomous driving scheme. For example, after the driver selects the adaptive cruise mode (ACC), the corresponding driving scheme is: if the vehicle ahead is unobstructed, the adaptive cruise (ACC) will maintain the set maximum cruising speed and drive forward; if a vehicle is detected ahead, the adaptive cruise (ACC) will reduce the speed as needed and maintain a distance from the vehicle ahead based on the selected time until the appropriate cruising speed is reached.

[0004] However, existing technologies are unable to evaluate the smoothness of autonomous driving solutions. Therefore, there is an urgent need to develop a method to evaluate autonomous driving solutions in order to improve the smoothness of autonomous driving. Summary of the invention

[0005] In order to overcome the problems existing in the related art, the present disclosure provides a vehicle maneuverability assessment method, device, electronic device and storage medium.

[0006] According to a first aspect of an embodiment of the present disclosure, a method for evaluating vehicle maneuverability is provided, comprising:

[0007] Obtaining high-precision maps and driving data of a road section passed by a vehicle to be evaluated, wherein the vehicle to be evaluated is traveling based on the autonomous driving solution to be tested;

[0008] Determining at least one lateral maneuvering assessment parameter and at least one longitudinal maneuvering assessment parameter according to the high-precision map and the driving data;

[0009] Calculating a lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter, the lateral maneuvering smoothness being used to evaluate a directional control capability of the vehicle during motion;

[0010] calculating a longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter, the longitudinal maneuvering smoothness being used to evaluate a speed control capability during vehicle motion;

[0011] The maneuverability smoothness of the vehicle to be evaluated when executing the automatic driving scheme to be tested is determined based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness.

[0012] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameter includes a trajectory deviation degree; determining at least one lateral maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining a first driving curve and a second driving curve of the vehicle to be evaluated based on the driving data and the high-precision map, the first driving curve being used to indicate a real trajectory of the vehicle to be evaluated, and the second driving curve being used to indicate a planned trajectory of the vehicle to be evaluated on the high-precision map; and calculating the trajectory deviation degree based on the first driving curve and the second driving curve.

[0013] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameter includes trajectory smoothness; determining at least one lateral maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining a first driving curve and a third driving curve of the vehicle to be evaluated based on the driving data and the high-precision map, the first driving curve being used to indicate a real trajectory of the vehicle to be evaluated, and the third driving curve being used to indicate a road reference curve of the vehicle to be evaluated on the high-precision map; determining the trajectory smoothness based on the correlation between the first driving curve and the third driving curve.

[0014] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameter includes running angle smoothness; determining at least one lateral maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the real trajectory points and road reference points of the vehicle to be evaluated based on the driving data and the high-precision map; determining a first driving curve and a third driving curve of the vehicle to be evaluated based on the real trajectory points and the road reference points, respectively, the first driving curve being used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve being used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map; determining a first curvature corresponding to the real trajectory point based on the first driving curve; determining a second curvature corresponding to the road reference point based on the third driving curve; and calculating the running angle smoothness based on the first curvature and the second curvature.

[0015] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameter includes the number of times the vehicle deviates from the road; determining at least one lateral maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the real trajectory points of the vehicle to be evaluated, the heading points passed by the vehicle to be evaluated during driving, and the time period and lane type corresponding to the driving process of the vehicle to be evaluated based on the driving data and the high-precision map; determining the coordinates of the real trajectory points of the vehicle to be evaluated in the high-precision map; based on the coordinates, the heading points and the lane type, determining the time period of deviating from the track from the time period corresponding to the driving process of the vehicle to be evaluated; and calculating the number of times the vehicle deviates from the track based on the time period of deviating from the track.

[0016] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes the number of emergency brakes; determining at least one longitudinal maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the total vehicle driving time and the acceleration change curve of the vehicle to be evaluated based on the high-precision map and the driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, the preset time interval and the acceleration change curve; calculating the number of emergency brakes based on the acceleration corresponding to each time node and the preset braking acceleration threshold.

[0017] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes an average acceleration; determining at least one longitudinal maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the total vehicle driving time and the speed change curve of the vehicle to be evaluated based on the high-precision map and the driving data; determining the speed corresponding to each time node based on the total vehicle driving time, a preset time interval and the speed change curve; calculating the average acceleration based on the speed corresponding to each time node and the number of time nodes in the total vehicle driving time.

[0018] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes an average jerk; determining at least one longitudinal maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the total vehicle driving time and the acceleration change curve of the vehicle to be evaluated based on the high-precision map and the driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, a preset time interval and the acceleration change curve; calculating the average jerk based on the acceleration corresponding to each time node and the number of time nodes in the total vehicle driving time.

[0019] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes the number of sudden accelerations; determining at least one longitudinal maneuvering evaluation parameter based on the high-precision map and the driving data includes: determining the total vehicle driving time and the acceleration change curve of the vehicle to be evaluated based on the high-precision map and the driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, the preset time interval and the acceleration change curve; calculating the number of sudden accelerations based on the acceleration corresponding to each time node and the preset acceleration threshold.

[0020] In one embodiment of the present disclosure, the calculation of lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter includes: constructing a lateral judgment matrix based on the lateral maneuvering evaluation parameter, each element in the lateral judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding lateral maneuvering evaluation parameters; performing a consistency check on the lateral judgment matrix; when the lateral judgment matrix passes the consistency check, determining the weight of each lateral maneuvering evaluation parameter based on the value of each element in the lateral judgment matrix; and performing weighted summation processing on all the lateral maneuvering evaluation parameters according to the weight of each lateral maneuvering evaluation parameter to obtain the lateral maneuvering smoothness.

[0021] In one embodiment of the present disclosure, the calculation of the longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter includes: constructing a longitudinal judgment matrix based on the longitudinal maneuvering evaluation parameters, each element in the longitudinal judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding longitudinal maneuvering evaluation parameters; performing a consistency check on the longitudinal judgment matrix; when the longitudinal judgment matrix passes the consistency check, determining the weight of each longitudinal maneuvering evaluation parameter based on the value of each element in the longitudinal judgment matrix; and performing weighted summation processing on all the longitudinal maneuvering evaluation parameters according to the weight of each longitudinal maneuvering evaluation parameter to obtain the longitudinal maneuvering smoothness.

[0022] In one embodiment of the present disclosure, determining the maneuverability smoothness of the vehicle to be evaluated based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness includes: constructing a smoothness judgment matrix based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness, wherein each element in the smoothness judgment matrix is ​​used to indicate a parameter importance comparison result of the lateral maneuverability smoothness and the longitudinal maneuverability smoothness; performing a consistency check on the smoothness judgment matrix; when the smoothness judgment matrix passes the consistency check, determining the weights of the longitudinal maneuverability smoothness and the lateral maneuverability smoothness based on the values ​​of each element in the smoothness judgment matrix; and performing a weighted summation process on the longitudinal maneuverability evaluation degree and the lateral maneuverability smoothness according to the weights of the longitudinal maneuverability smoothness and the lateral maneuverability smoothness to obtain the maneuverability smoothness.

