Heterogeneous network connection fleet communication topological structure optimization method based on multi-objective optimization

Abstract

A heterogeneous networked fleet communication topological structure optimization method based on multi-objective optimization relates to the technical field of networked fleet control, and comprises the following steps: S1, obtaining an initial communication topological population based on a population initialization strategy, and taking the initial communication topological population as a parent communication topological population; s2, loading dynamic models of the leading vehicle and the following vehicle, and a self-adaptive distributed controller of the following vehicle; s3, selecting a parent communication topology from the parent communication topology population, and obtaining a child communication topology after improving a crossover operator, variation and a competitive evolution mechanism; s4, comparing the parent communication topology with the child communication topology according to the fitness function, wherein the optimal parent communication topology is used as the parent communication topology of the next population; s5, judging whether the communication topology population reaches the maximum population number or not, and if not, circulating S3 and S4 until the maximum population number is reached; and if yes, screening is carried out, and an optimal communication topological structure is output, so that the overall cooperative performance of the heterogeneous network connection motorcade is improved.

Description

Technical Field

[0001] This invention relates to the field of connected vehicle fleet control technology, specifically a method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization. Background Technology

[0002] With the rapid development of intelligent transportation systems, heterogeneous connected vehicle fleets have shown great potential in improving traffic efficiency, reducing energy consumption, and enhancing driving safety. However, in the practical application of connected vehicle fleets, two core challenges are faced: first, the heterogeneity of vehicles (such as vehicle type, power system, sensor accuracy, etc.) makes it difficult to adapt to traditional homogeneous fleet control methods, and the inconsistency of dynamic characteristics can cause queue stability problems; second, the design of the communication topology directly affects the efficiency of information interaction and the accuracy of controller output.

[0003] Existing technologies for heterogeneous connected vehicle fleets have significant shortcomings. Firstly, to fully consider the inherent impact and constraints of vehicle heterogeneity on communication topology, many technologies employ fixed communication topologies or optimize based on a single performance metric (such as energy consumption minimization or queue stability). This makes it difficult to achieve coordinated optimization and balance of multiple objectives, such as energy consumption, stability, and traffic efficiency, in complex and ever-changing real-world traffic scenarios. Secondly, traditional optimization algorithms (such as genetic algorithms and particle swarm optimization) used to solve for optimal communication topologies suffer from inherent defects when dealing with such high-dimensional, nonlinear, and multi-constrained complex solution spaces. These defects include slow convergence speed and a tendency to get trapped in local optima, making it difficult to efficiently and reliably search for globally optimal or near-optimal communication topologies. Summary of the Invention

[0004] The purpose of this invention is to provide a method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization, so as to obtain the optimal communication topology and thereby improve the overall collaborative performance of heterogeneous connected vehicle fleets.

[0005] To achieve the above objectives, the specific solution adopted in this invention is as follows: a method for optimizing the communication topology of a heterogeneous connected vehicle fleet based on multi-objective optimization, obtaining real driving cycle data; constructing a multi-objective optimization function for the communication topology; training the multi-objective optimization function based on an improved non-dominated sorting genetic algorithm to obtain the optimal solution set (obtaining the optimal communication topology); the optimization method based on the improved non-dominated sorting genetic algorithm includes the following steps:

[0006] S1. Obtain the initial communication topology population based on the population initialization strategy and use it as the parent generation communication topology population.

[0007] S2, loads the dynamic models of the lead vehicle and the follower vehicle, as well as the adaptive distributed controller of the follower vehicle;

[0008] S3: Select the parent generation communication topology from the parent generation communication topology population, and obtain the offspring communication topology after improving the crossover operator, mutation and competitive evolution mechanism;

[0009] S4. Compare the parent generation communication topology and the offspring generation communication topology based on the fitness function, including the optimal parent generation communication topology as the next population.

[0010] S5 determines whether the communication topology population has reached the maximum population size. If not, repeat S3 and S4 until the maximum population size is reached. If so, perform filtering and output the optimal communication topology structure.

