Intelligent network connection automobile open road grading and test demonstration advanced evaluation method

Abstract

The invention provides an intelligent networked automobile open road grading and test demonstration advanced evaluation method, which comprises the following steps of road evaluation index selection, road evaluation index grading, road comprehensive evaluation model construction, model application, advanced evaluation index selection, advanced evaluation model construction and model application. A set of intelligent networked automobile road evaluation index system is designed, a comprehensive model is established, and a set of automatic evaluation of open roads is designed by means of automatic inspection vehicles, manpower, systems, platforms and the like. According to the method, vehicle running performance indexes such as vehicle running speed and acceleration on a road and test experience indexes such as accumulated test mileage and accumulated test duration are selected, evaluation indexes are rich, road test and demonstration application subjects and vehicle comprehensive capability can be accurately measured, and meanwhile, vehicle running dynamic safety monitoring is realized; a set of evaluation model is also established, and an advanced evaluation method realized by means of roadside equipment, intelligent network connection automobiles and the like is designed.

Description

Technical Field

[0001] The present invention relates to the field of traffic technology, and in particular to an open road classification and test demonstration advanced evaluation method for intelligent connected vehicles. Background Art

[0002] With the continuous development and maturity of current autonomous driving technology, intelligent connected vehicles have become the direction of industry development and a must-win for industrial development in various cities. In order to promote the development of the intelligent connected vehicle industry and to provide a good foundation for the intelligent connected vehicle industry, on the one hand, we will promote a wider range of open roads to provide a wider road space for the operation of autonomous driving vehicles. On the other hand, in terms of management methods, we will clarify the management requirements and application process for intelligent connected vehicles on the road to improve the approval and management efficiency. At present, before social roads are open to autonomous driving vehicles for road testing and demonstration applications, they need to be evaluated according to certain standards before they can be opened. Relevant intelligent connected vehicle companies can only carry out activities and achieve different levels of road test demonstration advancement after obtaining the qualifications for open road testing and demonstration applications in accordance with procedures.

[0003] As the intelligent connected vehicle industry is still in the exploratory development stage, although the software and hardware technologies related to autonomous driving technology are constantly improving with the deepening of testing and demonstration activities, the corresponding road assessment and advanced management technologies and methods are still relatively backward. At present, traditional manual means are mostly used for road assessment and a simple mileage accumulation model is used for intelligent connected vehicle management, which is inconsistent with the intelligent and information-based characteristics of intelligent connected vehicles and cannot meet the needs of efficient and safe management.

[0004] At present, the open road evaluation of intelligent networked vehicles is mainly based on the established road evaluation index system. For road evaluation indicators such as traffic flow, traffic speed, road section form, etc., each indicator is investigated and data is obtained through manual means, and then the road environment complexity level is comprehensively evaluated based on the performance of each indicator data. In terms of open road test demonstration and advanced management, the intelligent networked vehicle applicants are evaluated mainly based on the published management measures. The evaluation indicators are mostly simple measures of the company's road testing and demonstration application experience, such as cumulative test mileage, cumulative test time, etc., combined with the subject's indicator data and safety assurance capabilities, and the subject's road testing and demonstration application advanced qualifications are comprehensively reviewed. However, the existing technology has defects. First, the various indicators in the existing road evaluation index system are mainly qualitative, and the weights of the various indicators are not clearly distinguished. In the implementation process, data is mostly obtained by manual investigation, which has the disadvantages of low investigation efficiency, high manpower and time costs, and large subjective influence of investigators. In addition, the qualitative evaluation model only supports the management of roads in a single level mode, which is not precise and safe enough. At the current stage when cities are implementing the full-domain opening of intelligent networked vehicles, the workload of using this technical means to evaluate roads is large, and it is impossible to conduct real-time supervision of changes in the open road environment, and the road operation risk is relatively high. Second, the existing advanced management of intelligent networked vehicle road testing and demonstration applications mainly reviews the company's cumulative testing experience, such as cumulative test mileage, cumulative test time, and basic indicators such as no responsible traffic accidents. It is impossible to effectively evaluate the performance of intelligent networked vehicles after they are put on the road and the impact on normal road traffic. In addition, the relevant certificates mainly rely on the company to submit paper certification materials, which are then manually reviewed, which is inefficient. Summary of the invention

[0005] In view of the above problems, the present invention aims to realize dynamic and automated evaluation and grading of open roads for intelligent networked vehicles, and solve the problems of low efficiency, insufficient precision and safety, and inability to monitor changes in open road environment in real time in the existing road evaluation index system. In order to solve the above problems, the present invention provides an advanced evaluation method for open road grading and test demonstration of intelligent networked vehicles, which has good feasibility, specifically: An open road assessment method for an intelligent connected vehicle comprises the following steps: S1. Selection of road assessment indicators: Building a road assessment indicator system , is the total number of indicators, For the road assessment indicators, and complete the selection of road assessment indicators; S2. Road Assessment Index Classification: Selection roads as samples, determine the value range of n road evaluation indicators and the classification scheme of road evaluation indicators; S3, constructing a comprehensive road assessment model: calculating the road assessment index weight and the road comprehensive assessment value according to the value range selected by the road assessment index and the result of the road assessment index classification, completing the construction of the comprehensive road assessment model, and the model is used to judge the road level; S4. Application of the road comprehensive evaluation model: The road evaluation index value of the open road to be evaluated is connected to the road comprehensive evaluation model constructed in step S3, and a road classification scheme of the open road to be evaluated is output.

[0006] Further preferably, the step S2 specifically comprises the following steps: S21, Quantification of indicators; Road evaluation index The index value is , using the formula or , calculate the road evaluation index The index value is obtained by the number of times or length proportion of the corresponding road elements. L represents the length of the road under investigation. Represents road evaluation index Number of times it appears on the road, Represents road evaluation index The length of the road occupied, wherein the road elements include at least one of potholes and roadside parking; S22. Standardization of scoring; setting road indicators The grading standard values ​​are A, B, and C. When the score is normalized to ,when When the score is normalized to ,when When the score is normalized to ; A, B, and C are set to 10, 20, and 30 respectively. Arranged from small to large, they are divided into three intervals: , There are three dividing values.

