Intelligent networked automobile automatic driving performance evaluation method based on simulation test

Abstract

The invention relates to the technical field of intelligent automobile test and evaluation, in particular to an intelligent networked automobile automatic driving performance evaluation method based on simulation test, which comprises the following steps: S1, generating a reference steady-state traffic flow in a simulation platform; s2, injecting traffic flow disturbance parameters into the reference steady-state traffic flow, and recording whole-course operation data of the reference steady-state traffic flow; s3, extracting performance indexes of the tested vehicle; s4, calculating a deviation degree, and quantifying the deviation degree as a performance attenuation rate; s5, performing weighted fusion to obtain a comprehensive disturbance tolerance index; and S6, outputting an evaluation conclusion of the automatic driving performance of the tested vehicle according to the comprehensive disturbance tolerance index. According to the method, the reference traffic flow is constructed, the disturbance is injected, the performance attenuation rate is quantified, and then the disturbance tolerance index is generated through weighting, so that objective, comparable and interpretable comprehensive evaluation of the automatic driving performance of the intelligent networked automobile is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent vehicle test evaluation, and particularly relates to an intelligent connected vehicle automatic driving performance evaluation method based on simulation test. BACKGROUND

[0002] With the development of intelligent connected vehicles, the application of automatic driving systems in highway, urban road and mixed traffic scenarios is increasingly widespread; in order to ensure the safety and reliability of automatic driving vehicles, their performance in complex traffic environments must be comprehensively evaluated; traditional real vehicle road tests have problems such as high cost, long cycle, poor repeatability and high safety risk, so more and more research is turning to virtual testing methods based on simulation platforms.

[0003] However, existing simulation test evaluation methods mostly stay at the qualitative analysis level, lack a unified quantitative index system, and especially when facing traffic flow disturbances, how to accurately extract key performance indicators such as vehicle longitudinal acceleration and lateral acceleration, and further measure their differences from the baseline state, has not yet formed a systematic evaluation method; at the same time, existing research often only outputs a single performance indicator, which is difficult to reflect the overall disturbance adaptability and stability of the vehicle. Therefore, there is an urgent need for an intelligent connected vehicle automatic driving performance evaluation method based on simulation test to solve the above problems. SUMMARY

[0004] Based on the above purpose, the present application provides an intelligent connected vehicle automatic driving performance evaluation method based on simulation test.

[0005] The intelligent connected vehicle automatic driving performance evaluation method based on simulation test comprises the following steps: S1: generating a baseline steady traffic flow in a simulation platform, and embedding a to-be-evaluated intelligent connected vehicle as a measured vehicle in the baseline steady traffic flow; S2: injecting traffic flow disturbance parameters into the baseline steady traffic flow to convert it into a dynamic disturbance traffic flow, and controlling the measured vehicle to automatically drive in the dynamic disturbance traffic flow and recording its whole running data; S3: extracting performance indicators of the measured vehicle from the whole running data, including longitudinal acceleration and lateral acceleration; S4: for each performance indicator, calculating the degree of deviation between it in the dynamic disturbance traffic flow and the baseline value in the baseline steady traffic flow, and quantifying the degree of deviation as a performance decay rate; S5: based on the performance decay rates of all performance indicators, weighting and fusing according to preset index weight coefficients to obtain a comprehensive disturbance tolerance index; S6: outputting an evaluation conclusion of the automatic driving performance of the measured vehicle according to the numerical range of the comprehensive disturbance tolerance index.

[0006] Optionally, the S1 specifically comprises: S11: setting road structure, traffic signal and vehicle generation parameters in the simulation platform, initializing the flow, speed and type distribution of background vehicles, and constructing a basic traffic network environment; S12: continuously injecting background vehicles based on a microscopic simulation model, and monitoring overall vehicle speed fluctuations until the mean vehicle speed and vehicle distance distribution tend to be stable, thereby generating a baseline steady-state traffic flow; S13: selecting a position of a middle background vehicle in the baseline steady-state traffic flow, replacing and embedding the intelligent connected vehicle to be evaluated, and inheriting the initial state parameters of the vehicle as the simulation starting point.