[0011] As one optimization scheme of the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: the construction of the multi-objective optimization function of the communication topology includes defining the performance evaluation index of the communication topology and constructing the multi-objective optimization function based on the performance evaluation index.

[0012] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization, the performance evaluation indicators include latency stability margin, driving comfort, tracking performance, safety, and communication link cost.

[0013] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization, the time delay stability margin index is:

[0014]

[0015]

[0016] Where c represents the coupling strength of vehicles within the platoon, and k p and k v Let λ represent the controller gain of the following vehicle within a heterogeneous connected vehicle fleet. i Let λ represent the i-th eigenvalue of the communication topology matrix. max This represents the largest eigenvalue of the communication topology matrix.

[0017] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization, the multi-objective optimization function is:

[0018] f(t)=w1f1+w2f2+w3f3+w4f4+w5f5

[0019] Among them, w1, w2, w3, w4 and w5 are weighting factors, f1 is the time delay stability margin index function, f2 is the driving comfort function, f3 is the tracking function, f4 is the safety function, and f5 is the communication link cost function.

[0020] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: In S1, an initial communication topology population is obtained based on a population initialization strategy, multiple individuals are generated using chaotic mapping to form a basic communication topology population, and the classic communication topology is inserted into the basic communication topology population in a random distribution to obtain the initial communication topology population.

[0021] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: the chaotic mapping is:

[0022]

[0023] Where, α g =0.2, β g =0.5, where n represents the number of following vehicles in the convoy. The generated initial communication topology population values ​​are rounded down.

[0024] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: In S2, the dynamic model of the lead vehicle is as follows:

[0025]

[0026] Where v0(t) is the velocity of the lead car and a0(t) is the acceleration of the lead car.

[0027] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: In S2, the dynamic model of the following vehicle is:

[0028]

[0029]

[0030] Where, p i (t) and v i (t) represents the longitudinal position and velocity of the following vehicle i, respectively; u i (t) is the controller output of the following vehicle i, which controls the speed change of the following vehicle; M a M is the sum of the masses of all vehicles in a heterogeneous connected vehicle fleet. i For the mass of the following vehicle i; τ i C is the time constant for the response speed of the vehicle's powertrain system. ea v is the equivalent drag coefficient of the following vehicle i; w (t) represents the relative ambient wind speed during the car's movement; g represents the gravitational acceleration, with a value of 9.8 m / s². 2μ represents the rolling resistance coefficient of the following vehicle i; θ is the road gradient, a i (t) represents the acceleration of the following vehicle i.

[0031] As another optimization scheme for the above-mentioned heterogeneous connected vehicle fleet communication topology optimization method based on multi-objective optimization: In S2, the improved crossover operator is:

[0032] When q≤0.5,

[0033]

[0034] When q > 0.5,

[0035]

[0036] Where q represents the random probability variable of the crossover operation, and g 1,i and g 2,i Let m represent the decimal values ​​of the i-th gene segment in the parent and mother generations, respectively. 1,i and m 2,i represents the decimal value of the i-th gene segment in the two generated offspring communication topologies, respectively, and |N(0,1)| is a normally distributed random variable.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] This invention provides a method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization. It considers the heterogeneity of vehicles and the instability of communication in connected vehicle fleets, reflects the diversity of vehicles in actual connected vehicle fleets, and improves the performance of vehicle control. At the same time, it introduces a population initialization strategy, an improved crossover operator, and a competitive evolution mechanism, enabling the NSGA-II algorithm to effectively explore the solution space and obtain the optimal communication topology. Attached Figure Description

[0039] Figure 1 This is a flowchart of an optimization method based on an improved non-dominated sorting genetic algorithm;

[0040] Figure 2 This is a framework diagram of an improved non-dominated sorting genetic algorithm. Detailed Implementation

[0041] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. Parts not described or disclosed in detail in the following embodiments of the present invention should be understood as prior art known or should be known by those skilled in the art.

