Intelligent networked vehicle formation driving multi-performance evaluation method and system

Abstract

The invention discloses an intelligent connected vehicle (CAV) formation driving multi-performance evaluation method and system, and the method comprises the steps: constructing a CAV queue control model based on different car-following strategies, so as to adapt to different traffic conditions and formation driving performance requirements; a multi-criterion decision evaluation system is established, different weights are set for different criteria, and the diversified requirements of CAV formation driving performance are met; a fuzzy logarithmic least square method (FLLS) is applied to determine weights of different decision makers for performance criteria, so that the system can fully reflect the comparison priorities of different decision makers for different criteria; a GRA-TOPSIS method is constructed in combination with gray correlation analysis (GRA) and an ideal solution similarity sorting method (TOPSIS), and comprehensive evaluation indexes for evaluating CAV formation driving multi-performance expression are calculated, so that evaluation results under different strategies are sorted. According to the invention, decision support can be provided for a decision maker to select the optimal CAV formation driving strategy in a complex traffic environment.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology and autonomous driving technology, and in particular to a multi-performance evaluation method and system for intelligent networked vehicle platooning. Background Art

[0002] With the rapid development of intelligent transportation and autonomous driving technologies, connected vehicles (CAVs) have gradually become an important part of the intelligent transportation system. CAVs have significant advantages in reducing road congestion, improving traffic efficiency, and reducing energy consumption and emissions through information sharing and automatic control technology between vehicles. At present, CAV platoon technology (i.e. "platoon driving") has gradually attracted attention. Through the vehicle spacing control strategy, multiple CAVs are formed into a platoon to achieve more efficient road use and resource consumption management.

[0003] In CAV platoons, vehicle spacing control strategies are the core content. Common control strategies include fixed time interval (CTG) and fixed spacing (CS). The CTG strategy sets the time interval between vehicles and calculates the real-time spacing by multiplying the vehicle speeds. It is suitable for scenarios that require high energy efficiency and safety. The CS strategy maintains a fixed distance between vehicles, which can effectively improve traffic efficiency in low-traffic scenarios. In recent years, a hybrid control strategy (CC) combining CTG and CS has been proposed to take into account the advantages of different strategies. However, different control strategies have significant differences in emissions, energy consumption, traffic flow and safety. There is currently a lack of systematic evaluation methods for each strategy under multiple criteria. Summary of the invention

[0004] Purpose of the invention: In view of the problems existing in the prior art, the purpose of the present invention is to provide a multi-performance evaluation method and system for intelligent connected vehicle platooning, which combines the fuzzy logarithmic least squares method (FLLS), grey relational analysis (GRA) and TOPSIS method to evaluate the multi-criteria performance of CAV queues under different spacing control strategies, and comprehensively consider the preferences of decision makers to determine the optimal queue strategy.

[0005] Technical solution: In order to achieve the above-mentioned invention object, the present invention provides a method for evaluating the multi-performance of connected vehicle platoon driving, comprising the following steps:

[0006] (1) Establishing a CAV platoon control model that adapts to different traffic conditions based on different car-following strategies; the different car-following strategies include one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of mixed CTG and CS (CTG-CS);

[0007] (2) Establishing a multi-criteria evaluation system to meet the diverse needs of decision makers for platooning performance by setting different weights for different criteria; the multi-criteria evaluation system includes one or more performance standards of emissions, energy consumption, efficiency, and safety;

[0008] (3) Based on the decision makers' linguistic variable data, the fuzzy logarithmic least squares (FLLS) method is used to accurately calculate the weight of each criterion. The obtained weight can reflect the priority of different decision makers in comparing different criteria.

[0009] (4) Grey relational analysis (GRA) is embedded into the topology of ideal solution similarity (TOPSIS) to construct the GRA-TOPSIS method. The comprehensive evaluation index of CAV platooning performance is calculated based on simulation data or measured data, so as to rank the evaluation results of different strategies.

[0010] In some embodiments, the measurement of CAV platoon driving performance is evaluated and verified through data generated by a simulation system. First, vehicle position and speed data are generated to simulate the driving trajectory of the CAV queue. Then, the set following control model and control parameters are used to calculate one or more performance of the vehicle including emissions, energy consumption, efficiency and safety. The data is organized into a decision matrix, and finally, the FLLS and GRA-TOPSIS methods are used to analyze the data and calculate the normalized weight values ​​to obtain the performance of the CAV queue under different strategies.