[0007] Optionally, the S2 specifically comprises: S21: setting a set of disturbance parameters in the simulation platform, the set of disturbance parameters including lane-changing interference density, sudden acceleration and deceleration probability, and forward vehicle insertion frequency, and applying the set of disturbance parameters to the behavior rules of part of the background vehicles in the baseline steady-state traffic flow; S22: based on a disturbance triggering condition, starting a disturbance injection process to simulate vehicle sudden acceleration, sudden deceleration or frequent lane-changing behavior in a local road section or at a random position, thereby breaking the original steady-state flow state and constructing a dynamic disturbance traffic flow; S23: starting the automatic driving control program of the vehicle under test, taking over the lateral and longitudinal behavior execution according to the preset control strategy, and maintaining real-time interaction with the dynamic disturbance traffic flow, recording the speed, acceleration, path and behavior decision data of the vehicle under test throughout the journey, and forming a complete operation data sequence.

[0008] Optionally, the S3 specifically comprises: S31: during the simulation running of the vehicle under test, based on the data acquisition interface of the simulation platform, collecting the original running data of the speed, heading angle and time stamp of the vehicle under test at a fixed sampling period; S32: calculating the speed increment between adjacent time points according to the speed change at consecutive time points, and extracting the longitudinal acceleration sequence in combination with the time interval; S33: extracting the heading angle change rate per unit time according to the heading angle change trend at consecutive time points, and constructing a lateral acceleration sequence accordingly; S34: time-synchronizing and formatting the extracted longitudinal and lateral acceleration sequences to generate a complete performance index time sequence.

[0009] Optionally, the S32 specifically comprises: S321: based on the simulation platform, collecting the speed data sequence of the vehicle under test at a preset sampling period and recording the speed data of two consecutive time points and the longitudinal velocity value at the time interval and ; S322: calculating the velocity increment between adjacent time intervals , the formula being: wherein, represents the velocity change amount in the time interval ; S323: dividing the velocity increment by the corresponding time interval , extracting the longitudinal acceleration in the current time period, the formula being: wherein, represents the longitudinal acceleration value in the i-th sampling interval.

[0010] Optionally, the S33 specifically comprises: S331: in the simulation running process, collecting the heading angle data of the measured vehicle at a preset sampling period , recording the heading angle values at two adjacent time points and ; and ; S332: calculating the heading angle change rate in unit time , the formula being: wherein, represents the angular velocity of the vehicle in the i-th time interval; represents the heading angle of the vehicle at the time point ; S333: combining the longitudinal velocity of the vehicle at the current time point , calculating the lateral acceleration, the formula being: wherein, represents the lateral acceleration value at the i-th time point.

[0011] Optionally, the S4 specifically comprises: S41: collecting the performance index time sequence of the measured vehicle under the dynamic disturbance traffic flow, including the longitudinal acceleration sequence and the lateral acceleration sequence ; S42: constructing a benchmark steady-state traffic flow environment under the same simulation platform, and running the measured vehicle under the same control strategy to obtain the corresponding performance index sequence of the vehicle under the steady-state condition and ; S43: for each performance index, performing difference calculation according to the time point to obtain a performance deviation sequence; ​​​S44: taking the mean square error of each deviation sequence as the disturbance response amplitude, and introducing the mean square value of the reference performance sequence for normalization processing, respectively calculating the longitudinal and lateral performance attenuation rates With .

[0012] Optionally, the S5 specifically comprises: S51: setting a performance index weighted fusion rule, and constructing an index weight coefficient set containing the weights of each performance index , wherein and respectively represent the preset weights corresponding to the longitudinal acceleration performance index and the lateral acceleration performance index; S52: calling the calculated longitudinal performance attenuation rate and the lateral performance attenuation rate , and weighting and fusing them with the corresponding index weights to calculate a comprehensive disturbance tolerance index .

