Single index and performance evaluation method and system for longitudinal driving of intelligent vehicle

Abstract

The application discloses a kind of single index, performance evaluation method and system of intelligent automobile longitudinal driving, comprising: obtaining the driving data of mature driver, novice driver and intelligent automobile by testing the certain evaluation index of vehicle longitudinal performance in the same scene, the same working condition for multiple times;By respectively performing probability statistical analysis on the driving data corresponding to mature driver, novice driver and intelligent automobile, the driving characteristic probability distribution of mature driver, novice driver and intelligent automobile about the evaluation index is determined respectively;According to the driving characteristic probability distribution of mature driver and novice driver about the evaluation index, the high score evaluation standard and performance basic standard for calculating the index score corresponding to the evaluation index are determined respectively;Finally, based on the obtained high score evaluation standard and performance basic standard, the single index score of intelligent automobile about the evaluation index is quantitatively evaluated according to the driving characteristic probability distribution of intelligent automobile.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer-aided design of vehicles, and particularly relates to a single index and performance evaluation method and system for longitudinal driving of an intelligent vehicle. BACKGROUND

[0002] Forward longitudinal driving control is the most basic capability of an intelligent vehicle. At present, in the mode of a dedicated intelligent networked vehicle test site, an open road, or a whole vehicle in-loop simulation, the main way of testing and evaluating the driving performance of an intelligent vehicle is to set typical test scenes and test conditions to comprehensively evaluate the automatic driving level of the intelligent vehicle, such as typical test conditions of deceleration, slow driving, emergency, cut-in / cut-out, high-speed following, and stop-and-go driving. The test and evaluation stay at the "point" of specific conditions, and the evaluation results have a certain scientificity and rationality.

[0003] However, the evaluation results of the driving performance of the automatic driving system mainly reflect the functionality, and fail to comprehensively reflect the overall performance level of the system. Moreover, the driving experience of the driving system is mainly presented through subjective evaluation, the test results are not quantified, and the subjective evaluation has a high professional requirement for the evaluators, and the stability and consistency of the evaluation results are easily affected by the level and state of the evaluators. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application provides a single index and performance evaluation method and system for longitudinal driving of an intelligent vehicle, which can objectively and quantitatively evaluate the longitudinal driving performance of the intelligent vehicle. The specific technical solutions are as follows:

[0005] In a first aspect, a single index evaluation method for longitudinal driving of an intelligent vehicle is provided. In a first implementation manner of the first aspect, the method comprises:

[0006] Respectively acquiring a plurality of sets of driving data of a mature driver, a novice driver, and the intelligent vehicle about an evaluation index under the same test scene and the same condition;

[0007] Respectively determining driving characteristic probability distributions of the mature driver, the novice driver, and the intelligent vehicle about the evaluation index by statistically analyzing the corresponding driving data;

[0008] Respectively determining a high-score evaluation standard and a performance basic standard of the evaluation index according to characteristic values corresponding to the driving characteristic probability distributions of the mature driver and the novice driver;

[0009] Based on the high-score evaluation standard and the performance basic standard, calculating a single index score of the intelligent vehicle about the evaluation index according to the driving characteristic probability distribution of the intelligent vehicle.

[0010] In conjunction with the first feasible approach of the first aspect, in the second feasible approach of the first aspect, driving data on evaluation indicators from experienced drivers, novice drivers, and intelligent vehicles are obtained under the same test scenario and operating conditions, including:

[0011] Acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario;

[0012] The vehicle driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles are clustered to determine multiple sets of driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles under different working conditions.

[0013] In conjunction with the first possible implementation of the first aspect, in the third possible implementation of the first aspect, the high-score evaluation criteria include the mean and variance of the driving characteristic probability distribution corresponding to mature drivers and their corresponding score rate, and the basic performance criteria include the mean of the driving characteristic probability distribution corresponding to novice drivers and their corresponding score rate.

[0014] In conjunction with the first possible implementation of the first aspect, in the fourth possible implementation of the first aspect, the calculation of the individual indicator score of the intelligent vehicle regarding the evaluation index includes:

[0015]

[0016] Among them, E i σ i E represents the median and variance of the probability distribution of driving characteristics of intelligent vehicles with respect to evaluation index i, respectively. bi , E represents the average value of the probability distribution of driving characteristics for novice drivers and its corresponding score rate. gi σ gi , S σgi Let D be the mean, variance, and corresponding score rate of the probability distribution of driving characteristics for mature drivers. JSi The JS dispersion represents the probability distribution of driving characteristics and performance indicators of intelligent vehicles.

