A multi-vehicle dynamic routing method based on single and multi-intersections

Abstract

The present invention discloses a multi-vehicle dynamic routing method based on single and multiple intersections. The method comprises: a first roadside unit obtains first status information, the first status information including the vehicle classification category and the vehicle's destination; based on the first status information, the first roadside unit processes vehicles arriving at the intersection provided with the first roadside unit within a first time period through a constructed game model; the Nash equilibrium state of the game model is obtained according to the evolutionary game model, and the optimal proportion of vehicles for each strategy in each vehicle group is obtained, as well as the number of vehicles for each strategy in each vehicle group; and the travel direction of each vehicle is determined based on the number of vehicles and the time when the vehicle arrives at the intersection. By classifying different vehicle types and assigning corresponding weights, the present invention achieves the goal of both taking into account vehicle priority needs and balancing the road network.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a method for dynamically selecting paths for multiple vehicles based on single and multiple intersections. Background Art

[0002] As the number of vehicles on the road network continues to grow, traffic congestion is becoming increasingly severe. Expanding roads and restricting access to vehicles remain inadequate solutions. In a collaborative vehicle-road environment, vehicles and road infrastructure can share real-time traffic status information through vehicle-road communication technologies. This enables the transportation system to become interconnected, personalized, and dynamic. While effectively increasing the intelligence of road traffic control, it also significantly increases the complexity of traffic control.

[0003] To address the problem of dynamic multi-vehicle collaborative routing, one approach proposed in existing technologies is a decentralized strategy. This divides a large road network into smaller zones, where participating vehicles perform routing planning within each zone. However, due to the coupled motion trajectories of vehicles within different zones, this decentralized approach fails to maximize the positive impact of vehicle-based collaborative decision-making technology on routing optimization, increasing the uncertainty of traffic control performance. Another approach is a purely individualized strategy, which utilizes predictive models to forecast future traffic flows and dynamically optimize individual vehicle routes. This optimization approach relies heavily on the vehicle's own computing power and background data, making it challenging to scale and failing to fully realize the benefits of connected vehicles. Therefore, research on multi-vehicle collaborative routing requires further refinement to better meet the travel needs of diverse vehicles while maintaining a balanced road network and enhancing the model's universality.

[0004] In order to dynamically and properly consider the different traffic needs of individuals and the management performance of the traffic system, it is urgent to propose a new vehicle group collaborative routing decision-making algorithm to achieve a good balance between individual needs and road network coordination performance. Summary of the Invention

[0005] The present invention is proposed based on the above-mentioned requirements of the prior art. The technical problem to be solved by the present invention is to provide a multi-vehicle dynamic path selection method based on single intersections and multiple intersections to achieve a good balance between individual needs and road network coordination performance.

[0006] In order to solve the above problems, the present invention is implemented by adopting the following technical solutions:

[0007] A method for dynamic route selection of multiple vehicles based on a single intersection, the method comprising: a first roadside unit obtains first state information, the first state information including the classification category of the vehicle and the destination of the vehicle; the first roadside unit processes the vehicles arriving at the intersection with the first roadside unit within a first time period through a constructed game model based on the first state information, including: setting a temporary destination according to the destination of the vehicle; grouping the vehicles based on the road section where the vehicle is located and the temporary destination, and obtaining multiple vehicle groups; establishing a benefit relationship of the strategy based on the preset time consumed by the vehicle at the intersection and the travel time to the temporary destination; and The average benefit relationship of the vehicle group is established based on the vehicle proportion and the benefit relationship of the corresponding strategy, wherein the strategy includes the driving direction selected by the first roadside unit for the vehicle; when the classification category in the vehicle group is the first category, the proportion of vehicles adopting the corresponding strategy is increased based on the first weight, and when the classification category in the vehicle group is the second category, the proportion of vehicles adopting the corresponding strategy is reduced based on the second weight; the Nash equilibrium state of the game model is obtained according to the evolutionary game model, the optimal proportion of vehicles for each strategy in each vehicle group is obtained, and the number of vehicles for each strategy in each vehicle group is obtained; the driving direction of each vehicle is determined based on the number of vehicles and the time when the vehicle arrives at the intersection.

[0008] Optionally, the multi-vehicle dynamic path selection method based on a single intersection includes: the second roadside unit classifies the vehicle based on the vehicle request, wherein the classification categories include the first category and the second category; the second roadside unit obtains the first state information and the second state information of the vehicle, and the second state information includes the motion parameters of the vehicle during driving; the first roadside unit obtains the first state information, including: the second roadside unit determines whether the vehicle arrives at the intersection where the first roadside unit is provided within a first time period based on the second state information of the vehicle; if it can arrive, the second roadside unit sends the first state information of the vehicle to the first roadside unit.

[0009] Optionally, the number of vehicle groups is equal to the number of temporary destinations multiplied by the number of road sections where the vehicles are located.

[0010] Optionally, the method for calculating the driving time to a temporary destination includes: obtaining the road section that the vehicle passes through while driving towards the temporary destination based on the direction provided by the first roadside unit; wherein the passed road section is represented as 1, and the not passed road section is represented as 0; based on the shortest time for the vehicle to pass through the road section, the number of vehicles in the road section in the area controlled by the first roadside unit, the load capacity of the road section, and the number of vehicles in each vehicle group in the road section, obtaining the time delay of passing the road section; and determining the driving time to the temporary destination based on the passed road section and the corresponding time delay.

[0011] Optionally, the time delay for passing the road section is obtained based on the shortest time for a vehicle to pass the road section, the number of vehicles in the road section in the first roadside unit control area, the load capacity of the road section, and the number of vehicles in each vehicle group in the road section. The expression includes: in, Indicates road section The delay, is the road section from the i-th intersection to the j-th intersection, represents the shortest time for a vehicle to pass through the road section, α represents the first constant, represents the total number of vehicles moving from the i-th intersection to the j-th intersection in the shortest time period, Indicates road section The load capacity, β represents the second constant, p represents the vehicle group, S represents the total number of vehicle groups, a k，p represents the kth strategy selected, A p Represents a set of policies, Indicates that the road In the above example, the strategy a is selected in the vehicle group p. k，p The proportion of vehicles in p | indicates that vehicle group p is currently traveling on the road segment The number of vehicles, Indicates that the vehicles in group p follow strategy a k，p Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0. Indicates road section The current number of vehicles traveling.