[0013] Optionally, the performance index weighted fusion rule comprises: Rule 1: for the scenario where the longitudinal and lateral control performances of the autonomous vehicle are equally valued, the weight coefficient is set to , ; Rule 2: applicable to high-speed driving, straight lane or ramp following scenarios, with the longitudinal stability as the core, the weight coefficient is set to , 3; Rule 3: applicable to urban roads, lane changing or traffic dense areas, with the lateral control ability as the core, the weight coefficient is set to 3, 7.

[0014] Optionally, the S6 specifically comprises: S61: setting an evaluation grading rule of the disturbance tolerance index, dividing the numerical range of the comprehensive disturbance tolerance index into multiple performance grade intervals, and binding it with the preset performance grade labels in a mapping relationship; S62: matching the comprehensive disturbance tolerance index calculated by the measured vehicle under the dynamic disturbance traffic flow with the preset interval to determine its corresponding performance grade label; S63: outputting the corresponding evaluation conclusion, including the performance grade, evaluation explanation and optimization suggestion.

[0015] The beneficial effects of the present application are: The application can reproduce complex and changeable traffic scenes in a controllable environment by constructing a benchmark steady traffic flow in a simulation platform and injecting multiple types of disturbance parameters into the simulation platform; and the longitudinal acceleration and lateral acceleration data of the measured vehicle are collected to establish a complete performance index sequence, thereby providing a basis for quantitative analysis of performance differences.

[0016] The application calculates the performance attenuation rate under the disturbance condition and the benchmark condition, introduces a preset weight for weighted fusion, obtains a comprehensive disturbance tolerance index, and thereby realizes unified quantitative evaluation of the automatic driving system under different disturbance environments. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only illustrate the application, and other drawings can also be obtained by those skilled in the art without creative effort.

[0018] Fig. 1 The figure is a schematic diagram of the automobile automatic driving performance evaluation method of the embodiment of the application. Fig. 2 The figure is a schematic diagram of the process of extracting the performance index of the measured vehicle. DETAILED DESCRIPTION

[0019] The application will be described in detail below in combination with the drawings and specific embodiments. It should be noted that, in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and other alternative ways can also be used by those skilled in the art to implement; and the drawings are only used to more specifically describe the embodiments, and are not intended to specifically limit the application.

[0020] It should be noted that, in the specification, "one embodiment", "embodiment", "exemplary embodiment", "some embodiments" and the like indicate that the described embodiments can include specific features, structures or characteristics, but not necessarily every embodiment includes the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in combination with an embodiment, it should be within the knowledge of those skilled in the art to realize this feature, structure or characteristic in combination with other embodiments (whether or not explicitly described).

[0021] In general, terms can be understood, at least partly, from usage in context. For example, depending at least in part on context, the term "one or more" as used herein can describe any feature, structure, or characteristic in the singular or can describe combinations of features, structures, or characteristics, in the plural, both singular and plural. Further, the term "based on" can be understood as not necessarily being confined to factors that are explicitly enumerated, but rather can include additional factors not explicitly enumerated, depending at least in part on context.

[0022] As shown in Figs. 1-2 The intelligent connected vehicle automatic driving performance evaluation method based on simulation test comprises the following steps: S1: generating a benchmark steady traffic flow in a simulation platform, and embedding the intelligent connected vehicle to be evaluated as a measured vehicle in the benchmark steady traffic flow; S2: injecting a traffic flow disturbance parameter into the benchmark steady traffic flow to convert it into a dynamic disturbance traffic flow, and controlling the measured vehicle to automatically drive in the dynamic disturbance traffic flow and recording its whole running data; S3: extracting performance indicators of the measured vehicle from the whole running data, including longitudinal acceleration and lateral acceleration; S4: for each performance indicator, calculating the degree of deviation between it under the dynamic disturbance traffic flow and the benchmark value under the benchmark steady traffic flow, and quantifying the degree of deviation as a performance attenuation rate; S5: based on the performance attenuation rates of all performance indicators, weighting and fusing according to preset indicator weight coefficients to obtain a comprehensive disturbance tolerance index; S6: according to the numerical range of the comprehensive disturbance tolerance index, outputting an evaluation conclusion on the automatic driving performance of the measured vehicle.