[0011] In some embodiments, the fixed time interval (CTG) strategy dynamically adjusts the vehicle's following behavior by setting a fixed time interval and calculating the expected distance between vehicles based on the product of the time interval and the speed of the following vehicle; the fixed spacing (CS) strategy ensures a stable expected distance between vehicles by maintaining a fixed distance between vehicles; under the hybrid strategy, vehicles in the CAV queue are randomly distributed based on the CTG and CS strategies to form a hybrid queue mode, and the perception system and V2V communication system are used to obtain the speed, acceleration and position information of the leading vehicle to achieve real-time control.

[0012] In some embodiments, the hybrid platoon optimizes the comprehensive indicators of emissions, energy consumption, efficiency and safety, and determines the optimized vehicle distribution pattern and comprehensive spacing strategy based on fuzzy grey correlation analysis and ideal solution similarity ranking method (fuzzy GRA-TOPSIS), thereby improving the overall performance of platoon driving.

[0013] In some embodiments, the four evaluation indicators in the multi-criteria evaluation system including emissions, energy consumption, efficiency and safety correspond to carbon dioxide emissions, fuel consumption, traffic throughput and collision time respectively.

[0014] Furthermore, the FLLS method is used to accurately calculate the weight of each criterion, which specifically includes the following steps:

[0015] (3.1) Use linguistic variables to represent the fuzzy comparison matrix;

[0016] (3.2) The linguistic variables of the comparison matrix are converted into fuzzy numerical values ​​for calculation, and the comparison matrix of the kth decision maker is obtained:

[0017]

[0018] Where n represents the number of criteria, They represent the three element values ​​of the k-th decision maker's criterion i compared to criterion j, where the character u represents upper, m represents middle, and l represents lower;

[0019] (3.3) Based on the weighted geometric mean method, the fuzzy comparison matrices of K decision makers are summarized and the average comparison matrix is ​​calculated as follows:

[0020]

[0021] (3.4) Construct the FLLS method to determine the weight of each criterion:

[0022]

[0023] in, is the weight of the ith criterion, and Yes i The three elements of ij is the importance of criterion i compared to criterion j, Corresponding to three element values ​​respectively.

[0024] In some embodiments, K=1 or K>1, when K>1, the K decision makers are of the same type.

[0025] Furthermore, the performance ranking of different CAV queues is performed based on the GRA-TOPSIS method, which specifically includes the following steps:

[0026] (4.1) Numerical experiments on CAV platoons under different car-following strategies;

[0027] (4.2) Determine the decision matrix based on the simulation results;

[0028] (4.3) The values ​​of each criterion in the decision matrix of different CAV queues are normalized, that is, r ij :

[0029]

[0030] Where m represents the number of CAV queues, y ij represents the calculated value of the jth criterion of the i-th CAV queue;

[0031] (4.4) According to the criterion weight results and the obtained normalized results, the weighted normalized decision matrix is ​​calculated. For the convenience of calculation, the weights of each indicator are converted into clear numerical values. The calculation formula of the weighted normalized matrix is ​​as follows:

[0032]

[0033] x ij =ω j r ij

[0034] Among them, x ij is the weighted value of the jth criterion of the i-th CAV queue, and n is the number of criteria;

[0035] (4.5) Calculate the best possible ideal solution PIS and the worst possible solution NIS respectively:

[0036]

[0037] (4.6) The grey correlation coefficient between each indicator and PIS and NIS can be expressed as follows:

[0038]

[0039] (4.7) The grey relational grade of each CAV cohort for PIS and NIS is calculated as follows:

[0040]

[0041] (4.8) Determine the relative closeness of the grey relational degree of the CAV queues, and compare based on this value to obtain the sorting order of the CAV queues:

[0042]

[0043] Among them, μ i It represents the relative closeness of the grey relational degree of the i-th CAV queue, and its value is in the interval [0,1].