[0007] Further preferably, the step of calculating the road evaluation index weight in step S3 is as follows: S31, use the analytic hierarchy process to obtain the indicator weights; S311, fill in the indicator importance questionnaire using the 1-9 scale method, and convert the road evaluation indicator system Fill in all the elements in , obtain the comparison results of the importance of indicators, and compare any two road evaluation indicators. , The comparison result is defined as , where i, j are [1, ] any integer in ; S312: Obtain a judgment matrix based on the comparison results of the importances ; S313, calculate the judgment matrix The product of each row element is Dimensional vector ; S314, the The initial indicator weight distribution is obtained by standardizing and normalizing the dimension vector ; S315, calculate the maximum eigenvalue of the judgment matrix A , and perform consistency check using the formula , , if the calculated , then the consistency of the judgment matrix meets the requirements, that is, the initial indicator weight distribution can be adopted, where CI, CR, and RI are consistency index, consistency ratio, and average random consistency index respectively; S32, iteratively update the indicator weights using the entropy weight method; S321. Constructing a decision matrix Each column in the decision matrix is ​​all the road index data of a road, Indicates The first section The index value of each road evaluation index is is the total number of sample sections; S322. Calculate entropy value; introduce feature weight , which is the The first section The probability that the index value of a road evaluation index causes the entropy value of the entire evaluation system to change, , then The entropy value of the road evaluation index is , where In (N), Find N, The logarithmic function value of ; S323. Calculate indicator weights; introduce indicator difference coefficients , indicating the The dispersion degree of all evaluation values ​​of road evaluation indicators is calculated as follows: , obtain the indicator weight distribution after iterative update by entropy weight method , is the entropy value of the i-th road evaluation index; S33. Calculate the final indicator weight , road evaluation index The weights of , for The weight, , The indicator weight distribution is obtained after the hierarchical analysis method standardizes and normalizes the vector and the entropy weight method iteratively updates it.

[0008] Further preferably, the calculation of the comprehensive road evaluation value in step S3 comprises the following steps: S34. Index value based on the obtained road evaluation index , weight of each indicator , calculate the road segment The comprehensive evaluation value of S341, Computational Classical Domain :

[0009] For the division complexity level, For all levels, , express About the index values ​​of road evaluation indicators The value range of , Indicates the value range The upper and lower limits of S342, calculation section area : ; is the overall complexity of the road environment, Indicates the index value of the road evaluation index The overall value range of , is the lower limit of the value range, is the upper limit of the value range; S343, calculate the closeness , judge the closeness to each level: ; ; is the index value of the road evaluation index With value range distance; S344, calculation complexity evaluation eigenvalue: ; ; For the planned closeness, is the comprehensive evaluation value of the road of the nth road section.

[0010] Further preferably, the road level determination in step S3 includes: Comprehensive evaluation value of the road After sorting in ascending order, it is divided into 3 intervals , , , , , are the complexity evaluation characteristic values ​​of a certain road. The maximum value of all complexity evaluation eigenvalues; like , then the road belongs to Class I road, which indicates low complexity; if , then the road belongs to Class II road, which indicates medium complexity; if , then the road belongs to Class III road, and Class III road indicates high complexity.

[0011] Further preferably, the road assessment data is obtained in step S4 specifically as follows: Use patrol vehicles equipped with multiple high-definition cameras, high-precision locators, and edge computing gateways to collect video resources of road infrastructure; establish a scene library corresponding to road indicators, and automatically identify key data including target location, size, and length through deep learning algorithms and image recognition technology, and upload them to the intelligent connected vehicle system or platform; obtain road assessment index values ​​through manual inspections, traffic police, and operation index systems, and upload them to the intelligent connected vehicle system or platform; The road classification scheme is output in step S4, specifically: Relying on the relevant system or platform of intelligent connected vehicles, the comprehensive road evaluation model constructed in step S3 is deployed to the intelligent connected vehicle system or platform, and the road complexity is calculated and the road classification scheme is output through the obtained road evaluation index value.

[0012] An advanced evaluation method for open road test demonstration of intelligent connected vehicles, referring to the grading scheme of the aforementioned open road evaluation method of intelligent connected vehicles, comprises the following steps: M1. Selection of advanced evaluation indicators for test demonstration; M2. Constructing an advanced evaluation model; the advanced evaluation model includes a subject basic index for judging whether the subject basic index meets the grade requirements, a vehicle basic index model for judging whether the vehicle basic index meets the grade requirements, and a vehicle operation index model for judging whether the vehicle operation index meets the grade requirements; M3, application of advanced evaluation model: obtaining vehicle and subject data and connecting to the advanced evaluation model constructed in step S6 to determine whether it meets the advanced grade road standards.

[0013] Further preferably, in step M2: Main basic index model: score the main basic index, and set the main basic index score requirements corresponding to the three levels of roads as follows: , , , determine whether the scoring requirements are met; Vehicle basic index model: The cumulative test mileage of a vehicle is equal to the sum of the mileage on roads of various levels / areas, etc. The average vehicle takeover interval mileage is equal to the vehicle's cumulative test mileage divided by the cumulative number of vehicle takeovers ; Set the cumulative test road mileage requirements corresponding to the three levels as follows: , , The average takeover interval mileage requirements are , , ; Determine whether the cumulative test mileage and average takeover interval mileage required for the corresponding grade of road are met; Vehicle operation index model: The vehicle operation index includes the difference rate between the self-driving vehicle and the manually driven vehicle in terms of speed, acceleration, following distance, and number of stops of the intelligent connected vehicle, with a value of 0-100%; ;in , are the average speeds of autonomous vehicles and manually driven vehicles on the road section, is the speed difference between the autonomous driving vehicle and the manually driven vehicle on the road section, that is, the speed difference rate between the autonomous driving vehicle and the manually driven vehicle; ;in It is the frequency of sudden acceleration and deceleration in a unit time for autonomous driving vehicles and manual driving vehicles. The difference in the frequency of rapid acceleration and deceleration between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in acceleration between the autonomous driving vehicle and the manually driven vehicle; ;in is the average following distance between the autonomous driving vehicle and the manually driven vehicle, The following distance difference between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in following distance between the autonomous driving vehicle and the manually driven vehicle; ;in is the average number of stops for autonomous vehicles and manually driven vehicles, is the average number of stops of autonomous vehicles and manually driven vehicles, that is, the difference rate in the number of stops of autonomous vehicles and manually driven vehicles; The requirements for the difference rates of vehicle operation indicators corresponding to the three levels of roads are: , , ; Using manual driving test data, the sample mean of the test data is used as the average level of manual driving; introducing comprehensive indicators ,in, For the Items of evaluation indicators, For its weight, ; The comprehensive index requirements corresponding to the three road grades are , , , compare and determine whether it meets the requirements for advancement to a higher level road.