[0023] According to a second aspect of an embodiment of the present disclosure, there is provided a vehicle maneuverability stability assessment device, comprising: an acquisition module, a first data processing module, a second data processing module, and a third data processing module;

[0024] The acquisition module is used to acquire high-precision maps and driving data of a road section passed by a vehicle to be evaluated, wherein the vehicle to be evaluated is traveling based on the autonomous driving solution to be tested;

[0025] The first data processing module is used to determine at least one lateral maneuvering evaluation parameter and at least one longitudinal maneuvering evaluation parameter according to the high-precision map and the driving data;

[0026] The second data processing module is used to: calculate lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter, the lateral maneuvering smoothness is used to evaluate the directional control ability of the vehicle during movement; calculate longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter, the longitudinal maneuvering smoothness is used to evaluate the speed control ability of the vehicle during movement;

[0027] The third data processing module is used to determine the maneuverability smoothness of the vehicle to be evaluated when executing the automatic driving scheme to be tested based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness.

[0028] In one embodiment of the present disclosure, the first data processing module 220 is further used to calculate at least one evaluation parameter among the following 8 evaluation parameters.

[0029] 1. Track deviation

[0030] Determine a first driving curve and a second driving curve of the vehicle to be evaluated according to the driving data and the high-precision map, wherein the first driving curve is used to indicate the actual trajectory of the vehicle to be evaluated, and the second driving curve is used to indicate the planned trajectory of the vehicle to be evaluated on the high-precision map;

[0031] The trajectory deviation degree is calculated based on the first driving curve and the second driving curve.

[0032] 2. Trajectory smoothness

[0033] Determine a first driving curve and a third driving curve of the vehicle to be evaluated according to the driving data and the high-precision map, wherein the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map;

[0034] Based on the correlation between the first driving curve and the third driving curve, a trajectory smoothness is determined.

[0035] 3. Running angle smoothness

[0036] Determine the actual trajectory points and road reference points of the vehicle to be evaluated based on driving data and high-precision maps;

[0037] According to the real trajectory point and the road reference point, a first driving curve and a third driving curve of the vehicle to be evaluated are determined respectively, the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map;

[0038] Based on the first driving curve, determining a first curvature corresponding to the real trajectory point;

[0039] Determining a second curvature corresponding to the road reference point based on the third driving curve;

[0040] The running angle smoothness is calculated based on the first curvature and the second curvature.

[0041] 4. Number of times you left the road

[0042] Based on the driving data and high-precision maps, determine the actual trajectory points of the vehicle to be evaluated, the heading points passed by the vehicle to be evaluated during driving, and the time period and lane type corresponding to the driving process of the vehicle to be evaluated;

[0043] Determine the coordinates of the actual trajectory points of the vehicle to be evaluated in the high-precision map;

[0044] Based on the coordinates, the waypoints and the lane type, determine the time period of the vehicle being derailed from the time period corresponding to the driving process of the vehicle to be evaluated;

[0045] Based on the time period of de-orbiting, the number of de-orbiting times is calculated.

[0046] 5. Emergency Braking Times

[0047] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0048] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0049] The number of emergency braking times is calculated based on the acceleration corresponding to each time node and the preset braking acceleration threshold.

[0050] 6. Average acceleration

[0051] Determine the total driving time and speed change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0052] Determine the speed corresponding to each time node according to the total vehicle driving time, the preset time interval and the speed change curve;

[0053] The average acceleration is calculated based on the speed corresponding to each time node and the number of time nodes in the total vehicle travel time.

[0054] 7. Average jerk

[0055] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0056] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0057] The average jerk is calculated based on the acceleration corresponding to each time node and the number of time nodes in the total vehicle travel time.

[0058] 8. Rapid acceleration times

[0059] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0060] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0061] The number of sudden accelerations is calculated based on the acceleration corresponding to each time node and the preset acceleration threshold.

[0062] In one embodiment of the present disclosure, the second data processing module 230 is also used to: construct a lateral judgment matrix based on the lateral maneuvering evaluation parameters, each element in the lateral judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding lateral maneuvering evaluation parameters; perform a consistency check on the lateral judgment matrix; when the lateral judgment matrix passes the consistency check, determine the weight of each lateral maneuvering evaluation parameter based on the value of each element in the lateral judgment matrix; perform weighted sum processing on all lateral maneuvering evaluation parameters according to the weight of each lateral maneuvering evaluation parameter to obtain lateral maneuvering smoothness.

[0063] In one embodiment of the present disclosure, the second data processing module 230 is also used to: construct a longitudinal judgment matrix based on the longitudinal maneuvering evaluation parameters, each element in the longitudinal judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding longitudinal maneuvering evaluation parameters; perform a consistency check on the longitudinal judgment matrix; when the longitudinal judgment matrix passes the consistency check, determine the weight of each longitudinal maneuvering evaluation parameter based on the value of each element in the longitudinal judgment matrix; perform weighted summation processing on all longitudinal maneuvering evaluation parameters according to the weight of each longitudinal maneuvering evaluation parameter to obtain the longitudinal maneuvering smoothness.

[0064] In one embodiment of the present disclosure, the third data processing module 240 is also used to: construct a smoothness judgment matrix based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, each element in the smoothness judgment matrix is ​​used to indicate the parameter importance comparison result of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness; perform a consistency check on the smoothness judgment matrix; when the smoothness judgment matrix passes the consistency check, determine the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness based on the values ​​of each element in the smoothness judgment matrix; and perform weighted summation processing on the longitudinal maneuvering evaluation degree and the lateral maneuvering smoothness according to the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness to obtain the maneuvering smoothness.

[0065] According to the third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the vehicle maneuverability stability assessment method provided by the first aspect of the present disclosure.

[0066] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the vehicle maneuverability stability assessment method provided by the first aspect of the present disclosure is implemented.

[0067] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0068] Through the above-mentioned method, based on the lateral maneuvering evaluation parameters and the longitudinal maneuvering evaluation parameters, the digitization of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness are realized respectively, and then the maneuvering smoothness is realized based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, so as to realize the evaluation of the maneuvering smoothness of the vehicle when executing the automatic driving scheme. Since direction and speed are two important aspects of controlling vehicle driving, and the lateral maneuvering smoothness is used to evaluate the direction control ability of the vehicle during movement, and the longitudinal maneuvering smoothness is used to evaluate the speed control ability of the vehicle during movement, the evaluation of the vehicle maneuvering smoothness is closely related to the lateral maneuvering smoothness and the longitudinal maneuvering smoothness. Therefore, the present disclosure can make a relatively accurate evaluation of the smoothness of the vehicle when executing the automatic driving scheme to ensure the validity of the evaluation results. Improving the automatic driving scheme or the vehicle to be evaluated through the evaluation results is helpful to improve the smoothness of the automatic driving.

[0069] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0071] Figure 1 The present invention is a flow chart of a method for evaluating vehicle maneuverability according to an exemplary embodiment.

[0072] Figure 2 The present invention is a block diagram of a vehicle maneuverability assessment device according to an exemplary embodiment.

[0073] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0074] Exemplary embodiments will be described in detail below with reference to the accompanying drawings.