[0042] Example 1

[0043] A method for optimizing the communication topology of a heterogeneous connected vehicle fleet based on multi-objective optimization is proposed. This method acquires real driving cycle data and constructs a multi-objective optimization function for the communication topology. The construction of the multi-objective optimization function includes defining performance evaluation metrics for the communication topology and constructing the multi-objective optimization function based on these metrics. The performance evaluation metrics include latency stability margin, driving comfort, tracking performance, safety, and communication link cost. Specifically:

[0044] The latency stability margin metric measures the tolerance of a heterogeneous connected vehicle fleet control system to communication latency. A larger latency stability margin indicates that the heterogeneous connected vehicle fleet can maintain stable operation even when faced with a certain range of communication delays, and will not become unstable due to information transmission delays. This performance evaluation metric ensures the reliability and safety of the heterogeneous connected vehicle fleet under different communication environments, and its expression is:

[0045]

[0046]

[0047] Where c represents the coupling strength of vehicles within the platoon, and k p and k v Let λ represent the controller gain of the following vehicle within a heterogeneous connected vehicle fleet. i Let λ represent the i-th eigenvalue of the communication topology matrix. max This represents the largest eigenvalue of the communication topology matrix.

[0048] Driving comfort, specifically the standard deviation of acceleration, is used to represent the smoothness of changes in the speed of the following vehicle. It aims to characterize the smoothness of the following vehicle during operation, evaluating the dynamic performance of a heterogeneous connected vehicle fleet by measuring the smoothness of vehicle acceleration changes and their impact on passenger experience. Frequent or abrupt acceleration and deceleration not only significantly affect passenger experience but also lead to excessive wear on the vehicle's powertrain and energy waste, negatively impacting the long-term operating efficiency of the heterogeneous connected vehicle fleet. Therefore, a smaller driving comfort evaluation function value indicates a smoother following process, smaller acceleration changes, and higher energy efficiency. The driving comfort evaluation function is:

[0049]

[0050] Where std represents the standard deviation calculation function, used to calculate the standard deviation of acceleration over a specific time t; N represents the number of vehicles in a heterogeneous connected vehicle fleet; a i (t) represents the acceleration of the following vehicle i in the heterogeneous connected vehicle fleet.

[0051] Tracking performance, in this embodiment, is assessed by constructing a tracking performance evaluation function to calculate the average spacing error and speed error of a heterogeneous connected vehicle convoy, thereby better evaluating the tracking performance of the convoy. This performance evaluation index measures the ability of vehicles in a heterogeneous connected vehicle convoy to follow the vehicle in front, that is, whether the vehicle can quickly and accurately respond to changes in the speed and trajectory of the vehicle in front while maintaining a safe distance. The tracking performance evaluation function is as follows:

[0052]

[0053] Where N represents the number of vehicles in the heterogeneous connected vehicle fleet, SDE represents the spacing error sensitivity, SVE represents the speed error sensitivity, and t all p represents the total time. i (t) represents the position information of the following vehicle i, v i (t) represents the speed information of the following vehicle i, p i-1 (t) represents the position information of the following vehicle i-1, v i-1 (t) represents the speed information of the following vehicle i-1.

[0054] Safety, in the context of heterogeneous connected vehicle fleet control, is primarily reflected in the collision risk among vehicles within the fleet. Therefore, the actual collision time is chosen as the evaluation metric for the safety of heterogeneous connected vehicle fleets. This metric measures the collision risk between two adjacent vehicles within the fleet, with a smaller value indicating a lower collision risk. Its expression is:

[0055]

[0056]

[0057] Among them, ITC i (t) represents TTC i The reciprocal of (t), TTC i (t) represents the actual collision time of the vehicle.

[0058] Communication link cost is used to evaluate the communication topology cost of a heterogeneous connected vehicle fleet. The lengths of information transmission links within a heterogeneous connected vehicle fleet vary. The cost of each information transmission link is calculated equivalently based on its length. The link cost between adjacent vehicles is set to 1 and accumulated sequentially. Its expression is:

[0059]

[0060] Among them, Π i This represents the set of vehicles communicating with following vehicle i in the communication topology, where i represents following vehicle i, b represents the vehicle number communicating with following vehicle i, and N represents the number of following vehicles in the heterogeneous connected vehicle fleet.