[0017] Secondly, a method for evaluating the longitudinal driving performance of intelligent vehicles is provided, including:

[0018] Using any of the first to fourth possible implementation methods of the first aspect, the single index evaluation method for longitudinal driving of intelligent vehicles is used to calculate the single index score of intelligent vehicles for different evaluation indicators.

[0019] Based on the individual indicator scores and weights corresponding to each evaluation indicator at the next level, the indicator scores corresponding to each evaluation indicator at the previous level are calculated by weighting them level by level, and finally the comprehensive score of the intelligent vehicle is obtained.

[0020] Thirdly, a method for evaluating the longitudinal driving performance of intelligent vehicles is provided, including:

[0021] Acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different testing scenarios;

[0022] Using the longitudinal driving performance evaluation method for intelligent vehicles as described in the second aspect, the comprehensive score of intelligent vehicles under different test scenarios and operating conditions is determined.

[0023] Fourthly, a single-item evaluation system for longitudinal driving of intelligent vehicles is provided, including:

[0024] The acquisition module is configured to acquire multiple sets of driving data on evaluation indicators from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and working conditions.

[0025] The statistics module is configured to determine the probability distribution of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles with respect to the evaluation indicators by performing statistical probability analysis on the corresponding driving data.

[0026] The standard module is configured to determine the high-score evaluation standard and basic performance standard corresponding to the evaluation index based on the feature values ​​corresponding to the driving characteristic probability distributions of experienced drivers and novice drivers, respectively.

[0027] The evaluation module is configured to calculate the individual index scores of the intelligent vehicle with respect to the evaluation indicators based on the high-score evaluation criteria and the basic performance criteria, according to the probability distribution of the driving characteristics of the intelligent vehicle.

[0028] Fifthly, a system for evaluating the longitudinal driving performance of an intelligent vehicle is provided, including:

[0029] The single-item evaluation module is configured to use the single-item evaluation method for longitudinal driving of intelligent vehicles as described in any of the first to fourth implementable methods of the first aspect to calculate the single-item score of intelligent vehicles for different evaluation indicators.

[0030] The comprehensive evaluation module is configured to calculate the index scores corresponding to each evaluation index of the previous level by weighting the individual index scores and weights of each evaluation index of the next level, and finally obtain the comprehensive score of the intelligent vehicle.

[0031] Sixthly, a system for evaluating the longitudinal driving performance of an intelligent vehicle is provided, including:

[0032] The driving data module is configured to acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different test scenarios and operating conditions.

[0033] The performance evaluation module is configured to use the intelligent vehicle longitudinal driving performance evaluation method as described in the second aspect to determine the comprehensive score of the intelligent vehicle under different test scenarios and operating conditions.

[0034] Beneficial Effects: By employing the single-item index, performance evaluation method, and system for longitudinal driving of intelligent vehicles of this invention, statistical probability analysis can be performed on driving data from experienced and novice drivers obtained through multiple tests to determine the probability distribution of driving characteristics corresponding to experienced and novice drivers. Based on this, high-score evaluation standards and basic performance standards corresponding to various evaluation indicators used to evaluate the longitudinal driving performance of intelligent vehicles can be determined. Furthermore, by performing statistical probability distribution analysis on the driving data of intelligent vehicles obtained through multiple tests, the probability distribution of driving characteristics of intelligent vehicles with respect to various evaluation indicators can be determined. Combined with the previously determined high-score evaluation standards and basic performance standards, the performance of intelligent vehicles with respect to various evaluation indicators can be quantitatively evaluated. Finally, by combining the performance corresponding to all evaluation indicators, the performance of intelligent vehicles can be comprehensively and objectively evaluated. Attached Figure Description

[0035] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly described below. In all the drawings, the elements or parts are not necessarily drawn to scale.