[0014] Further preferably, the step M3 of acquiring vehicle and subject data is as follows: Use roadside cameras, radars, vehicle-mounted GPS equipment, operation index systems, and supervision platforms to collect vehicle operation data such as the speed, position, and acceleration of road network vehicles; call the vehicle operation data of the speed, position, and acceleration of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the vehicle basic indicator data of the test mileage and takeover times of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the intelligent connected vehicle enterprise data accessed by the intelligent connected vehicle platform.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention aims to achieve automation. First, it designs a set of intelligent networked vehicle road evaluation index system, establishes a comprehensive model, and designs a set of open road automation evaluation based on automatic inspection vehicles, manual, system and platform means, selects vehicle running speed, acceleration and other vehicle running performance indicators on the road and cumulative test mileage, cumulative test time and other test experience indicators, with rich evaluation indicators, which can accurately measure the comprehensive capabilities of road test and demonstration application subjects and vehicles, and realize dynamic safety monitoring of vehicle operation. Second, it designs a set of intelligent networked vehicle road test and demonstration application index system, establishes a set of evaluation models and designs advanced evaluation methods based on roadside equipment, intelligent networked vehicles and other means.

[0016] The present invention can monitor multiple road indicators, and based on the characteristics of dynamic and static data, it can update some data on a daily, monthly, and quarterly basis, realize dynamic road assessment and automatic road environment monitoring, reduce about 80% of manual inspection costs, and significantly improve assessment accuracy. Monitoring multiple indicators such as the speed and acceleration of intelligent networked vehicles can comprehensively measure the operating performance of autonomous driving vehicles, improve management precision and road operation safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a method flow chart of an open road assessment method for an intelligent connected vehicle of the present invention; Figure 2 It is a method flow chart of an advanced evaluation method for open road test demonstration of an intelligent networked vehicle of the present invention; Figure 3 This is a schematic diagram of the quantification of indicators in step S2 of Example 1 of the present invention; Figure 4 This is a schematic diagram of index classification in step S2 of Example 1 of the present invention; Figure 5 It is a schematic diagram of the accumulated mileage and classification cases of vehicle basic indicators in step M2 of embodiment 2 of the present invention; Figure 6 Schematic diagram of the distribution and classification of manual driving levels of vehicle operation indicators in step M2 of embodiment 2 of the present invention; Figure 7 It is a schematic diagram of the changes of latitude and longitude, speed and acceleration over time for the comparison of vehicle running indicators in step M3 of embodiment 2 of the present invention; Figure 8 It is a schematic diagram of calculating road network manually driven vehicle data for comparing vehicle operation indicators in step M3 of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Explanation of abbreviations and key term definitions: Open road: refers to roads in existing urban roads that are designated specifically for road testing and demonstration applications of intelligent connected vehicles, referred to as "open roads" for short.

[0020] Road testing: refers to the testing of the autonomous driving functions of intelligent connected vehicles conducted on designated sections of various types of roads used for the passage of social motor vehicles, such as highways (including expressways), urban roads, and regional areas.

[0021] Demonstration application: refers to the operation of intelligent connected vehicles carrying people and goods with pilot and trial effects on designated sections of various types of roads used for social motor vehicle traffic, such as highways (including expressways), urban roads, and regional areas. Road testing and demonstration application are referred to as test demonstration.

[0022] The preferred embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0023] Embodiment 1: An open road assessment method for an intelligent connected vehicle comprises the following steps: S1. Selection of road assessment indicators; Construction of road assessment indicator system , is the total number of indicators, For the road assessment indicators; S2. Classification of road assessment indicators; selection roads as samples, and determine the value range and classification scheme of the evaluation indicators of all n roads; the step S2 specifically includes the following steps: S21, Quantification of indicators; Road evaluation index The index value is , using the formula or , calculate the road evaluation index The index value is obtained by the number of times or length proportion of the corresponding road elements. L represents the length of the road under investigation. Represents road evaluation index Number of times it appears on the road, Represents road evaluation index The length of the road occupied, wherein the road elements include at least one of potholes and roadside parking; S22. Standardization of scoring; setting road indicators The grading standard values ​​are A, B, and C. When the score is normalized to ,when When the score is normalized to ,when When the score is normalized to ; A, B, and C are set to 10, 20, and 30 respectively. Arranged from small to large, they are divided into three intervals: , There are three dividing values.

[0024] S3, constructing a comprehensive road assessment model; calculating the road assessment index weight and the road comprehensive assessment value according to the value range selected by the road assessment index and the result of the road assessment index classification, completing the construction of the comprehensive road assessment model, and the model is used to judge the road level; the steps of calculating the road assessment index weight in step S3 are as follows: S31, use the analytic hierarchy process to obtain the indicator weights; S311, fill in the indicator importance questionnaire using the 1-9 scale method, and convert the road evaluation indicator system Fill in all the elements in , obtain the comparison results of the importance of indicators, and compare any two road evaluation indicators. , The comparison result is defined as , where i, j are [1, ] any integer in ; S312: Obtain a judgment matrix based on the comparison results of the importances ; S313, calculate the judgment matrix The product of each row element is Dimensional vector ; S314, the The initial indicator weight distribution is obtained by standardizing and normalizing the dimension vector ; S315, calculate the maximum eigenvalue of the judgment matrix A , and perform consistency check using the formula , , if the calculated , then the consistency of the judgment matrix meets the requirements, that is, the initial indicator weight distribution can be adopted, where CI, CR, and RI are consistency index, consistency ratio, and average random consistency index respectively; S32, iteratively update the indicator weights using the entropy weight method; S321. Constructing a decision matrix Each column in the decision matrix is ​​all the road index data of a road, Indicates The first section The index value of each road evaluation index is is the total number of sample sections; S322. Calculate entropy value; introduce feature weight , which is the The first section The probability that the index value of a road evaluation index causes the entropy value of the entire evaluation system to change, , then The entropy value of the road evaluation index is , where In (N), Find N, The logarithmic function value of ; S323. Calculate indicator weights; introduce indicator difference coefficients , indicating the The dispersion degree of all evaluation values ​​of road evaluation indicators is calculated as follows: , obtain the indicator weight distribution after iterative update by entropy weight method , is the entropy value of the i-th road evaluation index; S33. Calculate the final indicator weight , road evaluation index The weights of , for The weight, , The indicator weight distribution is obtained after the hierarchical analysis method standardizes and normalizes the vector and the entropy weight method iteratively updates it.