[0023] S1 specifically comprises: S11: setting road structure, traffic signal and vehicle generation parameters in the simulation platform, initializing the flow, speed and type distribution of background vehicles, and constructing a basic traffic network environment; S12: based on a microscopic simulation model, continuously injecting background vehicles and monitoring overall vehicle speed fluctuations until the mean vehicle speed and distance distribution tend to be stable, generating a benchmark steady traffic flow; S13: selecting a position of a middle background vehicle in the benchmark steady traffic flow, replacing and embedding the intelligent connected vehicle to be evaluated, and inheriting the initial state parameters of the vehicle as the simulation starting point; the above steps provide a unified reference benchmark for subsequent disturbance injection and performance comparison by constructing a stable and controllable simulation traffic flow environment and embedding the measured vehicle without disturbance, thereby improving the effectiveness and repeatability of the evaluation.

[0024] S2 specifically comprises: S21: Set the perturbation parameter set in the simulation platform, the perturbation parameter set includes lane-changing interference density, sudden acceleration and deceleration probability and forward vehicle insertion frequency, and apply it to the part of the background vehicle behavior rule of the reference steady-state traffic flow; S22: Based on the perturbation trigger condition, start the perturbation injection process, simulate the vehicle sudden acceleration, sudden deceleration or frequent lane-changing behavior in the local road section or random position, so as to break the original steady-state flow state and build a dynamic perturbation traffic flow; S23: Start the automatic driving control program of the measured vehicle, take over the lateral and longitudinal behavior execution according to the preset control strategy, and keep real-time interaction with the dynamic perturbation traffic flow, record the speed, acceleration, path and behavior decision data of the whole process, and form a complete running data sequence; the above steps can restore the dynamic response process under complex traffic change conditions by injecting multiple types of perturbation behaviors and controlling the automatic driving operation of the measured vehicle in the interference scene, which can provide effective support for subsequent performance deviation analysis and tolerance evaluation.

[0025] In the simulation process, the following perturbation parameter set table 1 is configured to perform perturbation injection operation on the reference steady-state traffic flow; Table 1 perturbation parameter set table Through the setting of the above table 1, the behavior mode of the background vehicle can be dynamically adjusted in the simulation platform, so that the traffic flow is transformed from steady state to non-steady state, thereby building a dynamic perturbation traffic flow environment.

[0026] S3 specifically includes: S31: In the simulation running process of the measured vehicle, based on the data acquisition interface of the simulation platform, the original running data of the speed, heading angle and time stamp of the measured vehicle is collected at a fixed sampling period; S32: According to the speed change at consecutive time points, the speed increment between adjacent time points is calculated, and the longitudinal acceleration sequence is extracted combined with the time interval, which is used to reflect the forward response characteristics of the vehicle; S33: According to the heading angle change trend at consecutive time points, the heading angle change rate per unit time is extracted, and the lateral acceleration sequence is constructed accordingly, which is used to depict the lateral control response ability of the vehicle; S34: The longitudinal and lateral acceleration sequences extracted are time-synchronized and formatted, and a complete performance index time sequence is generated, which is used as input data for subsequent performance deviation calculation and tolerance analysis; through the above steps, the longitudinal and lateral response data of the measured vehicle in the dynamic traffic environment can be accurately extracted, reflecting the real-time performance of the automatic driving system in acceleration control and path keeping, providing basic data support for performance decay rate evaluation.

[0027] S32 specifically includes: S321: Based on the simulation platform with a preset sampling period Collect speed data sequences of the vehicle under test and record two consecutive moments. and longitudinal velocity value and ; S322: Calculate the velocity increment between adjacent time moments The formula is: ,in, Indicates the time interval The change in velocity within, in units of ; S323: Divide the speed increment by the corresponding time interval. Extract the longitudinal acceleration within the current time period using the following formula: ,in, Indicates the first The longitudinal acceleration values ​​within each sampling interval, in units of By using the above-mentioned method based on speed difference and time interval calculation, a longitudinal acceleration sequence with high timeliness and controllable accuracy can be obtained, which reflects the acceleration control performance of the tested vehicle under dynamic disturbance conditions and provides a reliable basis for subsequent performance degradation quantification.

