A car-road cooperation roadside unit layout system and method for human-machine co-driving

Abstract

The present application relates to a kind of car-road coordination roadside unit layout methods for man-machine co-driving, it includes road network model optimization, time threshold optimization, objective function optimization, solution method optimization.The present application also relates to the car-road coordination roadside unit layout system for man-machine co-driving according to the layout method, the system includes information access module, roadside unit dynamic programming module, roadside unit edge computing module, switching control information publishing moduleThe car-road coordination roadside unit layout system and method for man-machine co-driving carry out system structure optimization and method topological optimization, establish better roadside unit layout topological model, and through the construction layout system, establish multi-objective optimization model, and propose based on the improved multi-objective quantum behavior particle swarm optimization algorithm, realize the information communication and control requirement of man-machine co-driving high density, high precision, low cost, while minimizing the number of roadside unit layout, maximum communication and control service vehicle coverage.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving, and specifically to a vehicle-road cooperative roadside unit deployment system and method for human-machine co-driving. Background Technology

[0002] The electrification, intelligentization, connectivity, and sharing of the automotive industry are gradually transitioning from concepts to practical applications. In recent years, the global development of autonomous vehicles has been rapid, with continuous increases in R&D investment and large-scale testing, verification, and demonstration applications. However, overall, according to SAE J3016 and GB / T40429, the development of autonomous driving globally is still in the early stages of L2-L3. Vehicles at this stage have two driving entities: a driver and an autonomous driving system. Traditional research focuses on single-vehicle intelligence, studying the switching of control between these entities. However, single-vehicle intelligent autonomous driving is easily affected by environmental conditions such as obstacles and inclement weather, and faces problems in target detection, trajectory prediction, and driver decision-making. Vehicle-to-everything (V2X) cooperative autonomous driving, through information interaction, collaborative perception, and collaborative decision-making control, can greatly expand the perception range and enhance the perception capabilities of a single vehicle, enabling deep decision-making in multiple scenarios and thus ensuring the safety of autonomous driving.

[0003] Although vehicle-road cooperative autonomous driving has become the clear technological roadmap for the development of autonomous driving in my country, different levels of autonomous vehicles have different requirements for road capabilities. Furthermore, my country's highway system is vast, and the conditions of roads vary across regions, leading to different needs for intelligent systems. Therefore, it is necessary to classify roads into intelligence levels based on the development of vehicle-road cooperative autonomous driving and the construction of intelligent roads in my country. Currently, research on road classification systems for vehicle-road cooperative autonomous driving has been conducted both domestically and internationally, dividing road intelligence levels into six levels from C0 to C5. It has been pointed out that high-level intelligent roads (C4-C5) are needed to support L2-L3 level autonomous vehicles to enhance their perception and decision-making capabilities.

[0004] As a crucial component of vehicle-road cooperative autonomous driving, Roadside Units (RSUs) have become an important research area due to their ability to interact with vehicles, provide road environment information to vehicles and pedestrians, support driving decisions, and assist in the networking and communication of Vehicular Ad-hoc Networks (VANETs). Besides RSU functional research, RSU deployment schemes are also a hot research topic. RSU deployment is constrained by various factors such as road network topology and geographical features, installation and maintenance costs, etc. In practical RSU deployment research, different deployment schemes for different scenarios and performance requirements are usually studied from multiple perspectives. According to different optimization objectives, related research mainly includes performance-optimized RSU deployment and cost-minimized RSU deployment.

[0005] In existing RSU deployment schemes, considering the high cost of RSU deployment, a hotspot deployment mode is often adopted, that is, placing RSUs at road intersections, congested or accident-prone road sections. The biggest advantage of this method is that the deployment and maintenance costs are low, and it can realize simple communication functions between vehicles and RSUs, providing vehicles with macro and long-term traffic information services. However, it is difficult to achieve precise control of vehicles and roads. Summary of the Invention

[0006] The technical problem this invention aims to solve is to overcome the shortcomings of existing RSU deployment schemes in achieving precise control of vehicles and roads. Starting from vehicle-road cooperative autonomous driving information interaction and collaboration, collaborative perception, and collaborative decision-making control, this invention optimizes the system structure and topology of the deployment of small-spaced, high-density vehicle-road cooperative roadside units (RSUs) in road networks. It establishes a better RSU deployment topology model to minimize the number of RSUs deployed and maximize the coverage of communication and control service vehicles. This provides a vehicle-road cooperative RSU deployment system and method for human-machine co-driving. By building a deployment system, establishing a multi-objective optimization model, and proposing an improved multi-objective quantum behavior particle swarm optimization algorithm, this invention solves the problem that traditional vehicle-road cooperative RSUs cannot adapt to precise control of vehicles and roads. It achieves high-density, high-precision, and low-cost information communication and control requirements for human-machine co-driving, minimizing the number of RSUs while maximizing the coverage of communication and control service vehicles.