[0036] Figure 1 A flowchart of a method for evaluating a single indicator of longitudinal driving of an intelligent vehicle according to an embodiment of the present invention;

[0037] Figure 2 A flowchart of a method for evaluating the longitudinal driving performance of an intelligent vehicle according to an embodiment of the present invention;

[0038] Figure 3 A flowchart of a method for evaluating the longitudinal driving performance of an intelligent vehicle according to an embodiment of the present invention;

[0039] Figure 4 This is a system block diagram of an intelligent vehicle longitudinal driving single-item index evaluation system provided in an embodiment of the present invention;

[0040] Figure 5 This is a system block diagram of an intelligent vehicle longitudinal driving performance evaluation system provided in an embodiment of the present invention;

[0041] Figure 6 This is a system block diagram of an intelligent vehicle longitudinal driving performance evaluation system provided in an embodiment of the present invention. Detailed Implementation

[0042] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0043] It should be understood that the longitudinal driving performance of intelligent vehicles is essentially vehicle speed control, while distance control, comfortable driving, and other aspects reflect the performance requirements of driving in different scenarios and operating conditions. Therefore, in this embodiment, the evaluation of longitudinal driving performance can be divided into four dimensions: distance control, comfortable driving, comprehensive energy consumption, and traffic efficiency.

[0044] Among them, the evaluation indicators for distance control include system response time, stopping distance, and stable driving distance between vehicles at various speed ranges. The evaluation indicators for comfortable driving include the vehicle's system response time, acceleration, and jerk under test scenarios such as acceleration of the vehicle in front, deceleration of the vehicle in front, cutting in and out of the vehicle in front, and driving on curves.

[0045] When evaluating the longitudinal driving performance of intelligent vehicles, each individual evaluation indicator can be scored separately to determine the individual indicator score corresponding to each individual evaluation indicator. Then, the longitudinal driving performance of intelligent vehicles can be comprehensively evaluated based on the individual indicator scores corresponding to multiple individual evaluation indicators.

[0046] like Figure 1 The flowchart shown illustrates the evaluation method for a single indicator of longitudinal driving performance of intelligent vehicles. This evaluation method includes:

[0047] Step 1: Obtain multiple sets of driving data on evaluation indicators from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and working conditions;

[0048] Step 2: By performing statistical probability analysis on the relevant driving data, determine the probability distribution of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles with respect to the evaluation indicators;

[0049] Step 3: Determine the high-score evaluation standard and basic performance standard corresponding to the evaluation index based on the feature values ​​corresponding to the driving characteristic probability distributions of experienced drivers and novice drivers, respectively.

[0050] Step 4: Based on the high-score evaluation criteria and basic performance criteria, and according to the probability distribution of the driving characteristics of the intelligent vehicle, calculate the individual index score of the intelligent vehicle with respect to the evaluation indicators.

[0051] Specifically, firstly, driving data from experienced drivers, novice drivers, and intelligent vehicles can be obtained by conducting multiple tests on a specific evaluation index of vehicle longitudinal performance under the same scenarios and operating conditions. Then, by performing probabilistic statistical analysis on the driving data of experienced drivers, novice drivers, and intelligent vehicles respectively, the probability distribution of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles regarding that evaluation index can be determined.

[0052] Subsequently, based on the probability distribution of driving characteristics for experienced and novice drivers regarding this evaluation indicator, high-score evaluation standards and basic performance standards can be determined for calculating the corresponding indicator score. Finally, based on the obtained high-score evaluation standards and basic performance standards, the individual indicator score of the intelligent vehicle regarding this evaluation indicator can be quantitatively evaluated according to the probability distribution of driving characteristics. This avoids the influence of factors such as the level and condition of the evaluators on the stability and consistency of the longitudinal driving performance evaluation results of the intelligent vehicle.

[0053] In this embodiment, optionally, in step 1, driving data on evaluation indicators for experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and operating conditions are obtained, including:

[0054] Acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario;

[0055] The vehicle driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles are clustered to determine multiple sets of driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles under different working conditions.

[0056] Specifically, vehicle driving data from experienced drivers in test scenarios can be collected through an onboard road testing system. The onboard road testing system includes an environmental perception and measurement system (such as a system for perceiving and measuring targets ahead, adjacent lane targets, and rear targets) and a vehicle dynamic parameter measurement system (such as speed, longitudinal acceleration / deceleration, lateral acceleration, and inter-vehicle distance). The laboratory vehicle-in-the-loop testing system includes a forward target simulation system (such as video simulation / injection, millimeter-wave radar simulation), a chassis dynamometer, and a continuous test scenario library.