[0044] A multi-performance evaluation system for intelligent connected vehicle platooning, comprising:

[0045] A car-following control module, used to establish a CAV platoon control model adapted to different traffic conditions based on different car-following strategies; the different car-following strategies include one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of mixed CTG and CS (CTG-CS);

[0046] An evaluation index module is used to establish a multi-criteria evaluation system, which satisfies the diverse needs of decision makers for platoon driving performance by setting different weights for different criteria; the multi-criteria evaluation system includes one or more performance standards of emissions, energy consumption, efficiency and safety;

[0047] The weight determination module is used to accurately calculate the weight of each criterion based on the decision maker's language variable data and apply the fuzzy logarithmic least squares (FLLS) method. The obtained weight can reflect the priority of different decision makers for the comparison between different criteria.

[0048] And a comprehensive ranking module, which is used to embed grey relational analysis (GRA) into the ideal solution similarity ranking method (TOPSIS) to construct a GRA-TOPSIS method, and calculate and evaluate the comprehensive evaluation indicators of CAV platoon driving multiple performances based on simulation data or measured data, so as to rank the evaluation results of different strategies.

[0049] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of a method for evaluating multiple performances of intelligent connected vehicle platooning.

[0050] Beneficial effects: Compared with the prior art, the present invention develops a hybrid multi-criteria decision-making (MCDM) method that combines the FLLS, GRA and TOPSIS methods to measure the CAV platoon driving performance under different spacing strategies, thereby determining the ranking results of the CAV queue. The present invention can evaluate the CAV platoon driving performance under different spacing strategies from the perspectives of two different decision makers, enterprise managers and administrative managers, taking into account emissions, energy consumption, efficiency and safety impacts, and can provide decision support for decision makers to select the best CAV platoon driving strategy in a complex traffic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flowchart of a multi-performance evaluation method for intelligent connected vehicle platooning provided by an embodiment of the present invention;

[0052] Figure 2 The figure is a schematic diagram of the CAV platoon under study, where (a) CTG-CAVP represents a CAV platoon using different CTG strategies; (b) CS-CAVP represents a CAV platoon using the CS strategy; (c) CC-CAVP represents a CAV platoon using both CTG and CS strategies. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0054] The present invention provides a multi-performance evaluation method for intelligent connected vehicle platooning, which aims to conduct a scientific and comprehensive evaluation of the emission, energy consumption, efficiency and safety of CAV platoons under different vehicle spacing control strategies. Through a series of modeling, calculation and optimization, the present invention can provide decision support for decision makers to select the best platoon driving strategy in complex traffic environments.

[0055] like Figure 1 As shown, an embodiment of the present invention discloses a multi-performance evaluation method for intelligent connected vehicle platooning, which mainly includes:

[0056] (1) Based on different car-following strategies, a CAV platoon control model that adapts to different traffic conditions is established to ensure the accuracy of platoon driving performance evaluation. The specific different car-following strategies include one or more of the constant time gap (CTG) strategy, the constant spacing (CS) strategy, and the combined strategy of mixed CTG and CS (CTG-CS).

[0057] (2) Establish a multi-criteria decision-making system and design a multi-criteria evaluation system that includes one or more performance standards such as emissions, energy consumption, efficiency, and safety. By setting different weights for different criteria, the diverse needs of decision makers (such as corporate decision makers and administrative managers) for platoon driving performance can be met.

[0058] (3) Based on the decision makers' linguistic variable data, the fuzzy logarithmic least squares (FLLS) method is used to accurately calculate the weight of each criterion. The obtained weight can reflect the priority of different decision makers in comparing different criteria.

[0059] (4) Grey relational analysis (GRA) is embedded into the topology of ideal solution similarity (TOPSIS) to construct the GRA-TOPSIS method. The comprehensive evaluation index of CAV platooning performance is calculated based on simulation data or measured data, so as to rank the evaluation results of different strategies.

[0060] The present invention can achieve optimization of different CAV platoon driving performance indicators by setting different vehicle spacing control strategies and controlling the following behavior of each CAV. In this embodiment, CTG, CS and hybrid CTG-CS (CC) control strategies are introduced to enable CAV to flexibly control the queue according to different road conditions, traffic flow and environmental requirements. The CTG strategy is suitable for safety control under high traffic flow. By maintaining a fixed time interval between vehicles, it can achieve dynamic adjustment of the vehicle spacing and enhance the stability and energy saving effect of the queue; the CS strategy is based on a fixed spacing and is more suitable for scenarios with less road flow, which helps to improve the efficiency and stability of CAV platoon driving; and the hybrid strategy has the advantages of both CTG and CS strategies, and dynamically switches under different traffic conditions to achieve a balance between safety and efficiency of the CAV queue. Each step is described in detail below.