[0075] It should be pointed out that the relevant embodiments and drawings are only for describing exemplary embodiments provided by the present disclosure, rather than all embodiments of the present disclosure, and it should not be understood that the present disclosure is limited to the relevant exemplary embodiments.

[0076] It should be noted that the terms "first", "second", etc. used in the present disclosure are only used to distinguish different steps, devices or modules, etc. The related terms neither represent any specific technical meanings nor indicate the order or interdependence between them.

[0077] It should be noted that the modification of the term "at least one" used in the present disclosure is illustrative rather than restrictive. Unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0078] It should be noted that the term "and / or" used in this disclosure is used to describe the association relationship between associated objects, and generally indicates that there are at least three association relationships. For example, A and / or B can at least indicate the existence of three association relationships: A exists alone, A and B exist at the same time, and B exists alone.

[0079] It should be noted that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. Unless otherwise specified, the scope of the present disclosure is not limited by the order of description of the steps in the relevant embodiments.

[0080] It should be noted that all actions of acquiring signals, information or data in the present disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the device is located and with the authorization given by the owner of the corresponding device.

[0081] Exemplary Methods

[0082] Figure 1 is a flow chart of a vehicle maneuverability evaluation method according to an exemplary embodiment. Figure 1 As shown, the following steps are included.

[0083] Step 1: Obtain high-precision maps and driving data of the road section that the vehicle to be evaluated passes through, and the vehicle to be evaluated drives based on the autonomous driving solution to be tested.

[0084] In the disclosed embodiments, the autonomous driving solution can be applied to vehicles driving under various road conditions. The autonomous driving solution is pre-set in the vehicle to be evaluated so that the vehicle to be evaluated can execute the autonomous driving solution. In addition to the elements that are included in ordinary maps, high-precision maps will also contain surrounding static information related to traffic, such as street lights, guardrails, traffic lights and other elements. For road information, high-precision maps will also have richer detailed information on lanes and intersections, for example, setting waypoints on lanes.

[0085] Step 2: Determine at least one lateral maneuvering evaluation parameter and at least one longitudinal maneuvering evaluation parameter based on the high-precision map and driving data.

[0086] In the disclosed embodiment, the motion smoothness of the vehicle during driving is closely related to the lateral maneuverability and longitudinal maneuverability of the vehicle itself. Among them, the lateral maneuverability smoothness is used to evaluate the direction control ability of the vehicle during motion, and the longitudinal maneuverability smoothness is used to evaluate the speed control ability of the vehicle during motion. The lateral maneuverability evaluation parameter is used to evaluate the lateral maneuverability smoothness, and the longitudinal maneuverability evaluation parameter is used to evaluate the longitudinal maneuverability smoothness.

[0087] Step 3: Calculate the lateral maneuvering smoothness based on at least one lateral maneuvering evaluation parameter, and calculate the longitudinal maneuvering smoothness based on at least one longitudinal maneuvering evaluation parameter.

[0088] In the disclosed embodiment, after the lateral maneuvering evaluation parameters are obtained, each lateral maneuvering evaluation parameter can be normalized. Then, according to the autonomous driving scheme, the weight of each lateral maneuvering evaluation parameter is determined. For example, the autonomous driving scheme is mainly used to verify the trajectory deviation and trajectory smoothness during driving, then the weights of trajectory deviation and trajectory smoothness are higher than other parameters. In addition, when the autonomous driving scheme is for special scenarios, such as a coal truck doing reciprocating linear motion in a mine; in this process, the tester is most concerned about the trajectory deviation, and the trajectory deviation can be directly used as the lateral maneuvering evaluation parameter.

[0089] After obtaining the longitudinal maneuvering evaluation parameters, each longitudinal maneuvering evaluation parameter can be normalized. Then, according to the autonomous driving scheme, the weight of each longitudinal maneuvering evaluation parameter is determined. For example, the autonomous driving scheme is mainly used to verify the number of emergency brakes and emergency accelerations during driving, then the weights of emergency brakes and emergency accelerations are higher than other parameters. In addition, when the autonomous driving scheme is for special scenarios, such as a vehicle driving downhill on a winding mountain road; in this process, the tester is most concerned about the number of emergency brakes, and the number of emergency brakes can be directly used as the longitudinal maneuvering evaluation parameter.

[0090] Step 4: Based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, determine the maneuvering smoothness of the vehicle to be evaluated when executing the autonomous driving scheme to be tested.

[0091] In the disclosed embodiment, after the lateral maneuverability and longitudinal maneuverability are obtained, the weights of the lateral maneuverability and longitudinal maneuverability are determined according to the automatic driving scheme, and the lateral maneuverability and longitudinal maneuverability are weightedly summed based on the weights to obtain the maneuverability of the vehicle to be evaluated.

[0092] It should be understood that the weights of the lateral maneuvering evaluation parameters required for the above-mentioned lateral maneuvering smoothness calculation, the weights of the longitudinal maneuvering evaluation parameters required for the longitudinal maneuvering smoothness calculation, and the weights of the lateral maneuvering smoothness and longitudinal maneuvering smoothness required for the maneuvering smoothness calculation can be calculated by the hierarchical analysis method, or can be manually pre-set, and the corresponding preset values ​​can also be obtained in combination with the type characteristics of the autonomous driving solution, and the embodiments of the present disclosure are not limited to this.

[0093] Through the above-mentioned method, based on the lateral maneuvering evaluation parameters and the longitudinal maneuvering evaluation parameters, the digitization of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness are realized respectively, and then the maneuvering smoothness is realized based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, so as to realize the evaluation of the maneuvering smoothness of the vehicle when executing the automatic driving scheme. Since direction and speed are two important aspects of controlling vehicle driving, and the lateral maneuvering smoothness is used to evaluate the direction control ability of the vehicle during movement, and the longitudinal maneuvering smoothness is used to evaluate the speed control ability of the vehicle during movement, the evaluation of the vehicle maneuvering smoothness is closely related to the lateral maneuvering smoothness and the longitudinal maneuvering smoothness. Therefore, the present disclosure can make a relatively accurate evaluation of the smoothness of the vehicle when executing the automatic driving scheme to ensure the validity of the evaluation results. Improving the automatic driving scheme or the vehicle to be evaluated through the evaluation results is helpful to improve the smoothness of the automatic driving.

[0094] In one embodiment of the present disclosure, the lateral maneuver evaluation parameters include one or more of trajectory deviation, trajectory smoothness, running angle smoothness, and number of times off track. Among them, trajectory deviation refers to the degree of deviation between the actual driving trajectory and the planned trajectory; trajectory smoothness refers to the degree of smoothness and fluctuation corresponding to the actual trajectory of the vehicle; running angle smoothness refers to the stability of the angle change during the vehicle's movement; and the number of times off the road refers to the number of times the vehicle leaves the driving road during movement.