[0012] Optionally, based on the preset time taken by the vehicle at the intersection and the travel time to the temporary destination, a benefit relationship of the strategy is constructed, and its expression includes: k,p =-(d crass,k,p +d link,k,p ), Among them, c k,p Indicates the vehicle execution strategy a in vehicle group p k,p The income, d crass，k，p Indicates the vehicle execution strategy a in vehicle group p k，p The preset time at the intersection, d link，k，p Indicates that the vehicles in vehicle group p execute strategy a at the intersection k,p Time of arrival at the temporary destination, represents the road section from the i-th intersection to the j-th intersection, L represents the number of road sections within the control area of ​​the first drive test unit, Indicates that the vehicles in group p follow strategy a k，p Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0, Indicates road section delay.

[0013] Optionally, based on the proportion of vehicles in each strategy in each vehicle group and the benefits of the corresponding strategy, an average benefit relationship in the vehicle group is constructed, and its expression includes: in, represents the average revenue of vehicle group p, a k，p represents the kth strategy selected, A p Represents a set of policies, Indicates that the road In the above example, the strategy a is selected in the vehicle group p. k，p The proportion of vehicles, c k，p Indicates the vehicle execution strategy a in vehicle group p k，p of income.

[0014] Optionally, the Nash equilibrium state of the game model is obtained according to the evolutionary game model to obtain the optimal proportion of vehicles for each strategy in each vehicle group, including: using the evolutionary game model to process the constructed game model to obtain an ordinary differential equation; using the fourth-order Runge-Kutta method to solve the ordinary differential equation to obtain the optimal proportion of vehicles for each strategy in each vehicle group.

[0015] A multi-vehicle dynamic routing method based on multiple intersections, the method comprising: obtaining vehicle information belonging to the third category at each intersection, wherein the vehicles belonging to the third category represent vehicles that have not reached a temporary destination and are traveling according to a strategy output by a first roadside unit at an upstream intersection; based on the vehicle information belonging to the third category at each intersection, processing the remaining vehicles that arrive at the corresponding intersection within a first time period at the intersection using the multi-vehicle dynamic routing method based on a single intersection is performed to obtain the driving direction of each vehicle controlled by the first roadside unit at each intersection.

[0016] Optionally, it includes: on non-intersecting road sections, vehicles travel to the temporary destination according to the shortest distance route; if there are multiple shortest distances, the vehicles in the vehicle group are evenly divided into multiple groups, and each group corresponds to the shortest distance route one by one.

[0017] Compared with the existing technology, the present invention proposes a multi-vehicle dynamic path selection method based on single and multi-intersections, which transforms the problem of collaborative routing of vehicle groups in macro-scale large-scale road network scenarios into a collaborative decision-making problem at micro-intersections, effectively reducing the difficulty of modeling and solving the problem of collaborative routing decision-making of vehicle groups in large-scale scenarios. By dividing different vehicle types and assigning corresponding weights, the goal of taking into account the vehicle priority needs and balancing the road network is achieved, and the universality of the model is also improved. The interests of individual vehicles and the collaborative operation of multiple vehicles are guaranteed, so that the average vehicle travel time, road network throughput and road load balancing indicators all have good performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 This is a flow chart of a method for dynamic route selection for multiple vehicles at a single intersection provided by this embodiment;

[0020] Figure 2 This is a schematic diagram of a scenario of collaborative routing decision-making for a group of vehicles in a vehicle-road collaborative environment provided by this embodiment;

[0021] Figure 3 This is a schematic diagram of a conflict in route planning for vehicles at a front and rear intersection provided by this embodiment;

[0022] Figure 4A Schematic diagram of the topology of a traffic network for multi-vehicle collaborative simulation at a single intersection provided by this embodiment;

[0023] Figure 4B Schematic diagram of the topology of the multi-intersection multi-vehicle collaborative simulation traffic network provided by this embodiment;

[0024] Figure 5A is a schematic diagram of the road network throughput under different traffic flows provided in Example 1;

[0025] Figure 5B This is a schematic diagram of average vehicle travel time under different traffic flows provided in Example 1;

[0026] Figure 6A The starting points O9 and O 10 Schematic diagram of average vehicle travel time under different vehicle priority ratios at a vehicle inflow rate of 60 veh / min;

[0027] Figure 6B The starting points O9 and O 10 Schematic diagram of average vehicle travel time under different vehicle priority ratios at a vehicle inflow rate of 80 veh / min;

[0028] Figure 7A 1 is a schematic diagram of vehicle density distribution at each moment when vehicles are traveling in the forward direction based on the method provided in Example 1;

[0029] Figure 7B 1 is a schematic diagram of vehicle density distribution at each moment when vehicles are traveling in the reverse direction based on the method provided in Example 1;

[0030] Figure 7CIt is the vehicle density distribution map of vehicles traveling forward at each moment based on the social routing algorithm;

[0031] Figure 7D It is the vehicle density distribution map of vehicles traveling in reverse at each moment based on the social routing algorithm;

[0032] Figure 7E This is a schematic diagram of vehicle density distribution at each moment when vehicles are traveling forward based on the shortest path algorithm;

[0033] Figure 7F This is a schematic diagram of vehicle density distribution at each moment when vehicles are traveling in the opposite direction based on the shortest path algorithm;

[0034] Figure 8A is a schematic diagram of road network throughput under different traffic flows provided in Example 2;

[0035] Figure 8B This is a schematic diagram of average vehicle travel time under different traffic flows provided in Example 2;

[0036] Figure 9A The starting points O9 and O 10 Schematic diagram of average vehicle travel time under different vehicle priority ratios at a vehicle inflow rate of 60 veh / min;

[0037] Figure 9B The starting points O9 and O 10 Schematic diagram of average vehicle travel time under different vehicle priority ratios at a vehicle inflow rate of 80veh / min. DETAILED DESCRIPTION

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0039] To facilitate understanding of the embodiments of the present invention, the following will be further explained with reference to specific embodiments in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the scope of protection of the present invention.