[0007] This vehicle-road cooperative roadside unit deployment system and method for human-machine co-driving includes road network model optimization, time threshold optimization, objective function optimization, and solution method optimization.

[0008] Furthermore, the road network model optimization specifically involves: prioritizing road intersections as candidate RSU placement locations; before placing roadside units, abstracting the road network into a mathematical model using a planar undirected graph, denoted as G = (V, E); where G consists of a set of vertices and the set of edges between them, and V is the set of candidate RSU placement locations, V = {V1, V2, L, V...} x}, and satisfy V=x; E is the set of road segments connecting the candidate RSU deployment locations, E={E1,E2,L,E y The set of road segment lengths is D = {D1, D2, L, D}. y}. For D k ∈D and D k >2R section E k Further segmentation is performed, and within each segment, equal intervals of length L are inserted. A new node is selected as a candidate location for RSU deployment to maximize road network coverage and achieve precise control of vehicles and roads. After the partitioning is complete, a final undirected graph G′=(V′,E′) is formed, where V′={V,N}, and N is the set of nodes formed during the partitioning of E∈G=(V,E), N={N1,N2,L,N}. z And |V′|=m, |E′|=p; And e(i,j)={l ij ,ρ ij ,v ij}

[0009] Furthermore, the time threshold optimization specifically involves: The set of vehicles in the road network is represented by C, where C = {C1, C2, L, C...} n Let there be an m×n matrix T = (t ab ), its element t ab (1≤a≤m,1≤b≤n,t ab ≥0) indicates that during a certain time period, vehicle C b In V′ a The time a vehicle stays within the communication coverage area of ​​the deployed RSU; assuming the minimum time required for a vehicle to connect to the RSU and successfully communicate is τ1, communication between the RSU and the vehicle can be completed through multiple RSUs. If, during this time period, vehicle C... b If the total time spent within the communication coverage area of ​​each RSU deployment node is greater than τ1, then vehicle C is determined to be... b Communication services are available during this period. The control commands transmitted by the RSU are time-sensitive and are represented by an m×m matrix. in This indicates that during a certain period of time, vehicle C d (C d ∈C b The time taken for vehicle C to travel on road segment e(i,j); let τ be the maximum time required for a vehicle to connect to the RSU and successfully receive a certain control command, and the relationship between τ and τ1 is: τ = τ1 + τ2, where τ2 is the total time the vehicle stays on road segment e(i,j) during the period of connecting to the RSU and successfully communicating. If, during a certain period, vehicle C... b If the time to successfully receive a control command is less than τ, then vehicle C is determined to be... b The service can be controlled during this period.

[0010] Furthermore, the objective function optimization is as follows: Since the number of RSUs deployed depends on the spacing L when creating new nodes between road segments at road intersections, a length L that is as long as possible is set to achieve a larger coverage area for the RSUs. The multi-objective optimization function is constructed as follows:

[0011]

[0012]

[0013]

[0014]

[0015]

[0016] t(C b )+t(C d )<τ,C b ∈C r C d ∈C r (6)

[0017] In equation (1), f1 represents the reciprocal of the number of RSUs deployed in the road network. The larger the value of f1, the fewer RSUs need to be deployed in the road network.

[0018] In equation (2), C s Let |C represent the set of vehicles that can be served by communication within a specified time period. s | represents the total number of vehicles that can be served by communication during this period, and f2 represents the coverage rate of vehicles that can be served by communication during this period. The constraints are as shown in equation (4), where t(C b ) indicates vehicle C b The total time spent within the RSU communication coverage area, if satisfying t(C b If )>τ1, determine vehicle C b It can be used for communication services;

[0019] In equation (3), C r Let |C represent the set of vehicles that can be controlled and served within a specified time period. r | represents the total number of vehicles that can be controlled and served during this period, and f3 represents the coverage rate of controlled and served vehicles during this period. Its constraints are as shown in equations (5) and (6), where t(C d ) indicates vehicle C d The total time spent on road segment e(i,j) during the period of connecting to the RSU and successfully communicating, if t(C b )+t(C d If τ < τ, determine vehicle C. d The service can be controlled.