[0095] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes one or more of the number of emergency brakes, the number of emergency accelerations, the average acceleration, and the average jerkiness. In addition, for some special scenarios, such as winding mountain roads, the average braking deceleration can also be selected as the longitudinal maneuvering evaluation parameter. Among them, the number of emergency brakes refers to the number of times the acceleration is urgently reduced during driving based on the autonomous driving solution; the number of emergency accelerations refers to the number of times the acceleration is urgently increased during driving based on the autonomous driving solution; the average acceleration refers to the rate of change of the speed of driving on the mission section during driving based on the autonomous driving solution; the average jerkiness refers to the rate of change of the acceleration of driving on the mission section during driving based on the autonomous driving solution; the average braking deceleration is used to evaluate the control ability of the control strategy based on the autonomous driving solution for braking.

[0096] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameters include a trajectory deviation degree; the above-mentioned step 2 includes the following steps: determining a first driving curve and a second driving curve of the vehicle to be evaluated based on driving data and a high-precision map; and calculating the trajectory deviation degree based on the first driving curve and the second driving curve.

[0097] Among them, the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the second driving curve is used to indicate the planned trajectory of the vehicle to be evaluated on the high-precision map. The planned trajectory is the running trajectory set by the autonomous driving solution to be tested on the high-precision map. During the driving process of the vehicle to be evaluated, its position in each frame in the world coordinate system is recorded to obtain the discrete points of the real trajectory of the vehicle. Then, curve fitting is performed on the discrete points of the real trajectory of the vehicle and the discrete points of the planned trajectory, respectively, to obtain the curve L1 (first driving curve) corresponding to the real trajectory of the vehicle and the curve L2 (second driving curve) corresponding to the planned trajectory. Among them, when fitting the curves, the Gaussian fitting method can be used.

[0098] When calculating the trajectory deviation based on the first driving curve and the second driving curve, the integral area S1 of the region between the curves L1 and L2 can be calculated first to measure the deviation between the actual trajectory and the planned trajectory; then the distance D of the planned trajectory is calculated based on L2; finally, the trajectory deviation is represented by a unit distance trajectory deviation value, and the unit distance trajectory deviation value is calculated by S1 (deviation between the actual trajectory and the planned trajectory) / D (planned trajectory distance).

[0099] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameters include trajectory smoothness; the above-mentioned step 2 includes the following steps: determining the first driving curve and the third driving curve of the vehicle to be evaluated based on the driving data and the high-precision map; determining the trajectory smoothness based on the correlation between the first driving curve and the third driving curve.

[0100] Among them, the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map. The calculation of the first driving curve requires obtaining the discrete points of the real trajectory of the vehicle. During the driving process of the vehicle to be evaluated, its position in each frame in the world coordinate system is recorded to obtain the discrete points of the real trajectory of the vehicle. The calculation of the third driving curve requires obtaining the road reference coordinate points. First, the reference line position list is determined; then all discrete points in the real trajectory of the vehicle are determined, and a waypoint is configured for each discrete point of the real trajectory according to the coordinates and reference line position list. The configured waypoint is the road reference point. Among them, in the high-precision map, taking the starting point of the lane as the starting point, a waypoint is generated at a preset distance along the center line of the current lane until the end of the lane; the reference line position list can be obtained based on the coordinates corresponding to all waypoints on the center line of the current lane. Finally, curve fitting is performed on the discrete points of the real trajectory of the vehicle to obtain the curve L1 (first driving curve) corresponding to the real trajectory of the vehicle; curve fitting is performed on all road reference points configured for all discrete points of the real trajectory to obtain the road reference curve in the two-dimensional coordinate system; then the road reference curve in the two-dimensional coordinate system is converted into world coordinates to obtain the curve L3 (third driving curve) corresponding to the road reference curve. When curve fitting, the Gaussian fitting method can be used.

[0101] In this embodiment, the trajectory smoothness is indicated by the correlation between the first driving curve and the third driving curve. The correlation between the first driving curve and the third driving curve is indicated by the correlation coefficient between the curves L1 and L3, and the value range of the correlation coefficient is [-1, 1]. Generally, the larger the correlation coefficient, the stronger the correlation between the first driving curve and the third driving curve, which means that the actual trajectory is smoother.

[0102] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameters include running angle smoothness; the above-mentioned step 2 includes the following steps: determining the real trajectory points and road reference points of the vehicle to be evaluated based on the driving data and the high-precision map; determining the first driving curve and the third driving curve of the vehicle to be evaluated respectively based on the real trajectory points and the road reference points; determining the first curvature corresponding to the real trajectory points based on the first driving curve; determining the second curvature corresponding to the road reference points based on the third driving curve; and calculating the running angle smoothness based on the first curvature and the second curvature.

[0103] The first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map. Specifically, the first driving curve and the third driving curve can be obtained by Gaussian fitting according to the discrete points of the real trajectory (also called "real trajectory points") and the road reference points. For other descriptions of the first driving curve and the third driving curve, please refer to the above embodiment, which will not be elaborated here.

[0104] The first curvature corresponding to the real track point needs to be determined based on the shape of the real track, and the shape of the real track is represented by the first driving curve. Based on this, the first curvature calculation formula is:

[0105]

[0106] Wherein, ρ1 is the first curvature, x is the function corresponding to the first driving curve, x′ is the first derivative of x, and x″ is the second derivative of x.

[0107] The second curvature corresponding to the road reference point needs to be determined based on the road shape of the lane corresponding to the road reference point, and the road shape of the lane is represented by the third driving curve. Based on this, the second curvature calculation formula is:

[0108]

[0109] Wherein, ρ2 is the second curvature, y is the function corresponding to the third driving curve, y′ is the first derivative of y, and y″ is the second derivative of y.

[0110] Afterwards, the running angle smoothness can be calculated based on the first curvature and the second curvature. Optionally, the absolute value Δρ of the error value between the first curvature and the second curvature is calculated first; then the average value of Δρ during the driving process of the vehicle to be evaluated based on the automatic driving scheme is calculated, and the running angle smoothness is indicated by the average value. Generally, the larger the average value of Δρ, the worse the running angle smoothness.

[0111] In one embodiment of the present disclosure, the lateral maneuvering evaluation parameter includes the number of times the vehicle is off the road; the above-mentioned step 2 includes the following steps: according to the driving data and the high-precision map, determining the real trajectory points of the vehicle to be evaluated, the heading points passed by the vehicle to be evaluated during driving, and the time period and lane type corresponding to the driving process of the vehicle to be evaluated; determining the coordinates of the real trajectory points of the vehicle to be evaluated in the high-precision map; based on the coordinates, heading points and lane types, determining the time period of off-track; based on the time period of off-track, calculating the number of times the vehicle is off-track.

[0112] The waypoint is pre-set on the lane of the high-precision map; lane types are not limited to drivable lanes, shoulders, sidewalks, and non-lanes.

[0113] When the vehicle to be evaluated is driving, each real track point is recorded as a time period. With the coordinates of the real track point in the high-precision map as the center, determine whether there is a waypoint within the preset range. If there is no waypoint, it means that the vehicle to be evaluated is off the road, and the time period corresponding to the real track point is recorded as the time period off the track; if there is a waypoint, determine whether the lane type is a drivable lane. If it is not a drivable lane, it means that the vehicle to be evaluated is off the road, and the time period corresponding to the real track point is recorded as the time period off the track.