[0057] The collected vehicle driving data includes data corresponding to evaluation indicators such as inter-vehicle distance, acceleration / deceleration, and emergency lane change / braking timing. It also includes driving data from experienced drivers under different operating conditions. By performing cluster analysis on all collected vehicle driving data, the driving data corresponding to experienced drivers under different operating conditions can be determined. Through multiple tests in the same test scenario, multiple sets of driving data corresponding to each evaluation indicator for experienced drivers can be obtained under the same test scenario but different operating conditions. From these, multiple sets of driving data corresponding to a specific evaluation indicator for an experienced driver can be extracted.

[0058] Similarly, the same method can be used to obtain multiple sets of driving data from novice drivers and intelligent vehicles under the same scenarios but different operating conditions. By matching operating conditions, multiple sets of driving data corresponding to various evaluation indicators for experienced drivers, novice drivers, and intelligent vehicles under the same scenarios and operating conditions can be determined, which can then be used to evaluate the driving performance of intelligent vehicles.

[0059] In this embodiment, optionally, in step 3, the high-score evaluation criteria include the average value and variance of the driving characteristic probability distribution corresponding to mature drivers and their corresponding score rate, and the basic performance criteria include the average value of the driving characteristic probability distribution corresponding to novice drivers and their corresponding score rate.

[0060] Specifically, after obtaining multiple sets of driving data from experienced drivers for a specific evaluation indicator, statistical probability analysis can be performed on these data sets to obtain the probability distribution of driving characteristics of experienced drivers for that evaluation indicator. Then, the score rate corresponding to the mean and variance of this probability distribution is manually set. Finally, the mean and variance of the probability distribution of driving characteristics of experienced drivers for that evaluation indicator, along with the corresponding score rate, form a high-scoring evaluation standard for evaluating the longitudinal driving performance of intelligent vehicles. Similarly, the mean and corresponding score rate of the probability distribution of driving characteristics of novice drivers for that evaluation indicator form a basic performance standard for evaluating the longitudinal driving performance of intelligent vehicles. Finally, the mean and variance of the probability distribution of driving characteristics of intelligent vehicles for that evaluation indicator are compared with both the basic performance standard and the high-scoring evaluation standard to quantitatively evaluate the individual indicator score of intelligent vehicles under the same test scenario and operating conditions.

[0061] In this embodiment, optionally, calculating the individual indicator score of the intelligent vehicle regarding the evaluation index includes:

[0062]

[0063] Among them, E i σ iE represents the median and variance of the probability distribution of driving characteristics of intelligent vehicles with respect to evaluation index i, respectively. bi , E represents the average value of the probability distribution of driving characteristics for novice drivers and its corresponding score rate. gi σ gi , S σgi Let D be the mean, variance, and corresponding score rate of the probability distribution of driving characteristics for mature drivers. JSi The JS dispersion represents the probability distribution of driving characteristics and performance indicators of intelligent vehicles.

[0064]

[0065] like Figure 2 The flowchart shown illustrates a method for evaluating the longitudinal driving performance of intelligent vehicles. This evaluation method includes:

[0066] Step S1: Using the above-mentioned evaluation method for single indicators of longitudinal driving of intelligent vehicles, calculate the single indicator scores of intelligent vehicles for different evaluation indicators;

[0067] Step S2: Based on the individual indicator scores and weights corresponding to each evaluation indicator at the next level, calculate the indicator scores corresponding to each evaluation indicator at the previous level by weighting them level by level, and finally obtain the comprehensive score of the intelligent vehicle.

[0068] Specifically, firstly, multiple sets of driving data from experienced drivers, novice drivers, and intelligent vehicles regarding various evaluation indicators can be collected under the same test scenarios and operating conditions. Then, using the aforementioned evaluation method for single indicators of longitudinal driving performance of intelligent vehicles, the individual indicator score for each evaluation indicator can be determined. Next, the evaluation indicators can be categorized according to the evaluation system, and the indicator score of the next higher level indicator can be calculated based on the individual indicator score and weight corresponding to each indicator. This step-by-step weighted calculation yields the comprehensive score of the longitudinal driving performance of the intelligent vehicle. This allows for a comprehensive and objective quantitative evaluation of the longitudinal driving performance of intelligent vehicles. The specific calculation formula for the indicator score is as follows:

[0069]

[0070] Among them, S t S is the score for the indicator at the next higher level. i+1 and ε i+1 This refers to the indicator scores and weights of the next level evaluation indicators corresponding to the previous level indicator.