[0020] Furthermore, the optimization of the solution method specifically involves: using the quantum behavior particle swarm optimization algorithm, eliminating the velocity update in the PSO algorithm, and only performing displacement updates, with the update equation being...

[0021]

[0022]

[0023]

[0024] Where i (i = 1, 2, ..., P) represents the i-th particle, P is the population size; j (j = 1, 2, ..., Q) represents the j-th dimension of the particle, Q is the dimension of the search space; k is the generation number; u i,j (t) and All are random numbers uniformly distributed in the interval [0,1]; X i,j (k), p i,j (k) and pbest i,j (k) represents the current position, attractor position, and best individual position of the i-th particle in the j-th dimension at the k-th generation; gbest j (k) represents the global best position of the j-th dimension particle swarm in the k-th generation; mbest j (k) represents the average best position of a particle in the kth generation and jth dimension, defined as the average of the best positions of all individual particles. α is called the expansion-contraction factor; when α < 1.782, the algorithm is guaranteed to converge.

[0025] To ensure the diversity of the solution set, the MOQPSO algorithm based on crowding distance sorting is adopted. An external archive is set to store the non-dominated solutions found during the search process in order to quickly approach the Pareto optimal front. The crowding distance sorting method in the second-generation non-dominated sorting genetic algorithm is used to maintain the diversity of the solution set of the multi-objective quantum behavior particle swarm optimization algorithm.

[0026] Furthermore, the MOQPSO algorithm based on crowding distance sorting specifically includes the following steps:

[0027] S1: Set the basic parameters of the algorithm, initialize all particles in the particle swarm within the search space, and initialize each particle with its initial pbest value. i Defined as the initial position of the particle;

[0028] S2: Evaluate all particles in the particle swarm and, based on the Pareto dominance, transfer the non-dominated particles into the external archive.

[0029] S3: Select gbest for each particle in the particle swarm, update the position of the particle according to equations (7) to (9), and perform mutation using a random mutation method with Gaussian distribution characteristics;

[0030] S4: Re-evaluate all particles in the swarm and update the individual best position pbest. If the updated particle dominates pbest, then the updated particle is set as the new pbest; if the two do not dominate each other, then one of them is randomly selected as the new pbest.

[0031] S5: Update the particles in the external archive using the crowding distance sorting method;

[0032] S6: Determine if the current generation has reached the maximum number of generations. If it has, proceed to step 8; otherwise, proceed to step 3.

[0033] S7: Output all particles in the external archive Archive as the final solution.

[0034] This vehicle-road cooperative roadside unit deployment system for human-machine co-driving utilizes the aforementioned human-machine co-driving vehicle-road cooperative roadside unit deployment method to communicate and control roadside units. The system includes an information access module, a roadside unit dynamic planning module, a roadside unit edge computing module, and a switching control information publishing module. The vehicle-road information access module is responsible for exchanging vehicle information. The roadside unit dynamic planning module dynamically accesses and releases roadside units to form a roadside unit grid of a certain size. The roadside unit edge computing module collects device status information from all roadside units within a certain range, providing computing power to execute the grid-based control switching for human-machine co-driving. The switching control information publishing module publishes control commands sent to the human-machine co-driving central control system.

[0035] Specifically, the input terminals of the information access module, roadside unit dynamic planning module, roadside unit edge computing module, and switching control information release module are all connected to the human-machine co-driving center control system, and the output terminals of the information access module, roadside unit dynamic planning module, roadside unit edge computing module, and switching control information release module are all communicatively connected to the roadside unit.