[0025] The steps of calculating the comprehensive evaluation value of the road in step S3 are as follows: S34. Index value based on the obtained road evaluation index , weight of each indicator , calculate the road segment The comprehensive evaluation value of S341, Computational Classical Domain :

[0026] For the division complexity level, For all levels, , express About the index values ​​of road evaluation indicators The value range of , Indicates the value range The upper and lower limits of can be determined in combination with the actual values ​​of the indicators; S342, calculation section area : ; is the overall complexity of the road environment, Indicates the index value of the road evaluation index The overall value range of , is the lower limit of the value range, is the upper limit of the value range; S343, calculate the closeness , judge the closeness to each level: ; ; is the index value of the road evaluation index With value range distance; S344, calculation complexity evaluation eigenvalue: ; ; For the planned closeness, is the comprehensive evaluation value of the road of the nth road section.

[0027] The road level determination in step S3 includes: Comprehensive evaluation value of the road After sorting in ascending order, it is divided into 3 intervals , , , , , are the complexity evaluation characteristic values ​​of a certain road. The maximum value of all complexity evaluation eigenvalues; like , then the road belongs to Class I road, which indicates low complexity; if , then the road belongs to Class II road, which indicates medium complexity; if , then the road belongs to Class III road, and Class III road indicates high complexity.

[0028] S4, application of comprehensive road assessment model; connecting the road assessment index value of the acquired road data to the comprehensive road assessment model constructed in step S3, and outputting the road classification scheme of the open road to be assessed; the road assessment data acquired in step S4 is specifically: Use patrol vehicles equipped with multiple high-definition cameras, high-precision locators, and edge computing gateways to collect video resources of road infrastructure; establish a scene library corresponding to road indicators, and automatically identify key data including target location, size, length, and upload them to the system or platform through deep learning algorithms and image recognition technology, as well as road surface quality, signs and markings; obtain road assessment index values ​​through manual inspections, traffic police, and operation index systems, and upload them to the intelligent connected vehicle system or platform; The road classification scheme is output in step S4, specifically: Relying on the relevant system or platform of intelligent connected vehicles, the comprehensive road evaluation model constructed in step S3 is deployed to the intelligent connected vehicle system or platform, and the road complexity is calculated and the road classification scheme is output through the obtained road evaluation index value.

[0029] An advanced evaluation method for open road test demonstration of intelligent connected vehicles, referring to the grading scheme of the open road evaluation method of intelligent connected vehicles, includes the following steps: M1. Selection of advanced evaluation indicators for test demonstration; M2. Constructing an advanced evaluation model; the advanced evaluation model includes a subject basic index for judging whether the subject basic index meets the grade requirements, a vehicle basic index model for judging whether the vehicle basic index meets the grade requirements, and a vehicle operation index model for judging whether the vehicle operation index meets the grade requirements; In the step M2: Main basic index model: score the main basic index, and set the main basic index score requirements corresponding to the three levels of roads as follows: , , , determine whether the scoring requirements are met; Vehicle basic index model: The cumulative test mileage of a vehicle is equal to the sum of the mileage on roads of various levels / areas, etc. The average vehicle takeover interval mileage is equal to the vehicle's cumulative test mileage divided by the cumulative number of vehicle takeovers ; Set the cumulative test road mileage requirements corresponding to the three levels as follows: , , The average takeover interval mileage requirements are , , ; Determine whether the cumulative test mileage and average takeover interval mileage required for the corresponding grade of road are met; Vehicle operation index model: The vehicle operation index includes the difference rate between the self-driving vehicle and the manually driven vehicle in terms of speed, acceleration, following distance, and number of stops of the intelligent connected vehicle, with a value of 0-100%; ;in , are the average speeds of autonomous vehicles and manually driven vehicles on the road section, is the speed difference between the autonomous driving vehicle and the manually driven vehicle on the road section, that is, the speed difference rate between the autonomous driving vehicle and the manually driven vehicle; ;in It is the frequency of sudden acceleration and deceleration in a unit time for autonomous driving vehicles and manual driving vehicles. The difference in the frequency of rapid acceleration and deceleration between the automatic and manual driving vehicles, that is, the difference rate in acceleration between the automatic driving vehicle and the manual driving vehicle; ;in is the average following distance between the autonomous driving vehicle and the manually driven vehicle, The following distance difference between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in following distance between the autonomous driving vehicle and the manually driven vehicle; ;in is the average number of stops for autonomous vehicles and manually driven vehicles, is the average number of stops of autonomous vehicles and manually driven vehicles, that is, the difference rate in the number of stops of autonomous vehicles and manually driven vehicles; The requirements for the difference rates of vehicle operation indicators corresponding to the three levels of roads are: , , ; Using manual driving test data, the sample mean of the test data is used as the average level of manual driving; introducing comprehensive indicators ,in, For the Items of evaluation indicators, For its weight, ; The comprehensive index requirements corresponding to the three road grades are , , , compare and determine whether it meets the requirements for advancement to a higher level road.

[0030] M3, application of advanced evaluation model: obtaining vehicle and subject data and connecting to the advanced evaluation model constructed in step S6 to determine whether it meets the advanced grade road standards.

[0031] The specific steps of obtaining vehicle and subject data in step M3 are as follows: Use roadside cameras, radars, vehicle-mounted GPS equipment, operation index systems, and supervision platforms to collect vehicle operation data such as the speed, position, and acceleration of road network vehicles; call the vehicle operation data of the speed, position, and acceleration of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the vehicle basic indicator data of the test mileage and takeover times of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the intelligent connected vehicle enterprise data accessed by the intelligent connected vehicle platform.

[0032] Embodiment 2: The present invention provides an open road assessment method for intelligent networked vehicles, the method flow chart is as follows: Figure 1 As shown, the following steps are included: S1. Selection of road assessment indicators: Analyze the factors causing accidents of traditional vehicles and intelligent connected vehicles in recent years, obtain the key factors leading to traffic accidents, and conduct interviews and surveys with experts and scholars related to intelligent connected vehicle technology and road traffic safety to extract factors that affect the road traffic safety of autonomous driving vehicles. With the goal of achieving automated collection with available tools, optimize the indicator system to obtain the final road assessment indicator system for people (pedestrian flow, hot spots, etc.), vehicles (operation index, truck ratio, etc.), roads (pavement quality, clear signs and markings, etc.), and the environment. , total number of indicators is 14, For the road evaluation indicators.