[0061] Therefore, the multi-objective optimization function is:

[0062] f(t)=w1f1+w2f2+w3f3+w4f4+w5f5

[0063] Among them, w1, w2, w3, w4 and w5 are weighting factors, f1 is the time delay stability margin index function, f2 is the driving comfort function, f3 is the tracking function, f4 is the safety function, and f5 is the communication link cost function.

[0064] The optimal communication topology is obtained by training a multi-objective optimization function based on an improved non-dominated sorting genetic algorithm (NSGA-II). The optimization method based on the improved non-dominated sorting genetic algorithm (NSGA-II) includes the following steps:

[0065] S1. Based on the population initialization strategy and driving cycle data, an initial communication topology population is obtained and used as the parent communication topology population. Multiple individuals are generated using chaotic mapping to form a basic communication topology population. Classical communication topologies are then randomly distributed and inserted into the basic communication topology population to obtain the initial communication topology population. Chaotic mapping can generate individuals with randomness, thereby increasing the diversity and exploration capability of the initial communication topology population and improving the global search performance of the NSGA-II algorithm. In other words, chaotic mapping provides a high-quality initial communication topology population for the NSGA-II algorithm, which helps the NSGA-II algorithm converge to the global optimum more quickly. In this embodiment, Circle chaotic mapping is used, specifically:

[0066]

[0067] Where, α g =0.2, β g =0.5, where n represents the number of following vehicles in a heterogeneous connected vehicle fleet. The generated initial communication topology population values ​​are rounded to ensure consistency with the binary format of the communication topology. The initial communication topology population generated using the Circle chaotic mapping has the advantage of uniform distribution, thereby expanding the exploration space during the population iteration process.

[0068] S2 loads the dynamic models of the lead vehicle and the following vehicle, as well as the adaptive distributed controller of the following vehicle. Based on the traditional vehicle dynamics model, considering the heterogeneity of vehicles caused by differences in vehicle power systems and the differences in vehicle mass, a third-order heterogeneous dynamics model is constructed. First, the dynamics model of the vehicles in the heterogeneous connected vehicle fleet is constructed in two parts: the lead vehicle and the following vehicle. The lead vehicle is driven by a human and its kinematic characteristics are described using a traditional kinematic model. The dynamics model of the lead vehicle is as follows:

[0069]

[0070] Where v0(t) is the velocity of the lead car and a0(t) is the acceleration of the lead car.

[0071] The following vehicles consist of autonomous vehicles. Taking into account the differences in powertrain transmission delay and actual load weight, a heterogeneous dynamics model is constructed to more accurately describe the kinematic characteristics of the vehicles. The dynamics model of the following vehicles is as follows:

[0072]

[0073]

[0074] Where, p i (t) and v i (t) represents the longitudinal position and velocity of the following vehicle i, respectively; u i (t) is the controller output of the following vehicle i, which controls the speed change of the following vehicle; M a M is the sum of the masses of all vehicles in a heterogeneous connected vehicle fleet. i For the mass of the following vehicle i; τ i C is the time constant for the response speed of the vehicle's powertrain system. ea v is the equivalent drag coefficient of the following vehicle i; w (t) represents the relative ambient wind speed during the car's movement; g represents the gravitational acceleration, with a value of 9.8 m / s². 2 μ represents the rolling resistance coefficient of the following vehicle i; θ is the road gradient, a i (t) represents the acceleration of the following vehicle i.

[0075] In the entire heterogeneous connected vehicle fleet, the vehicle's position, velocity, and acceleration information are used as state information. A third-order state-space model is then used to describe the longitudinal dynamics of each vehicle. The state-space expression of the longitudinal dynamics model is as follows:

[0076]

[0077] Wherein, the state vector x i =[p i ,v i ,a i ] T w i (t) represents the disturbance experienced by the vehicle.