[0071] The evaluation system is as follows:

[0072]

[0073]

[0074] For example, the above-mentioned evaluation method for longitudinal driving indicators of intelligent vehicles can be used to calculate the individual indicator scores of intelligent vehicles regarding parking distance, system response time under the condition of acceleration of the vehicle in front, vehicle longitudinal acceleration, vehicle jerk, system response time under the condition of deceleration of the vehicle in front, vehicle longitudinal acceleration, and vehicle jerk.

[0075] Then, based on the corresponding tertiary indicators, the scores of the secondary indicators to which the tertiary indicators belong can be calculated. For example, based on the system response time, longitudinal acceleration, and jerk of the vehicle in the preceding vehicle acceleration scenario, the score of the indicator corresponding to the preceding vehicle acceleration can be calculated.

[0076] Subsequently, the score of the primary indicator to which the secondary indicator belongs can be calculated based on the scores of the corresponding secondary indicators. For example, the score of the comfort driving indicator can be calculated based on the scores and weights of the indicators corresponding to the acceleration, deceleration, entry, exit, and cornering of the vehicle in front. For secondary indicators without tertiary evaluation indicators, the above-mentioned evaluation method for single indicators of longitudinal driving of intelligent vehicles can be directly used to calculate the corresponding indicator score. For example, the score of the distance control indicator can be calculated by weighting the scores and weights of the single indicators corresponding to parking distance and stable following distance.

[0077] Finally, based on the scores and weights of all primary indicators, a comprehensive score for the longitudinal driving performance of the intelligent vehicle is calculated. For example, the comprehensive score for the longitudinal driving performance of the intelligent vehicle is calculated by weighting the scores and weights of indicators corresponding to comfortable driving, distance control, comprehensive energy consumption, and traffic efficiency.

[0078] like Figure 3 The flowchart shown is for an evaluation method of longitudinal driving performance of intelligent vehicles. This evaluation method includes:

[0079] Step D1: Obtain multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different test scenarios;

[0080] Step D2: Using the above-described evaluation method for the longitudinal driving performance of intelligent vehicles, determine the comprehensive score of intelligent vehicles under different test scenarios and operating conditions.

[0081] Specifically, firstly, multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles can be acquired under different test scenarios. Then, all acquired vehicle driving data are clustered to determine the corresponding datasets for different test scenarios and operating conditions. These datasets include multiple sets of driving data from experienced drivers, novice drivers, and intelligent vehicles. Next, using the aforementioned longitudinal driving performance evaluation method for intelligent vehicles, corresponding comprehensive scores are calculated based on the datasets for different scenarios and operating conditions. This allows for a comprehensive and objective quantitative evaluation of the longitudinal driving performance of intelligent vehicles under different scenarios and conditions.

[0082] like Figure 4 The diagram shown is a system block diagram of a single-item evaluation system for longitudinal driving of intelligent vehicles. This evaluation system includes:

[0083] The acquisition module is configured to acquire multiple sets of driving data on evaluation indicators from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and working conditions.

[0084] The statistics module is configured to determine the probability distribution of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles with respect to the evaluation indicators by performing statistical probability analysis on the corresponding driving data.

[0085] The standard module is configured to determine the high-score evaluation standard and basic performance standard corresponding to the evaluation index based on the feature values ​​corresponding to the driving characteristic probability distributions of experienced drivers and novice drivers, respectively.

[0086] The evaluation module is configured to calculate the individual index scores of the intelligent vehicle with respect to the evaluation indicators based on the high-score evaluation criteria and the basic performance criteria, according to the probability distribution of the driving characteristics of the intelligent vehicle.