[0033] The 14 road assessment indicators are as follows:

[0034] S2. Classification of road assessment indicators; selection The roads are taken as samples to determine the value range and classification scheme of all road evaluation indicators; the step S2 specifically includes the following steps: S21. Quantification of indicators; combination Figure 3 As shown in the schematic diagram of the indicator quantification in step S2, Road evaluation index The index value is , using the formula or , calculate the road evaluation index The index value is obtained by the number of times or length proportion of the corresponding road elements. L represents the length of the road under investigation. Represents road evaluation index Number of times it appears on the road, Represents road evaluation index The length of the road, where road elements include potholes, roadside parking, etc.

[0035] S22. Index classification: The open roads for intelligent connected vehicles are planned to be managed according to a three-level open model. Correspondingly, the evaluation indicators of each road will also be divided into three levels. The higher the road level, the more complex the road is and the less road mileage it occupies.

[0036] There are two types of index classification. Type 1 is to continue the existing other standards and document classification methods for reclassification. Building land density, traffic operation index, non-motorized vehicle lane form, road traffic safety level. Taking the building density of a certain place as an example, the building land density is divided into five areas, including level 1 density four and five areas, level 2 density two and three areas, and level 3 density one area.

[0037] Type 2 is classified based on the range of index values. Other indicators include hotspot area distribution, traffic composition, road line shape, number and form of intersections, etc. Taking the hotspot area distribution as an example, the number of hotspot areas such as schools, hospitals, and business districts along the roads is mainly considered. The total number of hotspot areas in the city is divided by the total number of main and secondary roads, and the average hotspot area distribution density in the city is about 0.7 / km. Combined with the actual survey, the number of people in the area is generally concentrated in 0~2, which is divided into level one 0~0.5 / km, level two 0.5~1 / km, and level three 1~2 / km.

[0038] Combination Figure 4 Schematic diagram of index classification in step S2 of the embodiment of the present invention, if , then the road index At level 1, if , then the road index At level 2, if , then the road index At level 3; The indicators include hot spot area distribution, traffic composition, road alignment, and the number and form of intersections.

[0039] S23. Standardization of scoring; setting road assessment indicators The grading standard values ​​are A, B, and C. When the score is normalized to ,when When the score is normalized to ,when When the score is normalized to , obtain standardized indicator values ​​with scoring values ; A, B, and C are set to 10, 20, and 30 respectively. Arranged from small to large, they are divided into three intervals: , There are three dividing values.

[0040] S3. Construct a comprehensive road evaluation model; calculate the road evaluation index weights and the comprehensive road evaluation value to determine the road level and obtain the complexity evaluation characteristic value gather; The steps for calculating the road evaluation index weight in step S3 are as follows: S31, use the analytic hierarchy process to obtain the indicator weights; S311, through the platform to the road evaluation index system The importance survey form was sent to the experts for filling in. The survey form is as follows:

[0041] Use the 1-9 scale method to fill in the indicator importance comparison table, obtain the indicator importance pairwise comparison results, and compare any two road evaluation indicators. , The comparison result is defined as , where i, j are [1, ]; take 3 as an example to illustrate the digital scale. express The former is slightly more important than the latter. Indicates that the latter is slightly more important than the former.

[0042] S312: Obtain a judgment matrix based on the comparison results of the importances ; S313, calculate the judgment matrix The product of each row element is Dimensional vector ; S314, the The initial indicator weight distribution is obtained by standardizing and normalizing the dimension vector ; S315, calculate the maximum eigenvalue of the judgment matrix A And perform consistency check, , , if the calculated , then the consistency of the judgment matrix meets the requirements, that is, the initial indicator weight distribution can be adopted, where CI, CR, and RI are consistency index, consistency ratio, and average random consistency index respectively; S32. Use the entropy weight method to iteratively update the indicator weights; Entropy is an abstract concept that comes from the physical quantity that represents the state of a system in thermodynamics. It represents the degree of chaos of a system. The greater the entropy, the more chaotic the system is, because the system always develops in the direction of chaos, and the more chaotic the state, the more stable the system is. The entropy weight method iteratively updates the weight coefficients in the comprehensive evaluation, which can truly and objectively reflect the law of change of the system.

[0043] S321. Constructing a decision matrix Each column in the decision matrix is ​​all the road index data of a road, Indicates The first section Road evaluation index values, is the total number of sample sections; S322. Calculate entropy value; introduce feature weight , which is the The first section The probability that the index value of a road evaluation index causes the entropy value of the entire evaluation system to change, , then The entropy value of the road evaluation index is , where In (N), Find N, The logarithmic function value of ; S323. Calculate indicator weights; introduce indicator difference coefficients , indicating the The dispersion degree of all evaluation values ​​of road evaluation indicators is calculated as follows: The greater the difference of an evaluation indicator, the greater its entropy value, and the greater the amount of information it provides, so the greater the role it plays in the evaluation system and the higher its weight; Obtain the indicator weight distribution after iterative update of the entropy weight method , is the entropy value of the i-th road index; S33. Calculate the final weighting scheme , road evaluation index The weights of , for The weight, , The indicator weight distribution is obtained after the hierarchical analysis method standardizes and normalizes the vector and the entropy weight method iteratively updates it.

[0044] The steps of calculating the comprehensive evaluation value of the road in step S3 are as follows: S34. Index value based on the obtained road evaluation index , weight of each indicator , calculate the road segment The comprehensive evaluation value of S341, Computational Classical Domain :

[0045] For the division complexity level, For all levels, , express About the index values ​​of road evaluation indicators The value range of , Indicates the value range The upper and lower limits of can be determined in combination with the actual values ​​of the indicators; S342, calculation section area : ; is the overall complexity of the road environment, Indicates the index value of the road evaluation index The overall value range of , is the lower limit of the value range, is the upper limit of the value range; S343, calculate the closeness , judge the closeness to each level: ; ; is the index value of the road evaluation index With value range distance; S344, calculation complexity evaluation eigenvalue: ; ; For the planned closeness, is the comprehensive evaluation value of the road of the nth road section.

[0046] The road level determination in step S3 includes: Comprehensive evaluation value of the road After sorting from small to large, it is divided into 3 intervals , , , , , are the complexity evaluation characteristic values ​​of a certain road. The maximum value of all complexity evaluation eigenvalues; like , then the road belongs to Class I road, which indicates low complexity; if , then the road belongs to Class II road, which indicates medium complexity; if , then the road belongs to Class III road, and Class III road indicates high complexity.