[0036] This invention presents a vehicle-road cooperative roadside unit (RSU) deployment system and method for human-machine co-driving, overcoming the shortcomings of existing RSU deployment schemes in achieving precise control of vehicles and roads. Starting from collaborative information interaction, collaborative perception, and collaborative decision-making control in vehicle-road cooperative autonomous driving, it optimizes the system structure and topology of the deployment of RSUs with small spacing and high density in road networks, establishing a better RSU deployment topology model to minimize the number of RSUs deployed and maximize the coverage of communication and control service vehicles. By building a deployment system, establishing a multi-objective optimization model, and proposing an improved multi-objective quantum behavior particle swarm optimization algorithm, it solves the problem that traditional RSUs cannot adapt to precise control of vehicles and roads. This achieves high-density, high-precision, and low-cost information communication and control requirements for human-machine co-driving, minimizing the number of RSUs while maximizing the coverage of communication and control service vehicles. Attached Figure Description

[0037] The following description, in conjunction with the accompanying drawings, further illustrates a vehicle-road cooperative roadside unit deployment system and method for human-machine co-driving according to the present invention:

[0038] Figure 1 This is a road network model diagram of the optimized road network model described in the method for deploying roadside units in vehicle-road cooperative driving for human-machine co-driving.

[0039] Figure 2 It is an undirected graph representation of the road network model described in the method for deploying roadside units in vehicle-road cooperative driving for human-machine co-driving;

[0040] Figure 3 This is a schematic diagram of the vehicle driving path optimized by the time threshold in the vehicle-road cooperative roadside unit deployment system for human-machine co-driving.

[0041] Figure 4 This is a logical structure diagram of a vehicle-road cooperative roadside unit deployment system for human-machine co-driving. Detailed Implementation

[0042] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0043] In the description of this invention, it should be understood that the terms "left", "right", "front", "rear", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0044] The technical solution of the present invention will be further described below with specific embodiments, but the scope of protection of the present invention is not limited to the following embodiments.

[0045] Implementation method: The vehicle-road cooperative roadside unit deployment system and method for human-machine co-driving includes road network model optimization, time threshold optimization, objective function optimization, and solution method optimization.

[0046] like Figure 1 , 2 As shown, the road network model optimization specifically involves: prioritizing road intersections as candidate RSU placement locations; before placing roadside units, the road network is abstracted into a mathematical model using a planar undirected graph, denoted as G = (V, E); where G consists of a set of vertices and the set of edges between them, and V is the set of candidate RSU placement locations, V = {V1, V2, L, V...} x}, and satisfy V=x; E is the set of road segments connecting the candidate RSU deployment locations, E={E1,E2,L,E y The set of road segment lengths is D = {D1, D2, L, D}. y}. For D k ∈D and D k >2R section E k Further segmentation is performed, and within each segment, equal intervals of length L are inserted. A new node is selected as a candidate location for RSU deployment to maximize road network coverage and achieve precise control of vehicles and roads. After the partitioning is complete, a final undirected graph G′=(V′,E′) is formed, where V′={V,N}, and N is the set of nodes formed during the partitioning of E∈G=(V,E), N={N1,N2,L,N}. z And |V′|=m, |E′|=p; And e(i,j)={l ij ,ρ ij ,v ij} Figure 1 The road network shown at x=12, after being abstracted through mathematical modeling, yields the following result: Figure 2 The undirected graph shown.

[0047] In this embodiment, all RSUs have circular communication coverage areas with the same communication radius, R. RSUs can communicate with vehicles or other RSUs within their coverage area. However, at any given time, communication between a vehicle and an RSU is not limited to the coverage area of ​​only one RSU. Typically, the communication radius of an RSU is much larger than the road width, so the road width is negligible. This embodiment does not consider communication between vehicles, and for simplicity, it does not consider the uplink network conditions for communication between RSUs and vehicles. To achieve greater spatial coverage, RSUs are usually deployed in locations with significant spatial characteristics, such as road intersections, which not only offer greater data distribution advantages but also increase communication coverage by approximately 15%. Therefore, this embodiment prioritizes road intersections as candidate RSU deployment locations.

[0048] like Figure 3 As shown, the time threshold optimization specifically involves: the set of vehicles in the road network is represented by C, where C = {C1, C2, L, C...} n Let there be an m×n matrix T = (t ab ), its element t ab (1≤a≤m,1≤b≤n,t ab ≥0) indicates that during a certain time period, vehicle C b In V′ a The time a vehicle stays within the communication coverage area of ​​the deployed RSU; assuming the minimum time required for a vehicle to connect to the RSU and successfully communicate is τ1, communication between the RSU and the vehicle can be completed through multiple RSUs. If, during this time period, vehicle C... b If the total time spent within the communication coverage area of ​​each RSU deployment node is greater than τ1, then vehicle C is determined to be... b Communication services are available during this period. The control commands transmitted by the RSU are time-sensitive and are represented by an m×m matrix. in This indicates that during a certain period of time, vehicle C d (C d ∈C b The time taken for vehicle C to travel on road segment e(i,j); let τ be the maximum time required for a vehicle to connect to the RSU and successfully receive a certain control command, and the relationship between τ and τ1 is: τ = τ1 + τ2, where τ2 is the total time the vehicle stays on road segment e(i,j) during the period of connecting to the RSU and successfully communicating. If, during a certain period, vehicle C... b If the time to successfully receive a control command is less than τ, then vehicle C is determined to be... b The service can be controlled during this period.