[0061] (1) Establishment of CAV platoon following control model

[0062] The CAV platoon following control model is based on different spacing strategies, including one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of mixed CTG and CS (CTG-CS). Specifically, the CAV platoon following control model is:

[0063]

[0064]

[0065] In the formula, and represents the target distance between the preceding vehicle i-1 and the target vehicle i under the CTG and CS strategies; h is the fixed time interval, d s is the safe distance; p i (t),v i (t) and a i (t) represent the position, velocity and actual acceleration of vehicle i in the CAV queue at time t, respectively; g i Indicates time delay, satisfaction δ i represents the perception delay of vehicle i, represents the upper bound of the communication delay between vehicle i and the preceding vehicle i-1; and represents the spacing error between two consecutive CAVs using CTG and CS strategies; based on previous studies (Li et al., 2022; Zheng et al., 2023a), is the expected acceleration of vehicle i at time t using the Bian method based on the CTG strategy, is the expected acceleration of vehicle i at time t using the Ploeg method based on the CTG strategy; k s,1 , k v,1 , k a,1 , k s,2 , k v,2 , k a,2 represents the control parameters of the car-following control model using the CTG spacing strategy, k s,1 , k v,1 , k a,1 represents the parameters in the Bian method (Li et al., 2022), k s,2 , k v,2 , k a,2 represents the parameters in the Ploeg method (Zheng et al., 2023a); σ i represents the cumulative time delay between the lead vehicle and the target vehicle, that is, q 1 ,q 3 ,q 4 , and λ are the control parameters of the CS-based control model.

[0066] Among them, this embodiment adopts a linear third-order model as the dynamic model of all CAVs. The following control models of three representative CAV queues, CAVP with different CTG strategies, CAVP with CS strategy, and CAVP with CTG and CS strategies, are described as follows. CTG-CAVP: The expected acceleration of the CAVs following control model is calculated based on the above partial formula. For CTG-CAVP, CAVs with CTG strategy using Bian's and Ploeg's methods can be randomly distributed in queues of the same size. Therefore, this embodiment will study the mixed random distribution of CTG-based CAV control models in queues. A large number of randomly generated CAV queues can be used for case study analysis; CS-CAVP: Based on the above partial formula, the expected acceleration of the CS-based control model in CS-CAVP is calculated. For CS-CAVP, CAVs with CS strategy can form a CAV queue for case analysis; CC-CAVP: Based on the above formula, the expected acceleration of the CTG-based control model in CC-CAVP and the expected acceleration of the CS-based control model in CC-CAVP are calculated. For CC-CAVP, CAVs with CTG and CS strategies can be randomly distributed in cohorts with the same size. A large number of randomly generated CAV cohorts can be used for case study analysis.

[0067] In highway scenarios with unstable traffic flow, the CC strategy can switch the appropriate control mode according to the real-time traffic flow; in congested urban environments where road utilization needs to be improved, the CC strategy can effectively reduce queue gaps.

[0068] In the specific experiment, the vehicles in the CAV queue can be randomly distributed based on the two strategies of CTG and CS to form a mixed queue mode, and the perception system and V2V communication system are used to obtain the speed, acceleration and position information of the preceding vehicle to achieve real-time control. The mixed queue can improve the overall performance of the queue by optimizing the comprehensive indicators of emission, energy consumption, efficiency and safety performance, and determining the optimized vehicle distribution mode and comprehensive spacing strategy based on fuzzy grey correlation analysis and ideal solution similarity ranking method (fuzzy GRA-TOPSIS).

[0069] (2) Multi-criteria decision-making evaluation system

[0070] A multi-criteria evaluation system including emissions, energy consumption, efficiency and safety is designed to meet the diverse needs of decision makers for fleet performance through different weight settings; the standard system for measuring CAVs platoon driving performance consists of four performance criteria, namely emissions, energy consumption, efficiency and safety. These four evaluation indicators take into account four criteria, namely carbon dioxide emissions, fuel consumption, traffic throughput and collision time.