[0040] Example 1

[0041] This embodiment provides a method for dynamic routing of multiple vehicles based on a single intersection. Figure 1 As shown, including:

[0042] S1: A first roadside unit obtains first status information, where the first status information includes a classification category of a vehicle and a destination of the vehicle.

[0043] The first roadside unit includes an information collection unit, a game roadside unit and a data distribution unit, which are used to coordinate the vehicle routing planning within its control range. Figure 2 As shown in the figure, the areas I1, I2, ..., I9 represent the intersection conflict area, and the vehicle makes a specific path selection in this area. The control range of the first roadside unit includes the set intersection, surrounding intersections and the intersection sections between intersections, such as Figure 2 As shown, the control range is 9 intersections and 12 directed roads within the larger dotted box.

[0044] This step includes:

[0045] S100 The second roadside unit classifies the vehicle based on the vehicle request to obtain a first category and a second category.

[0046] The second roadside unit includes an information collection unit and a data distribution unit, which is used to directly exchange data with vehicles and other second roadside units.

[0047] In this embodiment, the first category includes priority vehicles, and the second category includes regular vehicles. Vehicles in the first category have urgent traffic needs and prefer to travel on cleared roads; vehicles in the second category are less urgent. As each vehicle travels, the assisted driving system registers an agent for the vehicle on the second roadside unit (RSU) on the current road segment. The second RSU classifies the vehicle based on the vehicle's request.

[0048] S110: The second roadside unit obtains first state information and second state information of the vehicle, where the second state information includes motion parameters of the vehicle during driving.

[0049] The motion parameters of a vehicle during driving include the vehicle's position, velocity, and acceleration.

[0050] S120 The second roadside unit determines whether the vehicle arrives at the intersection where the first roadside unit is located within a first time period based on the second state information of the vehicle.

[0051] The second RSU monitors in real time whether a registered agent vehicle is approaching the intersection where the first RSU is located. Based on the second status information, the second RSU determines whether the vehicle can reach the intersection within a first time period. If the vehicle can reach the intersection, the second RSU transmits the first status information to the first RSU and registers an agent for the vehicle. Otherwise, the second RSU continues to monitor the vehicle.

[0052] The first roadside unit plays a game for agents corresponding to all vehicles arriving within a first time period.

[0053] S2: The first roadside unit processes vehicles arriving at the intersection with the first roadside unit within a first time period through a constructed game model based on the first state information.

[0054] With intersections as points, road sections as edges, and the actual distances between road sections as edge weights, a directed weighted graph is constructed to represent the road model. Specifically, the expression of the directed weighted graph is H = (I, L), I i ∈I, Among them, I represents the point set of the graph, represents the set of intersections, and I i The point with the subscript i in the graph represents the i-th intersection in the intersection set, L represents the edge set of the graph, and represents the set of roads. Indicates that from intersection I i To intersection I j section of road.

[0055] In this embodiment, considering that there are many vehicles participating in route planning at the intersection and different vehicles have different urgency for passage, vehicles with high urgency for passage are given priority to a certain extent, and vehicles with relatively low urgency for passage need to give way so that they can travel on sections with relatively less congestion. However, at the same time, the traffic efficiency of all participating vehicles must also be considered. The game process is defined as G = (S, A).

[0056] Here, S represents a vehicle group consisting of multiple vehicle groups. A vehicle group p∈S contains multiple game players. All vehicles that must pass through the intersection with the first roadside unit within the first time period are game players. A represents the strategy set of all game players, that is, the possible actions that vehicles can take at the intersection, such as turning left, going straight, making a U-turn, and turning right.

[0057] For a vehicle group p containing multiple game individuals, its strategy set is A p =(a 1，p , a 2，p ,…,a k，p ,…,a m，p ), m is the number of optional actions at the intersection, a k，p Represents the strategy, that is, the optional actions of the vehicles at the intersection. The vehicle set is V p In this embodiment, the set V p The vehicles include the first type Car sperial and the second type of Car general There are two categories, each vehicle chooses a certain strategy. At this time, the state of vehicle group p is X p Represents a set of vehicles with different strategies. When the classification category in the vehicle group is the first type Carspecial When the first weight ρ is used to increase the proportion of vehicles that adopt the corresponding strategy, when the classification category in the vehicle group is the second type Car general When the proportion of vehicles adopting the corresponding strategy is reduced based on the second weight θ, the state of the vehicle group p is in Indicates that in vehicle group p, strategy a is selected k，p ∈A p The vehicle group state composed of all vehicle group states is Δ(X)={X p :p∈S}.

[0058] The goal of multi-vehicle coordination is to find a reasonable and effective diversion method, dispatching a reasonable number of vehicles to different roads so that no road is overused or idle, thus satisfying the priority needs of vehicles and maximizing the benefits of the game participants. Therefore, for the game model G = (S, A), the optimization goal is to find a vehicle group state Δ(X) such that no individual player can find a strategy that is better than their current strategy. To achieve this optimization goal, this embodiment uses the time required for vehicles to reach the temporary destination to evaluate the benefits of each strategy.

[0059] This step specifically includes:

[0060] S200 sets a temporary destination according to the vehicle's destination.

[0061] The vehicles participating in the game are grouped as follows: Figure 2 As shown, the intersections around the intersection where the first roadside unit is gaming are used as temporary destinations for the gaming vehicles. For example, the temporary destination of I1 can be selected from I2, ..., I9, which is specifically determined by the orientation of each vehicle's final destination relative to the current intersection.

[0062] S210 groups the vehicles based on the road sections and temporary destinations of the vehicles, and obtains a plurality of vehicle groups.

[0063] The first roadside unit groups vehicles according to their temporary destinations and the road section they are on, and records the number of agents in each group. Vehicles in the same group share the same temporary destination and are currently traveling on the same road section. For simplicity, this embodiment refers to the road section a vehicle is currently traveling on as its initial point. The number of vehicle groups in this model is equal to the number of temporary destinations multiplied by the number of initial points.

[0064] S220 establishes a benefit relationship of the strategy based on the time the vehicle spends at the intersection and the travel time to the temporary destination.

[0065] The strategy includes a driving direction selected by the first roadside unit for the vehicle.