[0049] For example, during a certain period of time, vehicle C b If the time to successfully receive a control command is less than τ, then vehicle C is considered to be...b Service can be controlled during this period. According to matrix T e Construction as Figure 3 The traffic scenario shown has four road intersections with a spacing greater than 2R, therefore a total of eight Road Units (RSUs) are deployed. During a specified time period, vehicles C1, C3, and C4 can receive communication services.

[0050]

[0051] The objective function optimization is as follows: Since the number of RSUs deployed depends on the spacing L when creating new nodes between road segments at road intersections, a length L that is as long as possible is set to achieve a larger coverage area for the RSUs. The multi-objective optimization function is constructed as follows:

[0052]

[0053]

[0054]

[0055]

[0056]

[0057] t(C b )+t(C d )<τ,C b ∈C r C d ∈C r (6)

[0058] In equation (1), f1 represents the reciprocal of the number of RSUs deployed in the road network. The larger the value of f1, the fewer RSUs need to be deployed in the road network.

[0059] In equation (2), C s Let |C represent the set of vehicles that can be served by communication within a specified time period. s | represents the total number of vehicles that can be served by communication during this period, and f2 represents the coverage rate of vehicles that can be served by communication during this period. The constraints are as shown in equation (4), where t(C b ) indicates vehicle C b The total time spent within the RSU communication coverage area, if satisfying t(C b If )>τ1, determine vehicle C b It can be used for communication services;

[0060] In equation (3), C r Let |C represent the set of vehicles that can be controlled and served within a specified time period. r| represents the total number of vehicles that can be controlled and served during this period, and f3 represents the coverage rate of controlled and served vehicles during this period. Its constraints are as shown in equations (5) and (6), where t(C d ) indicates vehicle C d The total time spent on road segment e(i,j) during the period of connecting to the RSU and successfully communicating, if t(C b )+t(C d If τ < τ, determine vehicle C. d The service can be controlled.

[0061] The optimization of the solution method specifically involves: using the quantum behavior particle swarm optimization algorithm, eliminating the velocity update in the PSO algorithm, and only performing displacement updates. The update equation is as follows:

[0062]

[0063]

[0064]

[0065] Where i (i = 1, 2, ..., P) represents the i-th particle, P is the population size; j (j = 1, 2, ..., Q) represents the j-th dimension of the particle, Q is the dimension of the search space; k is the generation number; u i,j (t) and All are random numbers uniformly distributed in the interval [0,1]; X i,j (k), p i,j (k) and pbest i,j (k) represents the current position, attractor position, and best individual position of the i-th particle in the j-th dimension at the k-th generation; gbest j (k) represents the global best position of the j-th dimension particle swarm in the k-th generation; mbest j (k) represents the average best position of a particle in the kth generation and jth dimension, defined as the average of the best positions of all individual particles. α is called the expansion-contraction factor; when α < 1.782, the algorithm is guaranteed to converge.

[0066] To ensure the diversity of the solution set, the MOQPSO algorithm based on crowding distance sorting is adopted. An external archive is set to store the non-dominated solutions found during the search process in order to quickly approach the Pareto optimal front. The crowding distance sorting method in the second-generation non-dominated sorting genetic algorithm is used to maintain the diversity of the solution set of the multi-objective quantum behavior particle swarm optimization algorithm.

[0067] The MOQPSO algorithm based on crowding distance sorting specifically includes the following steps:

[0068] S1: Set the basic parameters of the algorithm, initialize all particles in the particle swarm within the search space, and initialize each particle with its initial pbest value. i Defined as the initial position of the particle;

[0069] S2: Evaluate all particles in the particle swarm and, based on the Pareto dominance, transfer the non-dominated particles into the external archive.