[0071] (3) Fuzzy logarithmic least squares (FLLS) method is used to calculate the weight of the criteria. In the decision-making process, the weight of each criterion is set according to the preferences of different decision makers. Decision makers express the relative importance of different criteria through linguistic variables, which include "extremely important", "very important", "moderately important", etc. This embodiment uses fuzzy logarithmic least squares (FLLS) method to determine the weight of the criteria to reflect the relative importance of emissions, energy consumption, efficiency and safety in the evaluation system. The steps include:

[0072] (3.1) Use linguistic variables to represent the fuzzy comparison matrix;

[0073] (3.2) The linguistic variables of the comparison matrix are converted into fuzzy numerical values ​​for calculation, and the comparison matrix of the kth decision maker is obtained:

[0074]

[0075] Where n represents the number of criteria, They represent the three element values ​​of the k-th decision maker's criterion i compared to criterion j, where the character u represents upper, m represents middle, and l represents lower;

[0076] (3.3) Based on the weighted geometric mean method, the fuzzy comparison matrices of K (K ≥ 1) decision makers are summarized and the average comparison matrix is ​​calculated:

[0077]

[0078] (3.4) Construct the FLLS method to determine the weight of each criterion:

[0079]

[0080] in, is the weight of the ith criterion, and Yes i The three elements of ij is the importance of criterion i compared to criterion j, Corresponding to three element values ​​respectively.

[0081] (4) After determining the weights, the present invention uses the Grey Relational Analysis-Topsis Method (GRA-TOPSIS) to perform multi-criteria sorting on the CAV queue strategy. The specific steps are as follows:

[0082] (4.1) Numerical experiments on CAV platoons under different car-following strategies;

[0083] (4.2) Determine the decision matrix based on the simulation results;

[0084] (4.3) The values ​​of each criterion in the decision matrix of different CAV queues are normalized, that is, r ij :

[0085]

[0086] Where m represents the number of CAV queues, y ij represents the calculated value of the jth criterion of the i-th CAV queue;

[0087] (4.4) According to the criterion weight results and the obtained normalized results, the weighted normalized decision matrix is ​​calculated. For the convenience of calculation, the weight of each indicator is converted into a clear numerical value. In this embodiment The calculation formula of the weighted normalization matrix is ​​as follows:

[0088]

[0089] x ij =ω j r ij

[0090] Among them, x ij is the weighted value of the jth criterion of the i-th CAV queue, and n is the number of criteria;

[0091] (4.5) Calculate the best possible ideal solution PIS and the worst possible solution NIS respectively:

[0092]

[0093] (4.6) The grey correlation coefficient between each indicator and PIS and NIS can be expressed as follows:

[0094]

[0095] (4.7) The grey relational grade of each CAV cohort for PIS and NIS is calculated as follows:

[0096]

[0097] (4.8) Determine the relative closeness of the grey relational degree of the CAV queues, and compare based on this value to obtain the sorting order of the CAV queues:

[0098]

[0099] Among them, μ i It represents the relative closeness of the grey relational degree of the i-th CAV queue, and its value is in the interval [0,1].

[0100] (5) Simulation Experiment Verification,In order to verify the effectiveness of the system, a series of simulation experiments are designed in the embodiment of the present invention, using real vehicle trajectories in the NGSIM dataset.

[0101] The exogenous vehicle selection takes the NGSIM vehicle data with ID 1941 as an example to simulate the oscillation and dynamic changes caused by the exogenous vehicle in the queue. It is divided into two stages: the first stage is a large deceleration and acceleration fluctuation; the second stage is a uniform speed driving with a stable speed. The main parameters of the experiment include simulation time T = 80 seconds, time step Δt = 0.01 seconds, fixed time interval h = 1 second, stopping distance d_s = 10 meters or 20 meters, sensor delay δ = 0.02 seconds, communication delay θ between 0.02 seconds and 0.1 seconds, etc. Regarding data collection and decision matrix construction, under different strategies, the vehicle's carbon dioxide emissions, fuel consumption, traffic flow and collision time data are collected to form a decision matrix for analysis.

[0102] Through this simulation experiment, the differences in environment, energy consumption, efficiency and safety of different CAV platoon strategies were verified, providing reliable data support for the practical application of the system.