[0028] S33 specifically includes: S331: During the simulation, a preset sampling period is used. Collect the heading angle data of the vehicle under test and record it at two consecutive moments. and The heading angle below and ; S332: Calculate the rate of change of heading angle per unit time. The formula is: ,in, Indicates the vehicle is in angular velocity within a time interval, in units of ; Indicates the vehicle's time The heading angle, in radians; S333: Combined with the vehicle's current longitudinal velocity The formula for calculating lateral acceleration is: ,in, Indicates the first The lateral acceleration value at time t, in units of This is used to reflect the dynamic response capability of a vehicle in lateral handling in turbulent traffic environments. The above steps extract lateral acceleration by coupling the rate of change of heading angle with real-time speed, which can accurately reflect the lateral motion characteristics of the vehicle during lane changes, deviations, or emergency maneuvers, and improve the completeness of performance indicators and the sensitivity of evaluation.

[0029] S4 specifically includes: S41: Collect time series of performance indicators of the tested vehicle under dynamic traffic disturbance, including longitudinal acceleration series. With lateral acceleration sequence ,in and They represent the first The longitudinal and lateral acceleration values ​​at each sampling point, in units of ; S42: Construct a baseline steady-state traffic flow environment on the same simulation platform, and run the test vehicle with the same control strategy to obtain its corresponding performance index sequence under steady-state conditions. and ,in and The first The acceleration values ​​of each sampling point under reference conditions; S43: For each performance indicator, calculate the difference at each time point to obtain the performance deviation sequence. The calculation formula is as follows: ,in, and They represent the first The longitudinal and lateral acceleration deviation values ​​at each moment; S44: Using the mean square error of each deviation sequence as the perturbation response amplitude, and normalizing it by introducing the mean square value of the reference performance sequence, calculate the longitudinal and lateral performance degradation rates respectively. and The formula is: ; ,in, This represents the total number of sampling points for the performance index. These represent the longitudinal and lateral performance degradation rates, respectively. A larger value indicates a higher degree of performance deviation. The above steps, by introducing the normalized mean square deviation calculation method, realize the quantitative evaluation of the longitudinal and lateral control performance of vehicles under different traffic disturbances, so that the performance degradation trend has a unified measurement scale, which is convenient for subsequent cross-scenario comparison and evaluation.

[0030] S5 specifically includes: S51: Define the weighted fusion rules for performance metrics and construct a set of metric weight coefficients that includes the weights of each performance metric. ,in and respectively represent the preset weight corresponding to the longitudinal acceleration performance index and the lateral acceleration performance index, satisfying ; S52: call the calculated longitudinal performance decay rate and the lateral performance decay rate , and weight and integrate them with the corresponding index weight to calculate the comprehensive disturbance tolerance index , the calculation formula of which is: , wherein represents the comprehensive disturbance tolerance index of the measured vehicle in the current dynamic disturbance traffic environment, and the larger the value, the more obvious the vehicle performance decay and the lower the disturbance tolerance ability; the above steps integrate the multi-dimensional performance decay index into a single disturbance tolerance index by introducing a weighted integration mechanism, which can realize quantitative evaluation of the overall performance of different types of autonomous vehicles in complex traffic environments, and improve the comparability and practicality of the evaluation results.

[0031] The performance index weighted integration rule includes: Rule 1, for scenarios where the longitudinal and lateral control performance of the autonomous vehicle is equally important, the weight coefficient is set to , ; Rule 2: applicable to high-speed driving, straight lane or ramp following scenarios, with longitudinal stability as the core, the weight coefficient is set to , 3; Rule 3, applicable to urban roads, lane changes or traffic-intensive areas, with lateral control ability as the core, the weight coefficient is set to 3, 7; the above integration rules can be selectively enabled according to the configuration parameters of the evaluation platform, or the disturbance tolerance index can be compared according to different rules in batch simulation to realize flexible adaptation of evaluation strategies and analysis of index diversity.