[0087] Specifically, the evaluation system includes an acquisition module, a statistics module, a standards module, and an evaluation module. The acquisition module can acquire driving data from experienced drivers, novice drivers, and intelligent vehicles by conducting multiple tests on a specific evaluation index of vehicle longitudinal performance under the same scenarios and operating conditions. The statistics module, through probability statistical analysis of the driving data from experienced drivers, novice drivers, and intelligent vehicles, can determine the probability distribution of driving characteristics for each type of driver regarding that evaluation index.

[0088] The standard module can determine the high-scoring evaluation criteria and basic performance criteria for calculating the corresponding evaluation index based on the probability distribution of driving characteristics of experienced and novice drivers for that evaluation index. Based on the obtained high-scoring evaluation criteria and basic performance criteria, the evaluation module can quantitatively evaluate the individual index score of the intelligent vehicle for that evaluation index according to the probability distribution of driving characteristics. This avoids the influence of factors such as the level and condition of the evaluators on the stability and consistency of the longitudinal driving performance evaluation results of the intelligent vehicle.

[0089] like Figure 5 The system block diagram shown is for an intelligent vehicle longitudinal driving performance evaluation system. This evaluation system includes:

[0090] The single-item evaluation module is configured to use the above-mentioned evaluation method for single-item indicators of longitudinal driving of intelligent vehicles to calculate the single-item scores of intelligent vehicles for different evaluation indicators.

[0091] The comprehensive evaluation module is configured to calculate the index scores corresponding to each evaluation index of the previous level by weighting the individual index scores and weights of each evaluation index of the next level, and finally obtain the comprehensive score of the intelligent vehicle.

[0092] Specifically, the evaluation system includes a single-item evaluation module and a comprehensive evaluation module. The single-item evaluation module collects multiple sets of driving data from experienced drivers, novice drivers, and intelligent vehicles regarding various evaluation indicators under the same test scenarios and operating conditions. It then uses the aforementioned evaluation method for single-item longitudinal driving indicators of intelligent vehicles to determine the single-item score for each evaluation indicator. The comprehensive evaluation module categorizes the evaluation indicators according to the evaluation system and calculates the weighted score of the next higher-level indicator based on the single-item score and weight corresponding to each indicator. This hierarchical weighted calculation yields the comprehensive score for the longitudinal driving performance of the intelligent vehicle. This allows for a comprehensive and objective quantitative assessment of the longitudinal driving performance of intelligent vehicles.

[0093] like Figure 6 The system block diagram shown is for an intelligent vehicle longitudinal driving performance evaluation system. This evaluation system includes:

[0094] The driving data module is configured to acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different test scenarios and operating conditions.

[0095] The performance evaluation module is configured to use the aforementioned intelligent vehicle longitudinal driving performance evaluation method to determine the comprehensive score of the intelligent vehicle under different test scenarios and operating conditions.

[0096] Specifically, the evaluation system includes a driving data module and a performance evaluation module. The driving data module acquires multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different test scenarios. It then clusters all acquired vehicle driving data to determine the corresponding datasets for different test scenarios and operating conditions. These datasets include multiple sets of driving data for experienced drivers, novice drivers, and intelligent vehicles. The performance evaluation module employs the aforementioned longitudinal driving performance evaluation method for intelligent vehicles, calculating corresponding comprehensive scores based on the datasets for different scenarios and operating conditions. This allows for a comprehensive and objective quantitative evaluation of the longitudinal driving performance of intelligent vehicles under different scenarios and conditions.

[0097] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for evaluating a single indicator of longitudinal driving performance in intelligent vehicles, characterized in that, include: Multiple sets of driving data on evaluation indicators were obtained from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and working conditions. By performing statistical probability analysis on the relevant driving data, the probability distributions of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles with respect to the evaluation indicators are determined respectively. The average and variance of the probability distribution of driving characteristics of experienced drivers with respect to evaluation indicators, and the corresponding score rate, are used to form a high-scoring evaluation standard for evaluating the longitudinal driving performance of intelligent vehicles. The average and the corresponding score rate of the probability distribution of driving characteristics of novice drivers with respect to evaluation indicators are used to form a basic performance standard for evaluating the longitudinal driving performance of intelligent vehicles. Based on the high-score evaluation criteria and basic performance criteria, and according to the probability distribution of the driving characteristics of the intelligent vehicle, the individual index scores of the intelligent vehicle with respect to the evaluation indicators are calculated, including: ； ； ； in, , Let be the median and variance of the probability distribution of driving characteristics of intelligent vehicles with respect to evaluation index i, respectively. , These represent the average value of the probability distribution of driving characteristics for novice drivers and their corresponding score rates. , , , These represent the mean, variance, and corresponding score rate of the probability distribution of driving characteristics for mature drivers. The JS dispersion represents the probability distribution of driving characteristics and performance indicators of intelligent vehicles.