[0078] The controller uses the position, speed, and acceleration information of other vehicles in the heterogeneous connected vehicle fleet to control the acceleration of the following vehicle. The controller for following vehicle i is:

[0079]

[0080] Among them, Π i The set of all vehicles that exchange information with following vehicle i is represented by the communication topology matrix; c represents the coupling strength of the heterogeneous connected vehicle fleet under the influence of the communication topology; k p k v and k a Let Si and SV represent the controller gain of the following vehicle i, respectively; SDE and SVE represent the position error sensitivity and speed error sensitivity, respectively; τ(t) represents the communication delay of the information transmission link between vehicles in a heterogeneous connected vehicle fleet, which is continuously differentiable for any t ≥ 0, and

[0081] The controller gain problem is transformed into a linear matrix inequality problem, i.e.:

[0082]

[0083] Where, matrix Q > 0, scalar α d >0, β d >0, C1 = [1 0 0].

[0084] Matrix Q and scalar α obtained based on the linear matrix inequality method d The controller gain of vehicle i and the coupling strength between vehicles are:

[0085]

[0086] Where, K = [k p ,k v ,k a ] T ;λ min The smallest eigenvalue of the communication topology matrix T is represented by c, and the coupling strength c takes the value α. d / λ min .

[0087] S3: From the parent generation of communication topologies, select the parent generation communication topology and, after improving the crossover operator, mutation, and competitive evolution mechanism, obtain the offspring communication topology. The improved crossover operator is a normally distributed crossover operator. By introducing the randomness of the normal distribution, the search space of the algorithm can be significantly expanded, enhancing the global search capability and effectively avoiding local optima. This not only improves the performance of the NSGA-II algorithm in high-dimensional complex optimization problems but also provides a more reliable solution for optimizing the communication topology of heterogeneous connected vehicle fleets. The normally distributed crossover operator is:

[0088] When q≤0.5,

[0089]

[0090] When q > 0.5,

[0091]

[0092] Where q represents the random probability variable of the crossover operation, and g 1,i and g 2,i Let m represent the decimal values ​​of the i-th gene segment in the parent and mother generations, respectively. 1,i and m 2,i represents the decimal value of the i-th gene segment in the two generated offspring communication topologies, respectively, and |N(0,1)| is a normally distributed random variable.

[0093] A competitive evolutionary mechanism is introduced after the mutation operation to determine whether an information transmission link exists between the lead vehicle and the follower vehicles in the offspring communication topology. If no information transmission link exists, an information transmission link is constructed between the lead vehicle and the follower vehicles while keeping other communication links unchanged, generating a temporary communication topology. The fitness functions of the temporary communication topology and the parent communication topology are compared, and the optimal one is selected as the offspring communication topology. This method can improve the quality of individuals during the optimization process, thereby improving optimization efficiency.

[0094] S4. Compare the parent generation communication topology and the offspring generation communication topology based on the fitness function, including the optimal parent generation communication topology as the next population.

[0095] S5 determines whether the communication topology population has reached the maximum population size. If not, repeat S3 and S4 until the maximum population size is reached. If so, perform filtering and output the optimal solution set.

[0096] In this embodiment, inspired by the greedy strategy in reinforcement learning, a selection execution probability is added to the competitive evolution mechanism to avoid getting trapped in local optima. Each generated offspring communication topology has p. s The probability of executing a competitive evolutionary mechanism, and p s for:

[0097]

[0098] Where ε1 and ε2 represent the selection probabilities in the early and late iterations of the NSGA-II algorithm, respectively, and G is the population size generated by the algorithm iterations. max The maximum population size set for the algorithm.

[0099] The optimization method enables flexible control of the communication topology of heterogeneous connected fleets, which not only optimizes the overall fleet control accuracy in the region but also improves driving comfort. At the same time, the optimization method significantly improves learning efficiency and control effect, providing an efficient and accurate solution for complex fleet management.