[0066] In this step, the method for calculating the travel time to the temporary destination includes:

[0067] Obtain the road sections that the vehicle passes through while traveling to the temporary destination based on the direction provided by the first roadside unit; the passed road sections are represented by 1, and the unpassed road sections are represented by 0. Indicates strategy a k，p Whether it passes through the road section If it passes, it is 1, otherwise it is 0.

[0068] The time delay of passing the road section is obtained based on the shortest time for vehicles to pass through the road section, the number of vehicles in the road section in the first roadside unit control area, the load capacity of the road section and the number of vehicles in each vehicle group in the road section.

[0069] Its expressions include:

[0070]

[0071]

[0072] in, Indicates road section The delay, is the road section from the i-th intersection to the j-th intersection, Indicates the shortest time a vehicle takes to pass through a road section. represents the total number of vehicles moving from the i-th intersection to the j-th intersection in the shortest time period, Indicates road section The load capacity, α represents the first constant, β represents the second constant, p represents the vehicle group, S represents the total number of vehicle groups, a k，p represents the kth strategy selected at the i-th intersection, A p Represents a set of policies, Indicates that the road In the above example, the strategy a is selected in the vehicle group p. k,p The proportion of vehicles in p |Indicates that vehicle group p is currently traveling on the road segment The number of vehicles, Indicates that the vehicles in group p follow strategy a k,p Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0. Indicates road section The current number of vehicles traveling.

[0073] Determine the travel time to the temporary destination based on the road sections passed and the corresponding delays.

[0074] Its expressions include:

[0075]

[0076] Among them, d link,k,p Indicates the vehicle execution strategy a in vehicle group p k,p The time of arrival at the temporary destination.

[0077] Based on the time the vehicle spends at the intersection and the time it takes to reach the temporary destination, the benefit relationship of the strategy is established. Its expression includes:

[0078] c k，p =-(d crass，k,p +d link，k，p )

[0079] Among them, c k，p Indicates vehicle group p Vehicle execution strategy a k，p The income, d crass，k，p Indicates the vehicle execution strategy a in vehicle group p k，p Time spent at intersections.

[0080] In this embodiment, d crass，k，p is a preset threshold, where the delay for a vehicle to turn around at an intersection is 20s, the delay for a vehicle to turn left at an intersection is 15s, the delay for a vehicle to turn right at an intersection is 10s, and the delay for a vehicle to go straight at an intersection is 1s.

[0081] S230 establishes an average revenue relationship for each vehicle group based on the vehicle proportion of each strategy in each vehicle group and the revenue relationship of the corresponding strategy.

[0082] Its expressions include:

[0083]

[0084] in, represents the average revenue of vehicle group p.

[0085] Construct a game model based on constraints and average payoff relationships.

[0086] Its expressions include:

[0087]

[0088] V p ={Car sperial ,Car general}

[0089]

[0090]

[0091] μ>1

[0092] 0<θ≤1

[0093]

[0094]

[0095] in, Indicates that the road Above, the classification category in the car group p is the first category Car sperial的 Vehicle corresponding strategy a k，p The proportion of vehicles Indicates that the road Above, the classification category in the car group p is the second type Car general Vehicle corresponding strategy a k，p The proportion of vehicles in the vehicle, ρ represents the first weight, and θ represents the second weight.

[0096] In order to find a community state Δ(X) in which each player cannot find a strategy better than the one currently selected, this embodiment needs to find a Nash equilibrium point of the game model G, which can be expressed as follows:

[0097]

[0098] S3 obtains the Nash equilibrium state of the game model according to the evolutionary game model, obtains the optimal proportion of vehicles of each strategy in each vehicle group, and obtains the number of vehicles of each strategy in each vehicle group.

[0099] The constructed game model is processed using the evolutionary game model to obtain an ordinary differential equation.

[0100] The ordinary differential equation expression includes:

[0101] The fourth-order Runge-Kutta method is used to solve the ordinary differential equation to obtain the optimal proportion of vehicles in each strategy in each vehicle group.

[0102] For ordinary differential equations:

[0103] Known initial state The iteration step is h, then after one iteration The values ​​are:

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110] After multiple iterations, the game converges dynamically to a stable point. This stable point represents the Nash equilibrium state for all vehicle agents on the current first roadside unit. The first roadside unit then calculates the number of vehicles traveling in each direction for each agent group based on this Nash equilibrium state.

[0111] S4 determines the driving direction of each vehicle based on the number of vehicles and the time when the vehicle arrives at the intersection.

[0112] For each vehicle group and each driving direction, the first roadside unit (RSU) evenly divides the number of vehicles traveling in each direction into a corresponding number of vehicle proxies based on the number of vehicles traveling in each direction calculated in the previous step, and sets the driving direction for each vehicle proxies. The first RSU directly communicates with the vehicles and feeds the recommended routing plan back to the assisted driving system on the corresponding vehicle. The driver then selects the travel route based on the assisted driving system's recommendations.

[0113] The vehicle agent that completes the game and communication will be deleted by the first roadside unit, and the vehicle passing through the intersection will register its agent to the first roadside unit that is about to pass the next intersection, and play again to obtain new routing information.

[0114] In order to show that this embodiment has outstanding substantial effects, this embodiment uses the above method to conduct simulation experiments and analyze the results. Figure 4A and Figure 4B The simulation scenario shown. Figure 4AThe road model in the scenario includes 9 intersections and 24 two-way roads, with 10 road start points and 9 temporary end points. Vehicles enter the network at the start points and exit at the end points. When entering the network, vehicles choose a diagonal intersection as their destination with an 80% probability, and first-class vehicles are assigned a 10% probability. Vehicle arrivals follow a Poisson distribution, with each vehicle having an initial maximum speed of 20 m / s. As vehicles enter each road, their speed is calculated based on the vehicle density on that road. It is assumed that only one first roadside unit is located at the central intersection I1. The vehicle density on the road does not exceed 130 vehicles per kilometer; when the density reaches this value, vehicles queue at the road entrance. Roads within the control area are 1.2 to 1.5 km long, while roads outside the control area are 0.2 km long. The load capacity of all roads is 80 veh / km, α, β, and μ are 0.8, 0.6, and 1.2 respectively, the algorithm step size is set to 0.01, the number of iterations is 1000, and the delay when a vehicle passes through an intersection is a fixed value. The delays for turning around, turning left, turning right, and going straight are 20s, 15s, 10s, and 1s respectively. The time interval between two games in the single intersection multi-vehicle cooperative routing algorithm, that is, the first time period, is set to 30s. Figure 4B The road network model in this scenario includes 25 intersections and 60 two-way roads. The model has 10 road start points and 25 temporary end points. The remaining road settings are the same as those used in the simulation of the single-intersection multi-vehicle collaborative routing algorithm, with a γ value of 1.2. Vehicles enter the road network at their start points according to a Poisson distribution and exit at their end points. A primary roadside unit is placed at each intersection.