[0032] S6 specifically includes: S61: set the evaluation grading rule of the disturbance tolerance index, divide the numerical range of the comprehensive disturbance tolerance index into multiple performance level intervals, and bind it with the preset performance level label through a mapping relationship; Table 2 evaluation grading rule table The above table 1 divides the numerical range of the disturbance tolerance index R into five performance levels, effectively realizing quantitative grading evaluation of the automatic driving performance of the measured vehicle in the dynamic disturbance traffic environment; this rule can convert complex calculation results into intuitive level labels, facilitating developers to quickly identify the stability and robustness of the control system, and assisting in system tuning and strategy optimization; S62: The comprehensive disturbance tolerance index of the measured vehicle calculated under the dynamic disturbance traffic flow is matched with a preset interval to determine a corresponding performance level label; S63: A corresponding evaluation conclusion is output, including a performance level, an evaluation explanation and an optimization suggestion, for representing the comprehensive performance level of the automatic driving control strategy under the current disturbance condition; the above steps make the anti-disturbance capability evaluation result of the automatic driving system more interpretable and usable by constructing a grade division rule of the disturbance tolerance index and mapping the numerical result into an intuitive performance level label, facilitating subsequent control strategy optimization or vehicle comparison analysis.

[0033] The present application encompasses any alternatives, modifications, equivalent methods and schemes made on the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can also be fully understood without the description of these details by those skilled in the art. In addition, in order to avoid unnecessary confusion to the essence of the present application, well-known methods, processes, procedures, elements and circuits, etc. are not described in detail.

[0034] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principle of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method for evaluating performance of automatic driving of an intelligent connected vehicle based on simulation test, characterized in that, The method comprises the following steps: S1: generating a benchmark steady traffic flow in a simulation platform, and embedding a smart connected vehicle to be evaluated as a measured vehicle in the benchmark steady traffic flow; S2: injecting a traffic flow disturbance parameter into the benchmark steady traffic flow to convert it into a dynamic disturbance traffic flow, and controlling the measured vehicle to automatically drive in the dynamic disturbance traffic flow, and recording its whole-process running data; S3: extracting performance indicators of the measured vehicle from the whole-process running data, including longitudinal acceleration and lateral acceleration; S4: for each performance indicator, calculating a deviation degree between the performance indicator in the dynamic disturbance traffic flow and a benchmark value in the benchmark steady traffic flow, and quantifying the deviation degree as a performance attenuation rate; S5: based on the performance attenuation rates of all the performance indicators, performing weighted fusion according to preset indicator weight coefficients to obtain a comprehensive disturbance tolerance index; S6: outputting an evaluation conclusion on automatic driving performance of the measured vehicle according to a numerical range of the comprehensive disturbance tolerance index. 2.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 1, wherein The S1 specifically comprises: S11: setting road structure, traffic signal and vehicle generation parameters in the simulation platform, initializing traffic flow, speed and type distribution of background vehicles, and constructing a basic traffic network environment; S12: continuously injecting background vehicles based on a microscopic simulation model, and monitoring overall vehicle speed fluctuation until a mean vehicle speed and a distance distribution tend to be stable, to generate a benchmark steady traffic flow; S13: selecting a position of a middle background vehicle in the benchmark steady traffic flow, embedding the smart connected vehicle to be evaluated, and inheriting initial state parameters of the vehicle as a simulation starting point. 3.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 1, wherein, The S2 specifically comprises: S21: setting a disturbance parameter set in the simulation platform, the disturbance parameter set comprising lane-changing interference density, sudden acceleration and deceleration probability and forward vehicle insertion frequency, and applying the disturbance parameter set to behavior rules of part of the background vehicles in the benchmark steady traffic flow; S22: based on a disturbance triggering condition, starting a disturbance injection process, simulating vehicle sudden acceleration, sudden deceleration or frequent lane-changing behavior at a local road section or a random position, to break the original steady flow state and construct a dynamic disturbance traffic flow; S23: starting an automatic driving control program of the measured vehicle, taking over lateral and longitudinal behavior execution according to a preset control strategy, and keeping real-time interaction with the dynamic disturbance traffic flow, recording whole-process speed, acceleration, path and behavior decision data of the measured vehicle to form a complete running data sequence. 4.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 1, wherein, The S3 specifically comprises: S31: during simulation running of the measured vehicle, collecting original running data of speed, heading angle and time stamp of the measured vehicle at a fixed sampling period based on a data collection interface of the simulation platform; S32: calculating a speed increment between adjacent time points according to a speed change amount at continuous time points, and extracting a longitudinal acceleration sequence in combination with a time interval; S33: extracting a heading angle change rate per unit time according to a heading angle change trend at continuous time points, and constructing a lateral acceleration sequence accordingly; S34: performing time synchronization and format regularization on the extracted longitudinal and lateral acceleration sequences to generate a complete performance indicator time sequence. 5.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 4, wherein, The S32 specifically comprises: S321: based on the simulation platform with a preset sampling period Collecting the speed data sequence of the measured vehicle, recording the longitudinal speed values at two continuous time points and and ;​ S322: Calculate the speed increment between adjacent time instants , the formula is: wherein, denotes the speed change amount in the time interval ; S323: divide the speed increment by the corresponding time interval , extract the longitudinal acceleration in the current time period, the formula is: , wherein, represents the longitudinal acceleration value in the first sampling interval.