[0100] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization, characterized in that: Obtain real-world driving cycle data; Constructing a multi-objective optimization function for communication topology; The optimal communication topology is obtained by training the multi-objective optimization function based on the improved non-dominated sorting genetic algorithm. The optimization method based on the improved non-dominated sorting genetic algorithm includes the following steps: S1. Obtain the initial communication topology population based on the population initialization strategy, and use it as the mother generation and the parent generation communication topology population. S2, load the dynamic models of the lead vehicle and the follower vehicle, as well as the adaptive distributed controller of the follower vehicle; S3: Select the parent generation communication topology from the parent generation communication topology population, and obtain the offspring communication topology after improving the crossover operator, mutation and competitive evolution mechanism; S4. Compare the parent generation communication topology and the offspring generation communication topology based on the fitness function, including the optimal parent generation communication topology as the next population. S5, determine whether the communication topology population has reached the maximum population size. If not, loop through S3 and S4 until the maximum population size is reached. If so, then filter and output the optimal communication topology.

2. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that: Constructing a multi-objective optimization function for a communication topology involves defining performance evaluation metrics for the communication topology and constructing the multi-objective optimization function based on these metrics.

3. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 2, characterized in that: Performance evaluation metrics include latency stability margin, driving comfort, tracking performance, safety, and communication link cost.

4. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 3, characterized in that: The time delay stability margin index is: Where c represents the coupling strength of vehicles within the platoon, and k p and k v Let λ represent the controller gain of the following vehicle within a heterogeneous connected vehicle fleet. i Let λ represent the i-th eigenvalue of the communication topology matrix. max This represents the largest eigenvalue of the communication topology matrix.

5. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that, The multi-objective optimization function is: f(t)=w1f1+w2f2+w3f3+w4f4+w5f5 Among them, w1, w2, w3, w4 and w5 are weighting factors, f1 is the time delay stability margin index function, f2 is the driving comfort function, f3 is the tracking function, f4 is the safety function, and f5 is the communication link cost function.

6. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that: In step S1, an initial communication topology population is obtained based on a population initialization strategy. Multiple individuals are generated using chaotic mapping to form a basic communication topology population. Classical communication topologies are inserted into the basic communication topology population in a random distribution to obtain the initial communication topology population.

7. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 5, characterized in that: The chaotic mapping is: Where, α g =0.2, β g =0.5, where n represents the number of following vehicles in the convoy. The generated initial communication topology population values ​​are rounded down.

8. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that: In S2, the dynamic model of the lead vehicle is as follows: Where v0(t) is the velocity of the lead car and a0(t) is the acceleration of the lead car.

9. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that: In S2, the dynamic model of the following vehicle is: Where, p i (t) and v i (t) represents the longitudinal position and velocity of the following vehicle i, respectively; u i (t) is the controller output of the following vehicle i, which controls the speed change of the following vehicle; M a M is the sum of the masses of all vehicles in a heterogeneous connected vehicle fleet. i For the mass of the following vehicle i; τ i C is the time constant for the response speed of the vehicle's powertrain system. ea v is the equivalent drag coefficient of the following vehicle i; w (t) represents the relative ambient wind speed during the car's movement; g represents the gravitational acceleration, with a value of 9.8 m / s². 2 μ represents the rolling resistance coefficient of the following vehicle i; θ is the road gradient, a i (t) represents the acceleration of the following vehicle i.

10. The method for optimizing the communication topology of heterogeneous connected vehicle fleets based on multi-objective optimization as described in claim 1, characterized in that: In S2, the improved crossover operator is: When q≤0.5, When q > 0.5, Where q represents the random probability variable of the crossover operation, and g 1,i and g 2,i Let m represent the decimal values ​​of the i-th gene segment in the parent and mother generations, respectively. 1,i and m 2,i represents the decimal value of the i-th gene segment in the two generated offspring communication topologies, respectively, and |N(0,1)| is a normally distributed random variable.

Images