[0115] In this experiment, the social routing algorithm and the shortest path algorithm were used as the evaluation benchmarks, and the two important performance indicators for measuring the road traffic network were set as the average vehicle travel time and the throughput of the road network. The number of vehicles arriving at the destination during the simulation experiment was recorded as the throughput of the road network. For vehicles that have not yet reached the destination, their final travel time is uncertain, so the average travel time of all vehicles arriving at the destination is recorded as the average vehicle travel time. For starting points O1 to O8, the average vehicle inflow at each starting point is set to 10veh / min. In order to compare the performance of the multi-vehicle cooperative routing algorithm at a single intersection under different traffic flow conditions, for starting points O9 and O 10 Initially, the average vehicle inflow rate is set to 20 veh / min. The number of vehicles arriving at the destination within an hour and a half is simulated and counted, as well as the average travel time of different types of vehicles. Initially, the first type of vehicles accounts for 10%; then, the vehicle inflow rate of the two starting points is increased to 40 veh / min, 60 veh / min, and 80 veh / min, and the simulation experiment is repeated, and the results are counted. The road throughput under each flow rate is as follows Figure 5A As shown in the figure, the average travel time of vehicles is Figure 5B shown.

[0116] It can be seen that no matter how the traffic volume on the road changes, the multi-vehicle collaborative optimization algorithm proposed in this embodiment has the maximum throughput or a throughput close to the maximum. 10 When the vehicle inflow rate is 80veh / min, the shortest path algorithm does not divert traffic at all, resulting in vehicle congestion, which increases travel time dramatically and reduces throughput. 10 When the vehicle inflow rate is 20veh / min, the outflow number decreases instead of increasing. Compared with the shortest path algorithm, the social routing algorithm reroutes one vehicle in each group of vehicles every 30 seconds, which plays a certain role in diverting traffic. 10 When the vehicle inflow rate increases to 60veh / min, it still maintains a relatively good throughput. When the vehicle inflow rate increases to 80veh / min, there is also a significant improvement compared to the shortest path algorithm.

[0117] Regarding the average travel time of vehicles, the algorithm proposed in this embodiment obtains the minimum average travel time under different vehicle flows. In detail, the average travel time of the first type of vehicles is significantly shortened, while the average travel time of the second type of vehicles is only slightly increased, achieving a good balance between individual needs and road network coordination performance. When the traffic volume is small, the shortest path algorithm can quickly reach the destination because the vehicle chooses the shortest path and there is no congestion on the shortest path; however, as the traffic volume increases, the road gradually becomes congested, and the average travel time of vehicles continues to increase. Compared with the shortest path algorithm, the social routing algorithm has a significantly reduced average travel time, but when the traffic density continues to increase, the algorithm cannot achieve effective diversion to reduce the increase in delay caused by congestion.

[0118] In order to further evaluate whether the algorithm proposed in this embodiment can still balance individual needs and road network coordination performance under different priority vehicle ratio environments, the proportion of the second type of vehicles in the game vehicles is set to 10%, 15%, 20%, 25% and 30%, and the starting points O9 and O 10 The vehicle inflow rate is set to 60veh / min and 80veh / min respectively. The average travel time of the vehicles is further analyzed and the results are as follows Figure 6A and Figure 6B shown.

[0119] It can be seen that no matter how the vehicle inflow rate changes, or what proportion the second type of vehicles accounts for in all the gaming vehicles, the average travel time of the second type of vehicles is significantly lower than that of the first type of vehicles. However, the travel time of the first type of vehicles is not significantly higher than the average travel time of all the gaming vehicles. This fully demonstrates that the algorithm proposed in this embodiment can still balance individual needs and road network coordination performance under different ratios of priority vehicles.

[0120] In order to further judge whether the algorithm proposed in this embodiment has effectively diverted the vehicles, when the starting points O9 and O 10 When the average vehicle inflow rate is 80veh / min and the first type of vehicles accounts for 10%, the traffic density on the roads controlled by the first roadside unit is counted every 15 minutes during each simulation. Since there are vehicles traveling in both forward and reverse directions on the same road, the density of vehicles traveling in the forward direction is represented by a positive number, and the density of vehicles traveling in the reverse direction is represented by a negative number. Here, the directions from south to north and from west to east are assumed to be positive. The vehicle density distribution of the three algorithms at each moment is shown as follows: Figures 7A-7F As shown, Time slicing represents the time period, Crossing number represents the intersection number, and Traffic density represents the vehicle density.

[0121] from Figure 7A and Figure 7B It can be seen that the method proposed in this embodiment has played a good role in diverting traffic. Although the vehicle density on the forward road of road No. 1 and the reverse road of road No. 2 is higher than that on other roads, the overall distribution of vehicles is relatively balanced. And because vehicles are diverted to idle roads in a timely manner, there is no situation of excessive vehicle density and slow traffic on most roads. Figure 7E and Figure 7F The shortest path algorithm, because it failed to dynamically adjust routes and divert vehicles based on road conditions, resulted in saturated traffic density on some roads, while traffic density on others was extremely low. Furthermore, due to road congestion, vehicles continued to pile up, increasing the number of congested sections. It can be seen that at the 15th minute mark of the simulation, only the forward direction of Road 1 and the reverse direction of Road 2 had high traffic density. However, by the 90th minute of the simulation, traffic density had reached saturation on 14 roads, including the forward and reverse directions of Roads 1 and 2. Figure 7C and Figure 7D In the later stage of the simulation experiment, the social routing algorithm also showed that some roads had high traffic density and road congestion.