[0114] Based on the statistics of the time periods that are off the track, the number of off the track is calculated. Among them, if there are two or more time periods that are off the track are continuous time periods in all time periods corresponding to the real track points, these continuous time periods that are off the track are merged, and if the duration of the merged time period is not less than the preset duration, the number of off the track is increased by one; if the duration of the merged time period is less than the preset duration, the number of off the track is not increased. In addition, for the non-continuous time periods that are off the track in all time periods corresponding to the real track points, if the sum of the durations of all non-continuous time periods that are off the track exceeds the preset duration, the number of off the track is increased by one; otherwise, the number of off the track is not increased.

[0115] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes the number of emergency brakes; the above-mentioned step 2 includes the following steps: determining the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, the preset time interval and the acceleration change curve; calculating the number of emergency brakes based on the acceleration corresponding to each time node and the preset braking acceleration threshold.

[0116] The preset time interval is used to determine each time node. After the total vehicle travel time is divided into time nodes, the interval between two adjacent time nodes is equal to the preset time interval. Afterwards, the acceleration corresponding to each time node can be obtained from the acceleration change curve. Optionally, the calculation formula for the number of emergency brakes is:

[0117]

[0118] Where breaktimes is the number of emergency brakes, T is the number of time nodes in the total driving time of the vehicle (total number of time nodes), t is the tth time node, a t is the acceleration corresponding to the tth time node. μ1 represents the preset braking acceleration threshold, which is used to determine whether it is an emergency brake, and μ1 is less than or equal to 0. If the difference in acceleration corresponding to two adjacent time nodes is less than the preset braking acceleration threshold, the number of emergency brakes increases by one; otherwise, the number of emergency brakes does not increase.

[0119] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes an average acceleration; the above-mentioned step 2 includes the following steps: determining the total vehicle driving time and speed change curve of the vehicle to be evaluated based on the high-precision map and driving data; determining the speed corresponding to each time node based on the total vehicle driving time, the preset time interval and the speed change curve; calculating the average acceleration based on the speed corresponding to each time node and the number of time nodes in the total vehicle driving time.

[0120] The preset time interval is used to determine each time node. After the total vehicle travel time is divided into time nodes, the interval between two adjacent time nodes is equal to the preset time interval. Afterwards, the speed corresponding to each time node can be obtained from the speed change curve. Optionally, the average acceleration calculation formula is:

[0121]

[0122] Where Acc is the average acceleration, T is the number of time nodes in the total driving time of the vehicle (the total number of time nodes), t is the tth time node, and v t is the speed corresponding to the tth time node. In actual calculation, the acceleration along the front direction of the vehicle can be calculated. Generally, the smaller the average acceleration, the smoother the vehicle travels.

[0123] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameters include average jerk; the above-mentioned step 2 includes the following steps: determining the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, the preset time interval and the acceleration change curve; calculating the average jerk based on the acceleration corresponding to each time node and the number of time nodes in the total vehicle driving time.

[0124] The preset time interval is used to determine each time node. After the total vehicle travel time is divided into time nodes, the interval between two adjacent time nodes is equal to the preset time interval. Afterwards, the acceleration corresponding to each time node can be obtained from the acceleration change curve. Optionally, the calculation formula of the average jerk is:

[0125]

[0126] Where J is the average jerk, T is the number of time nodes in the total vehicle travel time (total number of time nodes), a t is the acceleration corresponding to the tth time node. In actual calculation, the jerk along the vehicle head direction can be calculated. Generally, the smaller the average jerk, the smoother the vehicle travels.

[0127] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameter includes the number of sudden accelerations; the above-mentioned step 2 includes the following steps: determining the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data; determining the acceleration corresponding to each time node based on the total vehicle driving time, the preset time interval and the acceleration change curve; calculating the number of sudden accelerations based on the acceleration corresponding to each time node and the preset acceleration threshold.

[0128] The preset time interval is used to determine each time node. After the total vehicle travel time is divided into time nodes, the interval between two adjacent time nodes is equal to the preset time interval. Afterwards, the acceleration corresponding to each time node can be obtained from the acceleration change curve. Optionally, the calculation formula for the number of rapid accelerations is:

[0129]

[0130] Among them, EmAccTimes is the number of rapid accelerations; T is the number of time nodes in the total driving time of the vehicle (the total number of time nodes), a t is the acceleration corresponding to the tth time node. μ2 represents the preset acceleration threshold, which is used to determine whether it is a sudden acceleration, and μ2 is greater than or equal to 0. If the difference between the accelerations corresponding to two adjacent time nodes is greater than the preset acceleration threshold, the number of sudden accelerations increases by one; otherwise, the number of sudden accelerations does not increase.

[0131] In one embodiment of the present disclosure, the longitudinal maneuvering evaluation parameters include an average braking deceleration; the above-mentioned step 2 includes the following steps: determining the deceleration period and the deceleration change curve within the deceleration period based on the high-precision map and driving data; determining the braking deceleration corresponding to each time node based on the deceleration period, the preset time interval and the deceleration change curve within the deceleration period; calculating the average braking deceleration based on the braking deceleration corresponding to each time node and the number of time nodes in the deceleration period.

[0132] The preset time interval is used to determine each time node. After the deceleration period of the vehicle to be evaluated is divided into time nodes, the interval between two adjacent time nodes is equal to the preset time interval. Afterwards, the braking deceleration corresponding to each time node can be obtained from the deceleration change curve corresponding to the deceleration period. Optionally, the calculation formula for the average braking deceleration is:

[0133]

[0134] Where, brake_acc is the average braking deceleration, m is the number of time nodes in the deceleration period (total number of time nodes), acc i Represents the braking deceleration of the vehicle to be evaluated at the i-th time node.

[0135] In one embodiment of the present disclosure, the above-mentioned step 3 includes the following steps: constructing a lateral judgment matrix based on the lateral maneuvering evaluation parameters; performing a consistency check on the lateral judgment matrix; when the lateral judgment matrix passes the consistency check, determining the weight of each lateral maneuvering evaluation parameter based on the value of each element in the lateral judgment matrix; performing weighted summation processing on all lateral maneuvering evaluation parameters according to the weight of each lateral maneuvering evaluation parameter to obtain the lateral maneuvering smoothness.

[0136] Among them, each element in the lateral judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding lateral maneuver evaluation parameters, that is, parameter a ij It is used to indicate the comparison result of the importance of lateral maneuver evaluation parameter i and lateral maneuver evaluation parameter j. In addition, the lateral judgment matrix satisfies:

[0137]

[0138] Among them, a ij is the quantitative value of the comparison result of the importance of parameter i and parameter j, a ji is the quantitative value of the comparison result of the importance of parameter j and parameter i, when i≠j, a ij and a ji Symmetric about the matrix diagonal. For example, the comparison results of the importance of parameter i and parameter j are shown in Table 1.