[0047] S4, model application; connecting the road assessment index value obtained from the road data to the comprehensive road assessment model constructed in step S3, and outputting the road classification scheme; The step S4 of obtaining the road assessment data is as follows: Use inspection vehicles equipped with multiple high-definition cameras, high-precision locators, and edge computing gateways to collect road infrastructure video resources; establish a scene library corresponding to road indicators, and automatically identify key data including target location, size, length, and upload them to the system or platform through deep learning algorithms and image recognition technology, including road quality, signs and markings; obtain road evaluation index values ​​through manual inspections, traffic police, and operation index systems, and upload them to the intelligent connected vehicle system or platform; advanced evaluation indicators are as follows:

[0048] The road classification scheme is output in step S4, specifically: Relying on the relevant system or platform of intelligent connected vehicles, the comprehensive road evaluation model constructed in step S3 is deployed to the intelligent connected vehicle system or platform, and the road complexity is calculated and the road classification scheme is output through the obtained road evaluation index value.

[0049] An advanced evaluation method for open road testing of intelligent connected vehicles is as follows: Figure 2 As shown, the grading scheme of the open road assessment method of the intelligent connected vehicle mentioned above in this embodiment includes the following steps: M1. Selection of advanced evaluation indicators for testing and demonstration: Based on the open road classification plan in the first part, formulate an advanced plan for road testing and demonstration applications of intelligent connected vehicles, and clarify the application conditions for intelligent connected vehicle applicants and their vehicles to conduct road testing and demonstration applications on roads of various grades.

[0050] (1) Entity Strength: The applicant's strength is comprehensively evaluated based on the company's registered capital, the size of the planned fleet, the size of the local workforce, qualifications and performance, safety assurance and compensation capabilities, etc.

[0051] (2) Vehicle testing experience: The accumulated test mileage, average takeover interval mileage, and the presence or absence of traffic violations are used as basic vehicle indicators to measure the accumulated testing experience of intelligent connected vehicles on the road.

[0052] (3) Vehicle performance: The difference rate and comprehensive score of the average road speed, acceleration, following distance and other indicators between automatic and manually driven vehicles are used as vehicle evaluation indicators to measure the driving performance of intelligent connected vehicles on open roads and their impact on road operation.

[0053] M2. Constructing an advanced evaluation model; the advanced evaluation model includes a subject basic index for judging whether the subject basic index meets the grade requirements, a vehicle basic index model for judging whether the vehicle basic index meets the grade requirements, and a vehicle operation index model for judging whether the vehicle operation index meets the grade requirements; In the step M2: The basic indicators of the subject are scored, and the registered capital, other urban road tests, demonstration applications and other information submitted by the enterprise are analyzed. The expert group comprehensively evaluates the subject's strength and issues a scoring result. The subject scoring requirements corresponding to the three levels of roads are set as follows: , , , to determine whether the score requirements are met, all applicants must meet the main scores required for the corresponding grade of road.

[0054] The vehicle's cumulative test mileage is equal to the sum of the mileage on each level of road / area road, etc. The average vehicle takeover interval mileage is equal to the vehicle's cumulative test mileage divided by the cumulative number of vehicle takeovers ; The cumulative test road mileage requirements for the three levels are: , , The average takeover interval mileage requirements are , , ; Determine whether the cumulative test mileage and average takeover interval mileage required for the corresponding grade of road are met.

[0055] Among them, the grading standards are mainly formulated with reference to the cumulative test mileage and average takeover interval data of existing intelligent connected vehicle companies. Figure 5 It is a schematic diagram of the cumulative mileage and classification cases of basic vehicle indicators. For example, the current cumulative test mileage of enterprises is basically distributed between 60 and 130. In order to meet the safety principle and based on the existing technical level of the enterprise, three road grade requirements are defined as 0, 80, and 100 respectively.

[0056] Vehicle operation indicators include the difference rate between intelligent connected vehicles and manually driven vehicles in speed, acceleration, following distance, and number of stops, with values ​​ranging from 0 to 100%; ;in , are the average speeds of autonomous vehicles and manually driven vehicles on the road section, is the speed difference between the autonomous driving vehicle and the manually driven vehicle on the road section, that is, the speed difference rate between the autonomous driving vehicle and the manually driven vehicle; ;in It is the frequency of sudden acceleration and deceleration in a unit time for autonomous driving vehicles and manual driving vehicles. The difference in the frequency of rapid acceleration and deceleration between the automatic and manual driving vehicles, that is, the difference rate in acceleration between the automatic driving vehicle and the manual driving vehicle; ;in is the average following distance between the autonomous driving vehicle and the manually driven vehicle, The following distance difference between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in following distance between the autonomous driving vehicle and the manually driven vehicle; ;in is the average number of stops for autonomous vehicles and manually driven vehicles, is the average number of stops of autonomous vehicles and manually driven vehicles, that is, the difference rate in the number of stops of autonomous vehicles and manually driven vehicles; The difference rate requirements for each indicator corresponding to the three levels of roads are: , , ; Combine Figure 6 The manual driving level distribution and classification diagram of vehicle operation indicators uses the manual driving test data and takes the sample mean of the test data as the average manual driving level. The difference rate between the vast majority of vehicles and the average level is basically maintained within 20%, and the difference rate value is determined based on this, such as , , .

[0057] In order to balance the performance differences between different indicators and comprehensively measure the overall situation of vehicle operation, a comprehensive indicator is introduced ,in, For the Items of evaluation indicators, For its weight, ; The comprehensive index requirements corresponding to the three road grades are , , , compare and judge whether it is enough to advance to a higher level road. Generally speaking .like This means that the difference in operating indicators between all intelligent connected vehicles and traditional vehicles on first-level roads must be within 20% before they can advance to higher-level roads, but the overall difference rate must be within 15%.

[0058] M3, application of advanced evaluation model: obtaining vehicle and subject data and connecting to the advanced evaluation model constructed in step S2 to determine whether it meets the advanced grade road standards.

[0059] The model application of the present invention mainly includes two modules. One is the open road classification module, which aims to achieve automatic evaluation. Based on the relevant systems or platforms of intelligent connected vehicles, an evaluation model for open roads of intelligent connected vehicles is constructed, and an open road classification algorithm is deployed on the system. Road data is obtained by accessing automatic inspection vehicles, manual inspections, other data platforms and other data means to achieve automatic evaluation and classification of open roads. The second is the test demonstration advanced module, which aims to achieve automatic evaluation. Based on the relevant systems or platforms of intelligent connected vehicles, an advanced evaluation model for road testing and demonstration applications of intelligent connected vehicles is constructed, and an advanced test demonstration management algorithm is deployed on the system. Data such as roadside cameras, key vehicle GPS, and intelligent connected vehicles are accessed to achieve automatic evaluation of test demonstration advanced.