[0103] (6) Application cases and decision maker analysis. The system is applied to the following two typical decision makers:

[0104] (6.1) Enterprise managers (DM1): focus on reducing energy consumption and improving queue efficiency. The FLLS model sets a higher energy consumption weight, making the CS strategy the best choice due to its higher energy efficiency.

[0105] (6.2) Administrative managers (DM2): They are more concerned about the safety and environmental benefits of vehicles. In this scenario, the CTG strategy is more suitable as the preferred strategy for administrative managers due to its advantages in safety and emissions.

[0106] Based on the same inventive concept, an embodiment of the present invention discloses a multi-performance evaluation system for intelligent connected vehicle platoon driving, including: a following control module, which is used to establish a CAV queue control model that adapts to different traffic conditions based on different following strategies; the different following strategies include one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of mixed CTG and CS (CTG-CS); an evaluation index module, which is used to establish a multi-criteria evaluation system, and by setting different weights for different criteria, it meets the diverse needs of decision makers for platoon driving performance; the multi-criteria evaluation system includes emissions, One or more performance criteria among energy consumption, efficiency and safety; a weight determination module, which is used to accurately calculate the weight of each criterion based on the decision maker's linguistic variable data and apply the fuzzy logarithmic least squares (FLLS) method. The obtained weight can reflect the priority of different decision makers for the comparison between different criteria; and a comprehensive ranking module, which is used to embed grey relational analysis (GRA) into the top-ranking method of ideal solution similarity (TOPSIS) to construct a GRA-TOPSIS method, and calculate and evaluate the comprehensive evaluation indicators of CAV platoon driving based on simulation data or measured data, so as to rank the evaluation results of different strategies.

[0107] The system provided in the embodiment of the present invention can be used to execute the method provided in the embodiment of the present invention, and has the corresponding functions and beneficial effects of the execution method. It is worth noting that the method in the embodiment of the above system can be used as a decision-making tool to determine the best queue among the candidate CAV queues for two different types of decision makers.

[0108] An embodiment of the present invention further discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for evaluating multiple performances of intelligent connected vehicle platoon driving.

[0109] It should be understood that the above embodiments and descriptions only describe the principles, main features and advantages of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, and these changes and improvements all fall within the scope of protection of the present invention.

Claims

1. A multi-performance evaluation method for intelligent connected vehicle platooning, characterized in that: The steps include: (1) Establishing a CAV platoon control model that adapts to different traffic conditions based on different car-following strategies; the different car-following strategies include one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of mixed CTG and CS (CTG-CS); (2) Establishing a multi-criteria evaluation system to meet the diverse needs of decision makers for platooning performance by setting different weights for different criteria; the multi-criteria evaluation system includes one or more performance standards of emissions, energy consumption, efficiency, and safety; (3) Based on the decision makers' linguistic variable data, the fuzzy logarithmic least squares (FLLS) method is used to accurately calculate the weight of each criterion. The obtained weight can reflect the priority of different decision makers in comparing different criteria. (4) Grey relational analysis (GRA) is embedded into the topology of ideal solution similarity (TOPSIS) to construct the GRA-TOPSIS method. The comprehensive evaluation index of CAV platooning performance is calculated based on simulation data or measured data, so as to rank the evaluation results of different strategies.

2. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 1, characterized in that: The measurement of CAV platoon driving performance is evaluated and verified through data generated by the simulation system. First, the vehicle position and speed data are generated to simulate the driving trajectory of the CAV queue. Then, the set following control model and control parameters are used to calculate one or more vehicle performance including emissions, energy consumption, efficiency and safety. The data is then organized into a decision matrix. Finally, the FLLS and GRA-TOPSIS methods are used for data analysis to calculate the normalized weight values ​​and obtain the performance of the CAV queue under different strategies.

3. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 1, characterized in that: The fixed time interval (CTG) strategy dynamically adjusts the following behavior of the vehicle by setting a fixed time interval and calculating the expected distance between vehicles based on the product of the time interval and the speed of the following vehicle. The fixed spacing (CS) strategy ensures a stable desired spacing between vehicles by maintaining a fixed distance between vehicles. Under the hybrid strategy, the vehicles in the CAV queue are randomly distributed based on the CTG and CS strategies to form a hybrid queue mode. The perception system and V2V communication system are used to obtain the speed, acceleration and position information of the preceding vehicle to achieve real-time control.

4. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 3, characterized in that: The mixed platoon optimizes the comprehensive indicators of emissions, energy consumption, efficiency and safety, and determines the optimized vehicle distribution pattern and comprehensive spacing strategy based on fuzzy grey correlation analysis and ideal solution similarity ranking method (fuzzy GRA-TOPSIS), thereby improving the overall performance of platooning.

5. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 1, characterized in that: The four evaluation indicators in the multi-criteria evaluation system including emissions, energy consumption, efficiency and safety correspond to carbon dioxide emissions, fuel consumption, traffic throughput and collision time respectively.

6. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 1, characterized in that: The FLLS method is used to accurately calculate the weight of each criterion, which specifically includes the following steps: (3.1) Use linguistic variables to represent the fuzzy comparison matrix; (3.2) The linguistic variables of the comparison matrix are converted into fuzzy numerical values ​​for calculation, and the comparison matrix of the kth decision maker is obtained: Where n represents the number of criteria, They represent the three element values ​​of the k-th decision maker's criterion i compared to criterion j, where the character u represents upper, m represents middle, and l represents lower; (3.3) Based on the weighted geometric mean method, the fuzzy comparison matrices of K decision makers are summarized and the average comparison matrix is ​​calculated as follows: (3.4) Construct the FLLS method to determine the weight of each criterion: in, is the weight of the ith criterion, and Yes i The three elements of ij is the importance of criterion i compared to criterion j, Corresponding to three element values ​​respectively.

7. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 6, characterized in that: When K=1 or K>1, K decision makers are of the same type.

8. The method for evaluating the multi-performance of intelligent connected vehicle platooning according to claim 1, characterized in that: The performance ranking of different CAV queues is performed based on the GRA-TOPSIS method, which includes the following steps: (4.1) Numerical experiments on CAV platoons under different car-following strategies; (4.2) Determine the decision matrix based on the simulation results; (4.3) The values ​​of each criterion in the decision matrix of different CAV queues are normalized, that is, r ij : Where m represents the number of CAV queues, y ij represents the calculated value of the jth criterion of the i-th CAV queue; (4.4) According to the criterion weight results and the obtained normalized results, the weighted normalized decision matrix is ​​calculated. For the convenience of calculation, the weights of each indicator are converted into clear numerical values. The calculation formula of the weighted normalized matrix is ​​as follows: x ij =ω j r ij Among them, x ij is the weighted value of the jth criterion of the i-th CAV queue, and n is the number of criteria; (4.5) Calculate the best possible ideal solution PIS and the worst possible solution NIS respectively: (4.6) The grey correlation coefficient between each indicator and PIS and NIS can be expressed as follows: (4.7) The grey relational grade of each CAV cohort for PIS and NIS is calculated as follows: (4.8) Determine the relative closeness of the grey relational degree of the CAV queues, and compare based on this value to obtain the sorting order of the CAV queues: Among them, μ i It represents the relative closeness of the grey relational degree of the i-th CAV queue, and its value is in the interval [0,1].

9. A multi-performance evaluation system for intelligent connected vehicle platooning, characterized in that: include: The car-following control module is used to establish a CAV platoon control model that adapts to different traffic conditions based on different car-following strategies; The different car-following strategies include one or more of a fixed time interval (CTG) strategy, a fixed spacing (CS) strategy, and a combination strategy of hybrid CTG and CS (CTG-CS); Evaluation index module, which is used to establish a multi-criteria evaluation system. By setting different weights for different criteria, it can meet the diverse needs of decision makers for platoon driving performance. The multi-criteria evaluation system includes one or more performance criteria among emissions, energy consumption, efficiency and safety; The weight determination module is used to accurately calculate the weight of each criterion based on the decision maker's language variable data and apply the fuzzy logarithmic least squares (FLLS) method. The obtained weight can reflect the priority of different decision makers for the comparison between different criteria. And a comprehensive ranking module, which is used to embed grey relational analysis (GRA) into the ideal solution similarity ranking method (TOPSIS) to construct a GRA-TOPSIS method, and calculate and evaluate the comprehensive evaluation indicators of CAV platoon driving multiple performances based on simulation data or measured data, so as to rank the evaluation results of different strategies.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by the processor, the steps of a multi-performance evaluation method for intelligent connected vehicle platoon driving are implemented according to any one of claims 1 to 8.

Images