[0070] S3: Select gbest for each particle in the particle swarm, update the position of the particle according to equations (7) to (9), and perform mutation using a random mutation method with Gaussian distribution characteristics;

[0071] S4: Re-evaluate all particles in the swarm and update the individual best position pbest. If the updated particle dominates pbest, then the updated particle is set as the new pbest; if the two do not dominate each other, then one of them is randomly selected as the new pbest.

[0072] S5: Update the particles in the external archive using the crowding distance sorting method;

[0073] S6: Determine if the current generation has reached the maximum number of generations. If it has, proceed to step 8; otherwise, proceed to step 3.

[0074] S7: Output all particles in the external archive Archive as the final solution.

[0075] Example 1: As Figure 4 As shown, this vehicle-road cooperative roadside unit deployment system for human-machine co-driving utilizes the aforementioned human-machine co-driving vehicle-road cooperative roadside unit deployment method to communicate and control roadside units. The system includes an information access module, a roadside unit dynamic planning module, a roadside unit edge computing module, and a switching control information publishing module. The vehicle-road information access module is responsible for exchanging vehicle information; the roadside unit dynamic planning module dynamically accesses and releases roadside units to form a roadside unit grid of a certain size; the roadside unit edge computing module collects the device status information of all roadside units within a certain range, providing computing power to execute the grid-based control switching of human-machine co-driving; and the switching control information publishing module publishes control commands sent to the human-machine co-driving central control system.

[0076] Example 2: In this vehicle-road cooperative roadside unit deployment system for human-machine co-driving, the input terminals of the information access module, roadside unit dynamic planning module, roadside unit edge computing module, and switching control information publishing module are all connected to the human-machine co-driving central control system. The output terminals of these modules are all communicatively connected to the roadside units. The remaining structures and components are as described in Example 1 and will not be repeated.

[0077] This vehicle-road cooperative roadside unit (RSU) deployment system and method for human-machine co-driving overcomes the shortcomings of existing RSU deployment schemes in achieving precise control of vehicles and roads. Starting with collaborative information interaction, perception, and decision-making control in vehicle-road cooperative autonomous driving, it optimizes the system structure and topology of the deployment of RSUs with small spacing and high density in road networks. A better RSU deployment topology model is established to minimize the number of RSUs deployed and maximize the coverage of communication and control service vehicles. By building a deployment system, establishing a multi-objective optimization model, and proposing an improved multi-objective quantum behavior particle swarm optimization algorithm, it solves the problem that traditional RSUs cannot adapt to precise control of vehicles and roads. This achieves high-density, high-precision, and low-cost information communication and control requirements for human-machine co-driving, minimizing the number of RSUs while maximizing the coverage of communication and control service vehicles.