2. The evaluation method for a single indicator of longitudinal driving of an intelligent vehicle according to claim 1, characterized in that, Obtain driving data on evaluation metrics from experienced drivers, novice drivers, and intelligent vehicles under the same test scenarios and operating conditions, including: Acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario; The vehicle driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles are clustered to determine multiple sets of driving data corresponding to experienced drivers, novice drivers, and intelligent vehicles under different working conditions.

3. A method for evaluating the longitudinal driving performance of an intelligent vehicle, characterized in that, include: Using the single-item evaluation method for longitudinal driving of intelligent vehicles as described in claim 1 or 2, the single-item scores of intelligent vehicles with respect to different evaluation indicators are calculated. The evaluation indicators are graded according to the evaluation system. Based on the individual indicator scores and weights of each evaluation indicator at the next lower level, the indicator scores of each evaluation indicator at the previous level are calculated by weighting them up level by level, and finally the comprehensive score of the intelligent vehicle is obtained.

4. A method for evaluating the longitudinal driving performance of an intelligent vehicle, characterized in that, include: Acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different testing scenarios; Using the longitudinal driving performance evaluation method for intelligent vehicles as described in claim 3, the comprehensive score of intelligent vehicles under different test scenarios and operating conditions is determined.

5. A single-item evaluation system for longitudinal driving of intelligent vehicles, characterized in that, include: The acquisition module is configured to acquire multiple sets of driving data on evaluation indicators from experienced drivers, novice drivers, and intelligent vehicles under the same test scenario and working conditions. The statistics module is configured to determine the probability distribution of driving characteristics of experienced drivers, novice drivers, and intelligent vehicles with respect to the evaluation indicators by performing statistical probability analysis on the corresponding driving data. The standard module is configured to form a high-scoring evaluation standard for evaluating the longitudinal driving performance of intelligent vehicles, based on the average and variance of the probability distribution of driving characteristics of mature drivers with respect to evaluation indicators, and the score rate corresponding to the average and variance. It also forms a basic performance standard for evaluating the longitudinal driving performance of intelligent vehicles, based on the average and score rate of the probability distribution of driving characteristics of novice drivers with respect to evaluation indicators. The evaluation module is configured to calculate the individual indicator scores of the intelligent vehicle regarding the evaluation indicators based on the high-score evaluation criteria and basic performance criteria, according to the probability distribution of the driving characteristics of the intelligent vehicle, including: ； ； ； in, , Let be the median and variance of the probability distribution of driving characteristics of intelligent vehicles with respect to evaluation index i, respectively. , These represent the average value of the probability distribution of driving characteristics for novice drivers and their corresponding score rates. , , , These represent the mean, variance, and corresponding score rate of the probability distribution of driving characteristics for mature drivers. The JS dispersion represents the probability distribution of driving characteristics and performance indicators of intelligent vehicles.

6. An intelligent vehicle longitudinal driving performance evaluation system, characterized in that, include: The single-item evaluation module is configured to use the single-item evaluation method for longitudinal driving of intelligent vehicles as described in claim 1 or 2 to calculate the single-item scores of intelligent vehicles for different evaluation indicators. The comprehensive evaluation module is configured to classify the various evaluation indicators according to the evaluation system, and calculate the indicator scores corresponding to the various evaluation indicators of the previous level by weighting them according to the individual indicator scores and weights of the next level of evaluation indicators, and finally obtain the comprehensive score of the intelligent vehicle.

7. An intelligent vehicle longitudinal driving performance evaluation system, characterized in that, include: The driving data module is configured to acquire multiple sets of vehicle driving data from experienced drivers, novice drivers, and intelligent vehicles under different test scenarios and operating conditions. The performance evaluation module is configured to use the intelligent vehicle longitudinal driving performance evaluation method as described in claim 3 to determine the comprehensive score of the intelligent vehicle under different test scenarios and different working conditions.

Images