[0122] Example 2

[0123] Large-scale road networks are composed of multiple intersections and road sections. The control ranges of different intersections in the road network overlap and the cognitive range is limited. In addition, when building a game model at a single intersection to calculate the benefits of each strategy, the time required to reach the temporary destination is estimated based on the current road state. However, when a vehicle travels to the next game intersection, the road state may deviate to varying degrees from the previously predicted state. In other words, the traffic state on unconnected roads has changed significantly, which may cause the vehicle to receive conflicting routing plans at the previous and next intersections. Figure 3 The route recommendation a vehicle receives at intersection I5 may be different from the one it receives at intersection I8. Therefore, coordinating vehicle routing decisions at different intersections in the road network to achieve flow balance, ensure continuity of traffic scheduling, and effectively allocate road resources is a major challenge.

[0124] Therefore, this embodiment provides a method for dynamic routing of multiple vehicles at multiple intersections, including:

[0125] S1 obtains information on vehicles belonging to the third category at each intersection. The vehicles belonging to the third category represent vehicles that have not reached the temporary destination and are traveling according to the strategy recommended by the first roadside unit at the upstream intersection.

[0126] Because there are many vehicles participating in route planning at multiple intersections, and each vehicle has different attributes, some vehicles have urgent traffic needs; some vehicles are following the strategy recommended by the previous game intersection and prefer to ensure the continuity of traffic scheduling when arriving at this intersection; some vehicles are just entering the game intersection; and some vehicles have already completed the recommended strategy at the previous intersection. Therefore, these vehicles still need to be classified. Unlike Example 1, the classification categories of this embodiment include the first, second, and third categories.

[0127] The first category includes priority vehicles, the second category includes ordinary vehicles, the third category includes directional vehicles, and vehicles belonging to the first category Car special There is an urgent need to pass through and hope to pass through the unobstructed road section; vehicles belonging to the second category Car general Vehicles that are not in a high traffic urgency or have completed the recommended strategy at the previous intersection; Vehicles belonging to the third category Car toward Vehicles with very clear road section directions drive according to the strategy recommended by the previous game intersection and prefer to ensure the continuity of traffic scheduling when arriving at the intersection.

[0128] The second roadside unit monitors the vehicle's strategy execution in real time. Specifically, the second roadside unit monitors the vehicle's status of executing the strategy obtained by the multi-vehicle dynamic path selection method based on a single intersection described in Example 1. If the temporary destination has not been reached, the second roadside unit continues to monitor, and the vehicle drives according to the strategy recommended by the first road test unit at the upstream intersection; if the temporary destination is reached, the second roadside unit sends the first status information of the vehicle to the first road test unit, so that it participates in the game and obtains the driving strategy for the next temporary destination.

[0129] The second roadside unit classifies the vehicle based on the vehicle request and the policy execution, including: if the vehicle sends a priority passage request to the second roadside unit and the second roadside unit receives the request, the vehicle belongs to the first category; if the vehicle arrives at the temporary end point and the second roadside unit does not receive or rejects the priority passage request, the vehicle belongs to the second category; if the vehicle does not arrive at the temporary end point, the vehicle belongs to the third category.

[0130] To take care of vehicles belonging to the first category Car special And ensure that the vehicle belongs to the third category Car toward The continuity of traffic dispatch and effective coordination between multiple intersections require a third type of vehicle, Car toward Do not participate in the intersection game, continue to move according to the previous strategy, and also need to belong to the second type of vehicle Car general Concessions can be made to a certain extent, but of course, the traffic efficiency of all participating vehicles needs to be considered.

[0131] Therefore, the vehicle information belonging to the third category at each intersection is obtained, and then the vehicles controlled by the first roadside unit at the current intersection are obtained. The first roadside unit, as shown in Example 1, processes the vehicles belonging to the first category and the vehicles belonging to the second category through the constructed game model, and then obtains the Nash equilibrium state of the game model according to the evolutionary game model, and then obtains the driving direction of the vehicles belonging to the first category and the vehicles belonging to the second category.

[0132] S2 is based on the vehicle information belonging to the third category at each intersection, and processes the remaining vehicles that arrive at the corresponding intersection within the first time period at the intersection using the multi-vehicle dynamic path selection method based on a single intersection described in Example 1 to obtain the driving direction of each vehicle controlled by the first roadside unit at each intersection.

[0133] Different from Example 1, this embodiment introduces a third type of vehicle, so the constructed game model is slightly different. The method for calculating the travel time to the temporary destination includes:

[0134] Obtain the road sections that the vehicle passes through while traveling to the temporary destination based on the direction provided by the first roadside unit; the passed road sections are represented by 1, and the unpassed road sections are represented by 0. Indicates strategy a k，p Whether it passes through the road section If it passes, it is 1, otherwise it is 0.

[0135] The time delay of passing the road section is obtained based on the shortest time for vehicles to pass through the road section, the number of vehicles in the road section in the first roadside unit control area, the number of third-category vehicles traveling on the road section, the load capacity of the road section and the number of vehicles in each vehicle group in the road section.

[0136] Its expressions include:

[0137]

[0138]

[0139] in, Indicates road section The delay, is the road section from the i-th intersection to the j-th intersection, Indicates the shortest time a vehicle takes to pass through a road section. represents the total number of vehicles moving from the i-th intersection to the j-th intersection in the shortest time period, Indicates road section The load capacity, α represents the first constant, β represents the second constant, p represents the vehicle group, S represents the total number of vehicle groups, a k，p represents the kth strategy selected at the i-th intersection, A p Represents a set of policies, Indicates that the road In the above example, the strategy a is selected in the vehicle group p. k，p The proportion of vehicles, p | indicates that vehicle group p is currently traveling on the road segment The number of vehicles, Indicates that the vehicles in group p follow strategy a k，p Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0, Indicates road section The current number of vehicles traveling, Indicates the road segment you are currently traveling on The third type of vehicle Car toward The number of

[0140] Determine the travel time to the temporary destination based on the road sections passed and the corresponding delays.

[0141] Its expressions include:

[0142]

[0143] Among them, d link，k，p Indicates the vehicle execution strategy a in vehicle group p k，p The time of arrival at the temporary destination.