[0060] The specific steps of obtaining vehicle and subject data in step M3 are as follows: Use roadside cameras, radars, vehicle-mounted GPS equipment, operation index systems, and supervision platforms to collect vehicle operation data on the speed, position, and acceleration of road network vehicles; call the vehicle operation data on the speed, position, and acceleration of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the basic vehicle indicator data on the test mileage and takeover times of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the intelligent connected vehicle enterprise data accessed by the intelligent connected vehicle platform; The vehicle operation data processing in step M3 is specifically as follows: Vehicle acceleration, speed and other data can be obtained using a mature vehicle-mounted GPS module; establish the coordinate axes of longitude and latitude trajectory points, speed, and acceleration over time, as follows Figure 7 Schematic diagram of the latitude and longitude, speed and acceleration of vehicle operation indicators over time, selected time point Vehicle location The latitude and longitude coordinates are , Time point Vehicle location The latitude and longitude coordinates are , Time difference between two points Small enough to be used as a vehicle passing by The instantaneous speed ; Distance between two points , is the radius of the Earth; instantaneous velocity , Approaches 0; thus, the coordinate axis of the vehicle speed changing with time is obtained. The speeds at the two points are , ,when Time difference between two points Small enough to be used as a vehicle passing by Acceleration ; Acceleration , from which a graph showing vehicle acceleration over time can be obtained; combined with Figure 8 Schematic diagram of the calculation of road network artificial driving vehicle data for vehicle operation index comparison. Detection line 1 and detection line 2 are defined on the road. The time difference between vehicle 1 and vehicle 2 passing through detection line 1 is , the time difference between vehicle 2 passing through detection line 1 and detection line 2 is , the speed at this time is , then the following distance of vehicle 2 to vehicle 1 is ; The vehicle operation index comparison in step M3 is specifically as follows: Intelligent connected car operation data After the vehicle-side equipment is connected to the government supervision platform for intelligent networked vehicles, the processed speed, acceleration and other indicators can be directly obtained; the operation differences between road network vehicles and intelligent networked vehicles on the road section can be calculated, and the difference rate of speed, acceleration and other indicators can be calculated for a certain road section; In step M3, whether to advance is determined specifically as follows: The vehicle and subject data, vehicle basic indicators, and vehicle operation index values ​​are calculated, and their compliance with the three levels of road advancement standards set in the early stage is judged to form an advancement result and determine whether to advance.

[0061] The present invention aims to achieve automation. First, it designs a set of intelligent networked vehicle road evaluation index system, establishes a comprehensive model, and designs a set of open road automation evaluation based on automatic inspection vehicles, manual, system and platform means, selects vehicle running speed, acceleration and other vehicle running performance indicators on the road and cumulative test mileage, cumulative test time and other test experience indicators, with rich evaluation indicators, which can accurately measure the comprehensive capabilities of road test and demonstration application subjects and vehicles, and realize dynamic safety monitoring of vehicle operation. Second, it designs a set of intelligent networked vehicle road test and demonstration application index system, establishes a set of evaluation models and designs advanced evaluation methods based on roadside equipment, intelligent networked vehicles and other means.

[0062] The present invention can monitor multiple road indicators, and based on the characteristics of dynamic and static data, it can update some data on a daily, monthly, and quarterly basis, realize dynamic road assessment and automatic road environment monitoring, reduce about 80% of manual inspection costs, and significantly improve assessment accuracy. Monitoring multiple indicators such as the speed and acceleration of intelligent networked vehicles can comprehensively measure the operating performance of autonomous driving vehicles, improve management precision and road operation safety.

[0063] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for evaluating an open road for an intelligent connected vehicle, characterized in that: The steps include: S1. Selection of road assessment indicators: Building a road assessment indicator system , is the total number of indicators, For the road assessment indicators, and complete the selection of road assessment indicators; S2. Road Assessment Index Classification: Selection roads as samples, determine the value range of n road evaluation indicators and the classification scheme of road evaluation indicators; S3, constructing a comprehensive road assessment model: calculating the road assessment index weight and the road comprehensive assessment value according to the value range selected by the road assessment index and the result of the road assessment index classification, completing the construction of the comprehensive road assessment model, and the model is used to judge the road level; S4. Application of the road comprehensive evaluation model: The road evaluation index value of the open road to be evaluated is connected to the road comprehensive evaluation model constructed in step S3, and a road classification scheme of the open road to be evaluated is output.

2. The method for evaluating an open road for an intelligent connected vehicle according to claim 1, characterized in that: The step S2 specifically includes the following steps: S21, Quantification of indicators; Road evaluation index The index value is , using the formula or , calculate the road evaluation index The index value is obtained by the number of times or length proportion of the corresponding road elements. L represents the length of the road under investigation. Represents road evaluation index Number of times it appears on the road, Represents road evaluation index The length of the road occupied, wherein the road elements include at least one of potholes and roadside parking; S22. Standardization of scoring; setting road indicators The grading standard values ​​are A, B, and C. When the score is normalized to ,when When the score is normalized to ,when When the score is normalized to ; A, B, and C are set to 10, 20, and 30 respectively. Arranged from small to large, they are divided into three intervals: , There are three dividing values.

3. The open road assessment method for intelligent connected vehicles according to claim 1, characterized in that: The steps for calculating the road evaluation index weight in step S3 are as follows: S31, use the analytic hierarchy process to obtain the indicator weights; S311, fill in the indicator importance questionnaire using the 1-9 scale method, and convert the road evaluation indicator system Fill in all the elements in , obtain the comparison results of the importance of indicators, and compare any two road evaluation indicators. , The comparison result is defined as , where i, j are [1, ] any integer in ; S312: Obtain a judgment matrix based on the comparison results of the importances ; S313, calculate the judgment matrix The product of each row element is Dimensional vector ; S314, the The initial indicator weight distribution is obtained by standardizing and normalizing the dimension vector ; S315, calculate the maximum eigenvalue of the judgment matrix A , and perform consistency check using the formula , , if the calculated , then the consistency of the judgment matrix meets the requirements, that is, the initial indicator weight distribution can be adopted, where CI, CR, and RI are consistency index, consistency ratio, and average random consistency index respectively; S32, iteratively update the indicator weights using the entropy weight method; S321. Constructing a decision matrix Each column in the decision matrix is ​​all the road index data of a road, Indicates The first section The index value of each road evaluation index is is the total number of sample sections; S322. Calculate entropy value; introduce feature weight , which is the The first section The probability that the index value of a road evaluation index causes the entropy value of the entire evaluation system to change, , then The entropy value of the road evaluation index is , where In (N), Find N, The logarithmic function value of ; S323. Calculate indicator weights; introduce indicator difference coefficients , indicating the The dispersion degree of all evaluation values ​​of road evaluation indicators is calculated as follows: , obtain the indicator weight distribution after iterative update by entropy weight method , is the entropy value of the i-th road evaluation index; S33. Calculate the final indicator weight , road evaluation index The weights of , for The weight, , The indicator weight distribution is obtained after the hierarchical analysis method standardizes and normalizes the vector and the entropy weight method iteratively updates it.