6. The intelligent network connection automobile automatic driving performance evaluation method based on simulation test according to claim 5, characterized in that, The S33 specifically comprises: S331: during the simulation running, at a preset sampling period Collect the heading angle data of the measured vehicle, record the heading angle values at two continuous time points With the heading angle value at With ; S332: Calculate the rate of change of the heading angle in a unit of time , the formula is: , wherein, represents the angular velocity of the vehicle in the time interval; represents the heading angle of the vehicle at the time ; S333: combine the longitudinal speed of the vehicle at the current time , calculate the lateral acceleration, formula: , wherein, represents the lateral acceleration value at the time.

7. The intelligent network connection automobile automatic driving performance evaluation method based on simulation test according to claim 1, characterized in that, The S4 specifically comprises: S41: Collect the performance index time series of the vehicle under dynamic disturbance traffic flow, including the longitudinal acceleration sequence and the lateral acceleration sequence ; S42: Construct a benchmark steady-state traffic flow environment under the same simulation platform, and run the measured vehicle under the same control strategy to obtain the corresponding performance index sequence of the measured vehicle under steady-state conditions With ; S43: For each performance indicator, difference calculation is performed at time points to obtain a performance deviation sequence; S44: Take the mean square deviation of each deviation sequence as the disturbance response amplitude, and introduce the mean square value of the reference performance sequence for normalization processing, respectively calculate the longitudinal and transverse performance decay rates With . 8.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 1, wherein, The S5 specifically includes: S51: Set the performance index weighted fusion rule, and construct an index weight coefficient set containing the weights of each performance index wherein and respectively represent the preset weights corresponding to the longitudinal acceleration performance index and the lateral acceleration performance index. S52: call the calculated longitudinal performance attenuation rate and the lateral performance attenuation rate and weight them with the corresponding index weight to calculate the comprehensive disturbance tolerance index . 9.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 8, wherein, The performance indicator weighting fusion rule includes: Rule 1, for scenarios where longitudinal and lateral control performance of the autonomous vehicle are equally valued, sets the weight coefficient to , ; Rule 2: Applicable to high-speed driving, straight lane or ramp following scenarios, with longitudinal stability as the core, and the weight coefficient is set to , 3; Rule 3, applicable to urban roads, lane changing or traffic-intensive areas, with the core of lateral maneuvering ability, the weight coefficient is set to 3, 7. 10.The intelligent network connected vehicle automatic driving performance evaluation method based on simulation test of claim 8, wherein, The S6 specifically includes: S61: Set the evaluation grading rule of the disturbance tolerance index, divide the numerical range of the comprehensive disturbance tolerance index into multiple performance grading intervals, and bind the mapping relationship with the preset performance grading labels; S62: Match the comprehensive disturbance tolerance index calculated by the vehicle under the dynamic disturbance traffic flow with the preset interval to determine the corresponding performance grading label; S63: Output the corresponding evaluation conclusion, including the performance grade, evaluation explanation and optimization suggestion.

Images