[0139] Table 1 Quantitative values ​​of parameter importance comparison results

[0140]

[0141] Afterwards, the consistency check is performed on the horizontal judgment matrix. First, the maximum eigenvalue λ of the horizontal judgment matrix is ​​determined, and then the consistency index CI is determined based on the maximum eigenvalue λ and the matrix order n. Optionally, the calculation formula of CI is:

[0142]

[0143] If CI = 0, the horizontal judgment matrix has complete consistency; if CI is close to 0, the horizontal judgment matrix has satisfactory consistency; the larger the CI, the more serious the inconsistency of the horizontal judgment matrix. When the inconsistency is too large (CI is too large), adjust a ij The quantified values ​​(i.e., the quantified values ​​in Table 1) are calculated and the consistency index is recalculated until the CI is less than the preset value. When the CI is less than the preset value, the horizontal judgment matrix passes the consistency test. ij The quantized value of determines the weight of each lateral maneuvering evaluation parameter. Finally, all lateral maneuvering evaluation parameters are weighted and summed according to the weight of each lateral maneuvering evaluation parameter to determine the lateral maneuvering smoothness.

[0144] In one embodiment of the present disclosure, the longitudinal maneuvering smoothness can also be calculated in a similar manner. That is, the above step 3 includes the following steps: constructing a longitudinal judgment matrix based on the longitudinal maneuvering evaluation parameters, each element in the longitudinal judgment matrix is ​​used to indicate the parameter importance comparison result of two corresponding longitudinal maneuvering evaluation parameters; performing a consistency check on the longitudinal judgment matrix; when the longitudinal judgment matrix passes the consistency check, determining the weight of each longitudinal maneuvering evaluation parameter based on the value of each element in the longitudinal judgment matrix; performing weighted summation processing on all longitudinal maneuvering evaluation parameters according to the weight of each longitudinal maneuvering evaluation parameter to obtain the longitudinal maneuvering smoothness. For other introductions to the calculation of longitudinal maneuvering smoothness, please refer to the above introduction to lateral maneuvering smoothness, which will not be elaborated here.

[0145] In one embodiment of the present disclosure, maneuvering smoothness can also be calculated in a similar manner. That is, the above step 4 includes the following steps: constructing a smoothness judgment matrix based on the longitudinal maneuvering smoothness and the lateral maneuvering smoothness, each element in the smoothness judgment matrix is ​​used to indicate the parameter importance comparison result of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness; performing a consistency check on the smoothness judgment matrix; when the smoothness judgment matrix passes the consistency check, determining the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness based on the values ​​of each element in the smoothness judgment matrix; performing a weighted summation process on the longitudinal maneuvering evaluation degree and the lateral maneuvering smoothness according to the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness to obtain the maneuvering smoothness. For other introductions to the calculation of maneuvering smoothness, please refer to the above-mentioned introduction to the lateral maneuvering smoothness, which will not be elaborated here.

[0146] Exemplary Devices

[0147] Figure 2 is a block diagram of a vehicle maneuverability evaluation device according to an exemplary embodiment. Figure 2 , the device includes: an acquisition module 210, a first data processing module 220, a second data processing module 230 and a third data processing module 240;

[0148] An acquisition module 210 is used to acquire a high-precision map and driving data of a road section passed by the vehicle to be evaluated, and the vehicle to be evaluated drives based on the autonomous driving solution to be tested;

[0149] A first data processing module 220, configured to determine at least one lateral maneuvering evaluation parameter and at least one longitudinal maneuvering evaluation parameter according to the high-precision map and the driving data;

[0150] The second data processing module 230 is used to: calculate a lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter, wherein the lateral maneuvering smoothness is used to evaluate the directional control ability of the vehicle during movement; calculate a longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter, wherein the longitudinal maneuvering smoothness is used to evaluate the speed control ability of the vehicle during movement;

[0151] The third data processing module 240 is used to determine the maneuvering smoothness of the vehicle to be evaluated when executing the automatic driving scheme to be tested based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness.

[0152] In one embodiment of the present disclosure, the first data processing module 220 is further used to calculate at least one evaluation parameter among the following 8 evaluation parameters.

[0153] 1. Track deviation

[0154] Determine a first driving curve and a second driving curve of the vehicle to be evaluated according to the driving data and the high-precision map, wherein the first driving curve is used to indicate the actual trajectory of the vehicle to be evaluated, and the second driving curve is used to indicate the planned trajectory of the vehicle to be evaluated on the high-precision map;

[0155] The trajectory deviation degree is calculated based on the first driving curve and the second driving curve.

[0156] 2. Trajectory smoothness

[0157] Determine a first driving curve and a third driving curve of the vehicle to be evaluated according to the driving data and the high-precision map, wherein the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map;

[0158] Based on the correlation between the first driving curve and the third driving curve, a trajectory smoothness is determined.

[0159] 3. Running angle smoothness

[0160] Determine the actual trajectory points and road reference points of the vehicle to be evaluated based on driving data and high-precision maps;

[0161] According to the real trajectory point and the road reference point, a first driving curve and a third driving curve of the vehicle to be evaluated are determined respectively, the first driving curve is used to indicate the real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate the road reference curve of the vehicle to be evaluated on the high-precision map;

[0162] Based on the first driving curve, determining a first curvature corresponding to the real trajectory point;

[0163] Determining a second curvature corresponding to the road reference point based on the third driving curve;

[0164] The running angle smoothness is calculated based on the first curvature and the second curvature.

[0165] 4. Number of times you left the road

[0166] Based on the driving data and high-precision maps, determine the actual trajectory points of the vehicle to be evaluated, the heading points passed by the vehicle to be evaluated during driving, and the time period and lane type corresponding to the driving process of the vehicle to be evaluated;

[0167] Determine the coordinates of the actual trajectory points of the vehicle to be evaluated in the high-precision map;

[0168] Based on the coordinates, the waypoints and the lane type, determine the time period of the vehicle being derailed from the time period corresponding to the driving process of the vehicle to be evaluated;

[0169] Based on the time period of de-orbiting, the number of de-orbiting times is calculated.

[0170] 5. Emergency Braking Times

[0171] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0172] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0173] The number of emergency braking times is calculated based on the acceleration corresponding to each time node and the preset braking acceleration threshold.

[0174] 6. Average acceleration

[0175] Determine the total driving time and speed change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0176] Determine the speed corresponding to each time node according to the total vehicle driving time, the preset time interval and the speed change curve;

[0177] The average acceleration is calculated based on the speed corresponding to each time node and the number of time nodes in the total vehicle travel time.

[0178] 7. Average jerk

[0179] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0180] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0181] The average jerk is calculated based on the acceleration corresponding to each time node and the number of time nodes in the total vehicle travel time.

[0182] 8. Rapid acceleration times

[0183] Determine the total vehicle driving time and acceleration change curve of the vehicle to be evaluated based on the high-precision map and driving data;

[0184] Determine the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve;

[0185] The number of sudden accelerations is calculated based on the acceleration corresponding to each time node and the preset acceleration threshold.

[0186] In one embodiment of the present disclosure, the second data processing module 230 is further configured to:

[0187] A lateral judgment matrix is ​​constructed based on the lateral maneuver evaluation parameters, wherein each element in the lateral judgment matrix is ​​used to indicate a comparison result of the parameter importance of two corresponding lateral maneuver evaluation parameters;

[0188] Conduct consistency checks on the horizontal judgment matrix;

[0189] When the lateral judgment matrix passes the consistency test, the weight of each lateral maneuver evaluation parameter is determined based on the value of each element in the lateral judgment matrix;

[0190] All lateral maneuvering evaluation parameters are weighted and summed according to the weights of each lateral maneuvering evaluation parameter to obtain the lateral maneuvering smoothness.