4. The open road assessment method for intelligent connected vehicles according to claim 3, characterized in that: The steps of calculating the comprehensive evaluation value of the road in step S3 are as follows: S34. Index value based on the obtained road evaluation index , weight of each indicator , calculate the road segment The comprehensive evaluation value of S341, Computational Classical Domain : For the division complexity level, For all levels, , express About the index values ​​of road evaluation indicators The value range of , Indicates the value range The upper and lower limits of S342, calculation section area : ； is the overall complexity of the road environment, Indicates the index value of the road evaluation index The overall value range of , is the lower limit of the value range, is the upper limit of the value range; S343, calculate the closeness , judge the closeness to each level: ； ； is the index value of the road evaluation index With value range distance; S344, calculation complexity evaluation eigenvalue: ； ； For the planned closeness, is the comprehensive evaluation value of the road of the nth road section.

5. The open road assessment method for intelligent connected vehicles according to claim 4, characterized in that: The road level determination in step S3 includes: The comprehensive evaluation value of the road After sorting from small to large, it is divided into 3 intervals , , , , , are the complexity evaluation characteristic values ​​of a certain road. The maximum value of all complexity evaluation eigenvalues; like , then the road belongs to Class I road, and Class I road indicates low complexity; like , then the road belongs to Class II road, and Class II road indicates medium complexity; like , then the road belongs to Class III road, and Class III road indicates high complexity.

6. The open road assessment method for intelligent connected vehicles according to claim 1, characterized in that: The road assessment data is obtained in step S4 specifically as follows: Use patrol vehicles equipped with multiple high-definition cameras, high-precision locators, and edge computing gateways to collect video resources of road infrastructure; establish a scene library corresponding to road indicators, and automatically identify key data including target location, size, and length through deep learning algorithms and image recognition technology, and upload them to the intelligent connected vehicle system or platform; obtain road assessment index values ​​through manual inspections, traffic police, and operation index systems, and upload them to the intelligent connected vehicle system or platform; The road classification scheme is output in step S4, specifically: Relying on the relevant system or platform of intelligent connected vehicles, the comprehensive road evaluation model constructed in step S3 is deployed to the intelligent connected vehicle system or platform, and the road complexity is calculated and the road classification scheme is output through the obtained road evaluation index value.

7. A method for advanced evaluation of open road test demonstration of intelligent connected vehicles, referring to the grading scheme of the open road evaluation method of intelligent connected vehicles described in claim 1, characterized in that: The steps include: M1. Selection of advanced evaluation indicators for test demonstration; M2. Constructing an advanced evaluation model; the advanced evaluation model includes a subject basic index for judging whether the subject basic index meets the grade requirements, a vehicle basic index model for judging whether the vehicle basic index meets the grade requirements, and a vehicle operation index model for judging whether the vehicle operation index meets the grade requirements; M3, application of advanced evaluation model: obtaining vehicle and subject data and connecting to the advanced evaluation model constructed in step S6 to determine whether it meets the advanced grade road standards.

8. The method for advanced evaluation of open road test demonstration of intelligent connected vehicles according to claim 7, characterized in that: In the step M2: Main basic index model: score the main basic index, and set the main basic index score requirements corresponding to the three levels of roads as follows: , , , determine whether the scoring requirements are met; Vehicle basic index model: The cumulative test mileage of a vehicle is equal to the sum of the mileage on roads of various levels / areas, etc. ; The average vehicle takeover interval mileage is equal to the vehicle's cumulative test mileage divided by the cumulative number of vehicle takeovers ; Set the cumulative test road mileage requirements corresponding to the three levels as follows: , , The average takeover interval mileage requirements are , , ; Determine whether the cumulative test mileage and average takeover interval mileage required for the corresponding grade of road are met; Vehicle operation index model: The vehicle operation index includes the difference rate between the self-driving vehicle and the manually driven vehicle in terms of speed, acceleration, following distance, and number of stops of the intelligent connected vehicle, with a value of 0-100%; ;in , are the average speeds of autonomous vehicles and manually driven vehicles on the road section, is the speed difference between the autonomous driving vehicle and the manually driven vehicle on the road section, that is, the speed difference rate between the autonomous driving vehicle and the manually driven vehicle; ;in It is the frequency of sudden acceleration and deceleration in a unit time for autonomous driving vehicles and manual driving vehicles. The difference in the frequency of rapid acceleration and deceleration between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in acceleration between the autonomous driving vehicle and the manually driven vehicle; ;in is the average following distance between the autonomous driving vehicle and the manually driven vehicle, The following distance difference between the autonomous driving vehicle and the manually driven vehicle, that is, the difference rate in following distance between the autonomous driving vehicle and the manually driven vehicle; ;in is the average number of stops for autonomous vehicles and manually driven vehicles, is the average number of stops of autonomous vehicles and manually driven vehicles, that is, the difference rate in the number of stops of autonomous vehicles and manually driven vehicles; The requirements for the difference rates of vehicle operation indicators corresponding to the three levels of roads are: , , ; Using manual driving test data, the sample mean of the test data is used as the average level of manual driving; introducing comprehensive indicators ,in, For the Items of evaluation indicators, For its weight, ; The comprehensive index requirements corresponding to the three road grades are , , , compare and determine whether it meets the requirements for advancement to a higher level road.

9. The method for advanced evaluation of open road test demonstration of intelligent connected vehicles according to claim 7, characterized in that: The specific steps of obtaining vehicle and subject data in step M3 are as follows: Use roadside cameras, radars, vehicle-mounted GPS equipment, operation index systems, and supervision platforms to collect vehicle operation data such as the speed, position, and acceleration of road network vehicles; call the vehicle operation data of the speed, position, and acceleration of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the vehicle basic indicator data of the test mileage and takeover times of intelligent connected vehicles accessed by the intelligent connected vehicle platform; call the intelligent connected vehicle enterprise data accessed by the intelligent connected vehicle platform.

Images