[0144] Furthermore, a strategy benefit relationship is established based on the time a vehicle spends at the intersection and the travel time to the temporary destination. An average benefit relationship for each vehicle group is established based on the proportion of vehicles using each strategy in each vehicle group and the benefit relationship of the corresponding strategy. The strategy includes a driving direction selected for a vehicle by the first roadside unit; when a vehicle group is classified as the first category, the proportion of vehicles using the corresponding strategy is increased based on a third weight; and when a vehicle group is classified as the second category, the proportion of vehicles using the corresponding strategy is decreased based on a fourth weight.

[0145] Construct a game model based on constraints and average payoff relationships.

[0146] Its expressions include:

[0147]

[0148] V p ={Car sperial ,Car general}

[0149]

[0150]

[0151] γ>1

[0152] 0<τ≤1

[0153]

[0154]

[0155] in, Indicates that the road Above, the classification category in the car group p is the first category Car sperial Vehicle corresponding strategy a k， The proportion of vehicles with p, Indicates that the road Above, the classification category in the car group p is the second type Car general Vehicle corresponding strategy a k，p The proportion of vehicles, γ represents the third weight, and τ represents the fourth weight.

[0156] The first roadside unit collects the number of vehicles on each road section within its control area and the number of third-class vehicles Car toward The routing plan is constructed by calculating the travel time of each strategy to the temporary destination and obtaining the benefits of each strategy based on the preference.

[0157] In order to find a community state Δ(X) in which each player cannot find a strategy better than the one currently selected, this embodiment needs to find a Nash equilibrium point of the game model G, which can be expressed as follows:

[0158]

[0159] The Nash equilibrium state of the game model is determined based on the evolutionary game model. The optimal vehicle ratio for each strategy in each vehicle group is obtained, as well as the number of vehicles in each vehicle group for each strategy. Based on the number of vehicles and the time at which they arrive at the intersection, the driving direction of each vehicle controlled by the first roadside unit at each intersection is determined.

[0160] Furthermore, the first roadside unit and the second roadside unit can cooperate with the traffic light signal control to achieve coordinated decision-making between vehicle routing and traffic light right of way.

[0161] For the problem of multi-vehicle cooperative routing with different traffic demands in a multi-intersection road network scenario, the performance analysis of this embodiment is carried out based on the two evaluation indicators of road throughput and average vehicle travel time, just like the evaluation system of multi-vehicle cooperative problem in a single intersection scenario. Initially, the third type of vehicles accounted for 30%, and the first type of vehicles accounted for 10% of the game vehicles; the inflow rate from starting point O1 to O8 was set to 20veh / min, and the starting points O9 and O 10 The vehicle inflow rate is initially set to 20veh / min. Then, in order to evaluate the performance of the algorithm under the condition of heavy traffic on the road, the starting points O9 and O 10 The vehicle inflow rate is increased to 40 veh / min, 60 veh / min, and 80 veh / min in sequence, and simulations are performed under these four conditions.

[0162] like Figure 8A and Figure 8B As shown, this embodiment provides a multi-vehicle dynamic path selection method for multiple intersections, which always has the maximum road network throughput regardless of the vehicle inflow rate; the shortest path algorithm is used as the benchmark algorithm, and the shortest path algorithm is used as the starting point O9 and O 10 The vehicle inflow rate is 20veh / min and 40veh / min, it has a good performance. However, when the traffic volume increases to 60veh / min and 80veh / min, the network throughput capacity begins to decline significantly. 10When the inflow rate is 60veh / min, the throughput of the road network is still very large. 10 When the inflow rate increases to 80 veh / min, the performance begins to decline. This shows that these two algorithms cannot cope well with changes in traffic demand and have poor responsiveness.

[0163] In order to further evaluate whether the algorithm proposed in this embodiment can still balance individual needs and road network coordination performance under different priority vehicle ratio environments, in terms of numerical settings, initially, the third type of vehicles accounted for 30%, and the proportion of the first type of vehicles in the game vehicles was set to 10%, 15%, 20%, 25% and 30%. 10 The vehicle inflow rate is set to 60veh / min and 80veh / min respectively. The average travel time of the vehicles is further analyzed and the results are as follows Figure 9A and Figure 9B shown.

[0164] It can be seen that no matter how the vehicle inflow rate changes, or what proportion the first type of vehicles accounts for in all the gaming vehicles, the average travel time of the first type of vehicles is significantly lower than that of ordinary vehicles. However, the travel time of the second type of vehicles is not significantly higher than the average travel time of all the gaming vehicles. This fully demonstrates that the algorithm proposed in this embodiment can still balance individual needs and road network coordination performance under different ratios of priority vehicles.

[0165] In order to further judge whether the algorithm proposed in this embodiment has effectively diverted the vehicles, when the starting points O9 and O 10 The average vehicle inflow rate is 80 vehicles per minute. Initially, the third category of vehicles accounts for 30% and the second category of vehicles accounts for 10%. During each simulation, traffic density statistics on the roads within the control range of the first roadside unit are collected every 15 minutes. The vehicle density distributions for the three algorithms at each time are obtained. It can be seen that the shortest path algorithm, as an algorithm that completely disregards load balancing, only accumulates and creates congestion on the links with the shortest physical distance. The remaining roads, by comparison, have very little traffic. Compared to the shortest path algorithm, the social routing algorithm achieves a certain degree of road load balancing by rerouting one vehicle in each group, effectively assigning this vehicle to a less-loaded road. However, it can be seen that this algorithm still results in high vehicle density and congestion on many roads. The algorithm proposed in this embodiment, on the other hand, can promptly divert vehicles to roads with lower traffic density, allowing them to quickly reach their destination and exit the road network, preventing any road from reaching saturation with vehicle density. Overall, the multi-vehicle dynamic routing method proposed in this embodiment achieves a good traffic diversion effect.

[0166] Compared with the existing technology, Examples 1 and 2 of this invention propose a multi-vehicle dynamic path selection method based on single intersections and multiple intersections, which transforms the problem of collaborative routing of vehicle groups in a macroscopic large-scale road network scenario into a collaborative decision-making problem at a microscopic intersection, effectively reducing the difficulty of modeling and solving the problem of collaborative routing decision-making of vehicle groups in large-scale scenarios. By dividing different vehicle types and assigning corresponding weights, the goal of taking care of the vehicle priority demand and balancing the road network is achieved, and the universality of the model is also improved. The interests of a single vehicle and the collaborative operation of multiple vehicles are guaranteed, so that the average travel time of vehicles, the road network throughput and the load balancing indicators of the road all have good performance.