[0191] In one embodiment of the present disclosure, the second data processing module 230 is further configured to:

[0192] A longitudinal judgment matrix is ​​constructed based on the longitudinal maneuver evaluation parameters, wherein each element in the longitudinal judgment matrix is ​​used to indicate a comparison result of parameter importance of two corresponding longitudinal maneuver evaluation parameters;

[0193] Conduct consistency checks on the longitudinal judgment matrix;

[0194] When the longitudinal judgment matrix passes the consistency test, the weight of each longitudinal maneuver evaluation parameter is determined based on the value of each element in the longitudinal judgment matrix;

[0195] All longitudinal maneuvering evaluation parameters are weighted and summed according to the weights of each longitudinal maneuvering evaluation parameter to obtain the longitudinal maneuvering smoothness.

[0196] In one embodiment of the present disclosure, the third data processing module 240 is further configured to:

[0197] Based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, a smoothness judgment matrix is ​​constructed, and each element in the smoothness judgment matrix is ​​used to indicate the comparison result of the importance of the parameters of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness;

[0198] Conduct consistency check on the stationarity judgment matrix;

[0199] When the stability judgment matrix passes the consistency test, the weights of the longitudinal maneuvering stability and the lateral maneuvering stability are determined based on the values ​​of each element in the stability judgment matrix;

[0200] According to the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness, the longitudinal maneuvering evaluation degree and the lateral maneuvering smoothness are weighted and summed to obtain the maneuvering smoothness.

[0201] The exemplary device is an embodiment of the device corresponding to the above exemplary method. The specific operations of each module can be understood by referring to the description of the method embodiment, which will not be repeated here.

[0202] Exemplary Electronic Devices

[0203] Figure 3 1 is a block diagram of an electronic device 300 according to an exemplary embodiment. The electronic device 300 may be a computer device, a notebook computer, a server, a vehicle controller, a vehicle terminal, a vehicle computer or other types of electronic devices.

[0204] Reference Figure 3 The electronic device 300 may include at least one processor 310 and a memory 320. The processor 310 may execute instructions stored in the memory 320. The processor 310 is communicatively connected to the memory 320 via a data bus. In addition to the memory 320, the processor 310 may also be communicatively connected to an input device 330, an output device 340, and a communication device 350 via a data bus.

[0205] The processor 310 may be any conventional processor, such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), or a combination thereof.

[0206] The memory 320 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0207] In the disclosed embodiment, executable instructions are stored in the memory 320, and the processor 310 can read the executable instructions from the memory 320 and execute the instructions to implement all or part of the steps of the vehicle maneuverability stability evaluation method in the above exemplary embodiment.

[0208] Exemplary computer-readable storage media

[0209] In addition to the above methods and devices, exemplary embodiments of the present disclosure also include a computer program product or a computer-readable storage medium storing the computer program product. The computer product includes computer program instructions that can be executed by a processor to implement all or part of the steps described in the above exemplary embodiments.

[0210] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and also conventional procedural programming languages, such as "C" language or similar programming languages ​​and scripting languages ​​(e.g., Python). The program code may be executed entirely on the user computing device, partially on the user computing device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0211] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples of readable storage media include: static random access memory (SRAM) with one or more wires electrically connected, electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk, or any suitable combination of the above.

[0212] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. The present application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and embodiments are intended to be exemplary only, and the present disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes may be made without departing from the scope thereof.

Claims

1. A vehicle maneuverability assessment method, characterized in that: include: Obtaining high-precision maps and driving data of a road section passed by a vehicle to be evaluated, wherein the vehicle to be evaluated is traveling based on the autonomous driving solution to be tested; According to the high-precision map and the driving data, at least one lateral maneuvering evaluation parameter and at least one longitudinal maneuvering evaluation parameter are determined, wherein the lateral maneuvering evaluation parameter includes one or more of trajectory smoothness and running angle smoothness; wherein, The determining of trajectory smoothness according to the high-precision map and the driving data includes: Determine, according to the driving data and the high-precision map, a first driving curve and a third driving curve of the vehicle to be evaluated, wherein the first driving curve is used to indicate a real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate a road reference curve of the vehicle to be evaluated on the high-precision map; Determining the trajectory smoothness based on the correlation between the first driving curve and the third driving curve; and determining the running angle smoothness according to the high-precision map and the driving data, including: Determining the actual trajectory points and road reference points of the vehicle to be evaluated according to the driving data and the high-precision map; Determining a first driving curve and a third driving curve of the vehicle to be evaluated respectively according to the real trajectory point and the road reference point; Determining a first curvature corresponding to the real trajectory point based on the first driving curve; Determining a second curvature corresponding to the road reference point based on the third driving curve; The running angle smoothness is calculated based on the first curvature and the second curvature; and the calculation process of the third driving curve includes: Get the reference line position list based on the coordinates corresponding to all heading points on the current lane centerline; Determine all discrete points in the real trajectory of the vehicle, and configure a heading point for each discrete point of the real trajectory according to the coordinates and reference line position list, wherein the configured heading point is a road reference point; Performing curve fitting on all road reference points configured for all discrete points of the real trajectory to obtain a road reference curve in a two-dimensional coordinate system; Converting the road reference curve in the two-dimensional coordinate system into a world coordinate to obtain a third driving curve corresponding to the road reference curve; Calculating a lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter, the lateral maneuvering smoothness being used to evaluate a directional control capability of the vehicle during motion; calculating a longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter, the longitudinal maneuvering smoothness being used to evaluate a speed control capability during vehicle motion; The maneuverability smoothness of the vehicle to be evaluated when executing the automatic driving scheme to be tested is determined based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness.

2. The method according to claim 1, characterized in that The lateral maneuver assessment parameters also include trajectory deviation; The determining of the trajectory deviation according to the high-precision map and the driving data includes: Determine, according to the driving data and the high-precision map, a first driving curve and a second driving curve of the vehicle to be evaluated, wherein the first driving curve is used to indicate a real trajectory of the vehicle to be evaluated, and the second driving curve is used to indicate a planned trajectory of the vehicle to be evaluated on the high-precision map; The trajectory deviation degree is calculated based on the first driving curve and the second driving curve.

3. The method according to claim 1, characterized in that The lateral maneuver assessment parameters also include the number of times off the road; The determining the number of times of leaving the road according to the high-precision map and the driving data includes: Determine, based on the driving data and the high-precision map, the actual trajectory points of the vehicle to be evaluated, the heading points passed by the vehicle to be evaluated during driving, and the time period and lane type corresponding to the driving process of the vehicle to be evaluated; Determine the coordinates of the real trajectory point of the vehicle to be evaluated in the high-precision map; Determine, based on the coordinates, the heading point and the lane type, a time period of derailment from a time period corresponding to the driving process of the vehicle to be evaluated; The number of road departures is calculated based on the time period of the track departure.