[0078] The foregoing description illustrates the main features, basic principles, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments or examples described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the above embodiments or examples should be considered exemplary and not restrictive. The scope of the present invention is defined by the appended claims rather than the foregoing description, and therefore all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0079] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for deploying vehicle-road cooperative roadside units for human-machine co-driving, characterized by: This includes road network model optimization, time threshold optimization, objective function optimization, and solution method optimization; Specifically, the time threshold optimization is as follows: Collection of vehicles in the road network Let's represent this as a matrix. Its elements Indicates vehicles within a certain time period exist The time a vehicle spends within the communication coverage area of ​​the RSU deployed at the location; assuming the minimum time required for the vehicle to connect to the RSU and successfully communicate is... Communication between the RSU and the vehicle is completed through multiple RSUs. If the vehicle is within this time period... The total time spent within the communication coverage area of ​​each RSU deployment node is greater than Determine the vehicle Communication services are available during this period; The control commands transmitted by the RSU are time-sensitive and are configured with a matrix. ,in Indicates the number of vehicles on the road section within a certain time period. Time taken for passage; assuming the maximum time required for a vehicle to connect to the RSU and successfully receive a certain control command is... , and The relationship is: ,in Stopped on the road while the vehicle connected to the RSU and successfully communicated. The total time, if within a certain period of time, the vehicle The time to successfully receive a certain control command is less than Determine the vehicle Services can be controlled during this period; The objective function optimization aims to construct a multi-objective optimization function with the objectives of minimizing the number of roadside units deployed, maximizing the coverage of communication service vehicles, and maximizing the coverage of control service vehicles. The coverage of communication service vehicles is the ratio of the number of vehicles that can be served by communication within a specified time period to the total number of vehicles within that time period, and the coverage of control service vehicles is the ratio of the number of vehicles that can be controlled within a specified time period to the total number of vehicles within that time period. The optimization of the solution method is used to solve the multi-objective optimization function to obtain the roadside unit layout scheme; The road network model optimization specifically involves... Priority road intersections are candidate locations for RSU placement. Before placing roadside units, the road network is abstracted into a mathematical model using a planar undirected graph, denoted as . ;in, It consists of a set of vertices and the set of edges between them. Set up the candidate deployment locations for RSUs. And satisfy ; This is the set of road segments connecting the candidate RSU deployment locations. The set of road segment lengths is ; for and section of road Further division is made within the road segment by length. Insert at equal intervals in units A new node is selected as a candidate deployment location for RSUs to maximize coverage of the road network and achieve precise control of vehicles and roads. After the partitioning is completed, the final undirected graph is formed. , , yes The set of nodes formed during the partitioning process and , ; ,and ; The optimization of the solution method specifically involves using a quantum behavior particle swarm optimization algorithm, eliminating velocity updates in the PSO algorithm, and performing only displacement updates. The update equation is as follows: （7） （8） （9） in, Indicates the first One particle, For group size; The first particle represents the second particle. dimension, The dimension of the search space; For evolutionary algebra; and All A random number uniformly distributed over an interval; , and They represent the first time. Generation, First The first particle The current position of the dimension, the position of the attractor, and the best position of the individual; Indicates the first Generation, First The global best position of the 3D particle swarm; Indicates the particle at the 1st Generation, First The average best position is defined as the average of the best positions of all individual particles. The Called the expansion-contraction factor, when When the algorithm converges, it can be guaranteed to achieve convergence. To ensure the diversity of the solution set, the MOQPSO algorithm based on crowding distance sorting is adopted. An external archive is set to store the non-dominated solutions found during the search process in order to quickly approach the Pareto optimal front. The crowding distance sorting method in the second-generation non-dominated sorting genetic algorithm is adopted to maintain the diversity of the solution set of the multi-objective quantum behavior particle swarm optimization algorithm. The MOQPSO algorithm based on crowding distance sorting specifically includes the following steps: S1: Set the basic parameters of the algorithm, initialize all particles in the particle swarm within the search space, and initialize each particle's initial... Defined as the initial position of the particle; S2: Evaluate all particles in the particle swarm and, based on the Pareto dominance, transfer the non-dominated particles into the external archive. S3: Select each particle in the particle swarm. The positions of the particles are updated according to equations (7) to (9), and the mutation is performed using a random mutation method with Gaussian distribution characteristics. S4: Re-evaluate all particles in the swarm and update the individual best positions of the particles. If the updated particle dominates Then the updated particle will be used as the new... If neither can dominate the other, then one of them is randomly selected as the new [selector / selector]. ; S5: Update the particles in the external archive using the crowding distance sorting method; S6: Determine if the current generation has reached the maximum generation; if so, proceed to step 8; otherwise, proceed to step 3. S7: Output all particles in the external archive Archive as the final solution.

2. A vehicle-road cooperative roadside unit deployment system for human-machine co-driving, characterized by: The vehicle-road cooperative roadside unit deployment system, based on the roadside unit communication control of the human-machine co-driving vehicle-road cooperative roadside unit deployment method of claim 1, includes an information access module, a roadside unit dynamic planning module, a roadside unit edge computing module, and a switching control information publishing module. The information access module is used to handle the interaction of vehicle information; The roadside unit dynamic planning module is used to dynamically access and release roadside units to form a roadside unit grid of a certain size. The roadside unit edge computing module is used to collect the device status information of all roadside units within a certain range and provide computing power to perform grid-based control switching for human-machine co-driving. The switching control information publishing module is used to publish control commands sent to the human-machine co-driving center control system.

3. The vehicle-road cooperative roadside unit deployment system for human-machine co-driving as described in claim 2, characterized in that: The input terminals of the information access module, roadside unit dynamic planning module, roadside unit edge computing module, and switching control information release module are all connected to the human-machine co-driving center control system, and the output terminals of the information access module, roadside unit dynamic planning module, roadside unit edge computing module, and switching control information release module are all connected to the roadside unit for communication.

Images