4. The method according to claim 1, characterized in that: The longitudinal maneuver assessment parameters include the number of emergency brakes; The determining, according to the high-precision map and the driving data, at least one longitudinal maneuver assessment parameter comprises: Determining a total vehicle travel time and an acceleration change curve of the vehicle to be evaluated according to the high-precision map and the driving data; Determining the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve; The number of emergency braking times is calculated based on the acceleration corresponding to each of the time nodes and a preset braking acceleration threshold.

5. The method according to claim 1, characterized in that The longitudinal maneuver assessment parameters include average acceleration; The determining, according to the high-precision map and the driving data, at least one longitudinal maneuver assessment parameter comprises: Determining a total vehicle travel time and a speed change curve of the vehicle to be evaluated according to the high-precision map and the driving data; Determine the speed corresponding to each time node according to the total vehicle travel time, the preset time interval and the speed change curve; The average acceleration is calculated based on the speed corresponding to each of the time nodes and the number of time nodes in the total vehicle driving time.

6. The method according to claim 1, characterized in that The longitudinal maneuver assessment parameters include average jerk; The determining, according to the high-precision map and the driving data, at least one longitudinal maneuver assessment parameter comprises: Determining a total vehicle travel time and an acceleration change curve of the vehicle to be evaluated according to the high-precision map and the driving data; Determining the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve; The average jerk is calculated based on the acceleration corresponding to each of the time nodes and the number of time nodes in the total vehicle travel time.

7. The method according to claim 1, characterized in that The longitudinal maneuver assessment parameters include the number of rapid accelerations; The determining, according to the high-precision map and the driving data, at least one longitudinal maneuver assessment parameter comprises: Determining a total vehicle travel time and an acceleration change curve of the vehicle to be evaluated according to the high-precision map and the driving data; Determining the acceleration corresponding to each time node according to the total vehicle driving time, the preset time interval and the acceleration change curve; The number of rapid accelerations is calculated based on the acceleration corresponding to each of the time nodes and a preset acceleration threshold.

8. The method according to any one of claims 1 to 7, characterized in that: The calculating the lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter comprises: Constructing a lateral judgment matrix based on the lateral maneuver evaluation parameters, wherein each element in the lateral judgment matrix is ​​used to indicate a comparison result of parameter importance of two corresponding lateral maneuver evaluation parameters; Performing consistency check on the horizontal judgment matrix; When the lateral judgment matrix passes the consistency check, determining the weight of each lateral maneuver evaluation parameter based on the value of each element in the lateral judgment matrix; All the lateral maneuvering evaluation parameters are weighted and summed according to the weight of each lateral maneuvering evaluation parameter to obtain the lateral maneuvering smoothness.

9. The method according to any one of claims 1 to 7, characterized in that: The calculating the longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter comprises: constructing a longitudinal judgment matrix based on the longitudinal maneuver evaluation parameters, wherein each element in the longitudinal judgment matrix is ​​used to indicate a parameter importance comparison result of two corresponding longitudinal maneuver evaluation parameters; Performing a consistency check on the longitudinal judgment matrix; When the longitudinal judgment matrix passes the consistency check, determining the weight of each longitudinal maneuver evaluation parameter based on the value of each element in the longitudinal judgment matrix; All the longitudinal maneuvering evaluation parameters are weighted and summed according to the weight of each longitudinal maneuvering evaluation parameter to obtain the longitudinal maneuvering smoothness.

10. The method according to any one of claims 1 to 7, characterized in that: The determining the maneuverability of the vehicle to be evaluated according to the lateral maneuverability and the longitudinal maneuverability comprises: Based on the lateral maneuvering smoothness and the longitudinal maneuvering smoothness, constructing a smoothness judgment matrix, wherein each element in the smoothness judgment matrix is ​​used to indicate a comparison result of the importance of parameters of the lateral maneuvering smoothness and the longitudinal maneuvering smoothness; Performing a consistency check on the stability judgment matrix; When the stability judgment matrix passes the consistency check, determining the weights of the longitudinal maneuvering stability and the lateral maneuvering stability based on the values ​​of each element in the stability judgment matrix; According to the weights of the longitudinal maneuvering smoothness and the lateral maneuvering smoothness, weighted sum processing is performed on the longitudinal maneuvering evaluation degree and the lateral maneuvering smoothness to obtain the maneuvering smoothness.

11. A vehicle maneuverability assessment device, characterized in that: include: An acquisition module, a first data processing module, a second data processing module and a third data processing module; The acquisition module is used to acquire high-precision maps and driving data of a road section passed by a vehicle to be evaluated, wherein the vehicle to be evaluated is traveling based on the autonomous driving solution to be tested; The first data processing module is used to determine at least one lateral maneuvering evaluation parameter and at least one longitudinal maneuvering evaluation parameter according to the high-precision map and the driving data; the lateral maneuvering evaluation parameter includes one or more of trajectory smoothness and running angle smoothness; wherein, The determining of trajectory smoothness according to the high-precision map and the driving data includes: Determine, according to the driving data and the high-precision map, a first driving curve and a third driving curve of the vehicle to be evaluated, wherein the first driving curve is used to indicate a real trajectory of the vehicle to be evaluated, and the third driving curve is used to indicate a road reference curve of the vehicle to be evaluated on the high-precision map; Determining the trajectory smoothness based on the correlation between the first driving curve and the third driving curve; and determining the running angle smoothness according to the high-precision map and the driving data, including: Determining the actual trajectory points and road reference points of the vehicle to be evaluated according to the driving data and the high-precision map; Determining a first driving curve and a third driving curve of the vehicle to be evaluated respectively according to the real trajectory point and the road reference point; Determining a first curvature corresponding to the real trajectory point based on the first driving curve; Determining a second curvature corresponding to the road reference point based on the third driving curve; The running angle smoothness is calculated based on the first curvature and the second curvature; and the calculation process of the third driving curve includes: Get the reference line position list based on the coordinates corresponding to all heading points on the current lane centerline; Determine all discrete points in the real trajectory of the vehicle, and configure a heading point for each discrete point of the real trajectory according to the coordinates and reference line position list, wherein the configured heading point is a road reference point; Performing curve fitting on all road reference points configured for all discrete points of the real trajectory to obtain a road reference curve in a two-dimensional coordinate system; Converting the road reference curve in the two-dimensional coordinate system into a world coordinate to obtain a third driving curve corresponding to the road reference curve; The second data processing module is used to calculate a lateral maneuvering smoothness based on the at least one lateral maneuvering evaluation parameter, wherein the lateral maneuvering smoothness is used to evaluate the directional control ability of the vehicle during movement; and calculate a longitudinal maneuvering smoothness based on the at least one longitudinal maneuvering evaluation parameter, wherein the longitudinal maneuvering smoothness is used to evaluate the speed control ability of the vehicle during movement; The third data processing module is used to determine the maneuverability smoothness of the vehicle to be evaluated when executing the automatic driving scheme to be tested based on the lateral maneuverability smoothness and the longitudinal maneuverability smoothness.

12. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the vehicle maneuverability assessment method according to any one of claims 1 to 10 is implemented.

13. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed, the vehicle maneuverability assessment method according to any one of claims 1 to 10 is implemented.

Images