[0167] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for dynamic routing of multiple vehicles at a single intersection, characterized in that: include: The first roadside unit acquires first status information, where the first status information includes a classification category of the vehicle and a destination of the vehicle; Based on the first state information, the first roadside unit processes vehicles arriving at the intersection where the first roadside unit is located within a first time period using a constructed game model, including: setting a temporary destination according to the vehicle's destination; grouping the vehicles based on the road section and the temporary destination to obtain multiple vehicle groups; and establishing a strategy benefit relationship based on the preset time consumed by the vehicles at the intersection and the travel time to the temporary destination, wherein the strategy benefit relationship is expressed as follows: in, Indicates vehicle group Vehicle execution strategy of income, Indicates vehicle group Vehicle execution strategy The preset time at the intersection, Indicates vehicle group Vehicles in the intersection execute strategy Time of arrival at the temporary destination, Indicates that from Intersection to the The road section at the intersection, Indicates the number of road sections within the control area of ​​the first DT unit, Indicates vehicle group Vehicles follow the strategy Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0, Indicates road section Based on the vehicle proportion of each strategy in each vehicle group and the profit relationship of the corresponding strategy, the average profit relationship of the vehicle group is established, and its expression includes: in, Indicates vehicle group The average income, Indicates the selected strategy, Represents a set of policies, Indicates that the road Up, crew Select a strategy The proportion of vehicles Indicates vehicle group Vehicle execution strategy The benefit; wherein the strategy includes a driving direction selected by the first roadside unit for the vehicle; when the classification category in the vehicle group is the first category, the proportion of vehicles adopting the corresponding strategy is increased based on the first weight; when the classification category in the vehicle group is the second category, the proportion of vehicles adopting the corresponding strategy is reduced based on the second weight; Obtaining the Nash equilibrium state of the game model according to the evolutionary game model, obtaining the optimal proportion of vehicles for each strategy in each vehicle group, and obtaining the number of vehicles for each strategy in each vehicle group; Based on the number of vehicles and the time they arrive at the intersection, the direction of travel of each vehicle is determined.

2. The method for dynamic routing of multiple vehicles at a single intersection according to claim 1, characterized in that: The multi-vehicle dynamic path selection method based on a single intersection includes: The second roadside unit classifies the vehicle based on the vehicle request, wherein the classification categories include a first category and a second category; The second roadside unit acquires first state information and second state information of the vehicle, where the second state information includes motion parameters of the vehicle during driving; The first roadside unit obtains first status information, including: The second roadside unit determines whether the vehicle arrives at the intersection where the first roadside unit is located within a first time period based on the second state information of the vehicle; if the vehicle can arrive, the second roadside unit sends the first state information of the vehicle to the first roadside unit.

3. The method for dynamic routing of multiple vehicles at a single intersection according to claim 1, characterized in that: The number of vehicle groups is equal to the number of temporary destinations multiplied by the number of road sections where the vehicles are located.

4. The method for dynamic routing of multiple vehicles at a single intersection according to claim 1, characterized in that: Methods for calculating travel time to a temporary destination include: Obtain the road sections that the vehicle passes through while traveling toward the temporary destination based on the direction provided by the first roadside unit; where the passed road sections are represented as 1, and the unpassed road sections are represented as 0; Obtaining the time delay for passing the road section based on the shortest time for vehicles to pass the road section, the number of vehicles in the road section in the first roadside unit control area, the load capacity of the road section, and the number of vehicles in each vehicle group in the road section; Determine the travel time to the temporary destination based on the road sections passed and the corresponding delays.

5. The method for dynamic routing of multiple vehicles at a single intersection according to claim 4, characterized in that: The time delay for passing the road section is obtained based on the shortest time for vehicles to pass the road section, the number of vehicles in the road section in the first roadside unit control area, the load capacity of the road section, and the number of vehicles in each vehicle group in the road section. The expression includes: in, Indicates road section The delay, For the Intersection to the The road section at the intersection, Indicates the shortest time a vehicle takes to pass through a road section. represents the first constant, Indicates that within the shortest time period Move to the intersection The total number of vehicles at the intersection, Indicates road section Load capacity, represents the second constant, Indicates the vehicle group, Indicates the total number of vehicles. Indicates the selected strategy, Represents a set of policies, Indicates that the road Up, crew Select a strategy The proportion of vehicles Indicates vehicle group Currently traveling on the road The number of vehicles, Indicates vehicle group Vehicles follow the strategy Whether the vehicle passes through the road section If you pass by, The value of is 1, otherwise, The value of is 0, Indicates road section The current number of vehicles traveling.

6. The method for dynamic routing of multiple vehicles at a single intersection according to claim 1, characterized in that: The Nash equilibrium state of the game model is obtained according to the evolutionary game model, and the optimal proportion of vehicles in each strategy in each vehicle group is obtained, including: The constructed game model is processed using the evolutionary game model to obtain the ordinary differential equation; The fourth-order Runge-Kutta method is used to solve the ordinary differential equation to obtain the optimal proportion of vehicles in each strategy in each vehicle group.

7. A multi-vehicle dynamic path selection method based on multiple intersections, characterized in that: include: Obtaining information about vehicles belonging to the third category at each intersection, where the vehicles belonging to the third category represent vehicles that have not reached a temporary destination and are traveling according to a strategy output by a first roadside unit at an upstream intersection; Based on the vehicle information belonging to the third category at each intersection, the remaining vehicles arriving at the corresponding intersection within the first time period at the intersection are processed according to the multi-vehicle dynamic path selection method based on a single intersection as described in any one of claims 1 to 6 to obtain the driving direction of each vehicle controlled by the first roadside unit at each intersection.

8. The method for dynamic routing of multiple vehicles at multiple intersections according to claim 7, characterized in that: include: On non-intersecting road sections, vehicles travel to the temporary destination according to the shortest distance route. If there are multiple shortest distances, the vehicles in the vehicle group are evenly divided into multiple groups, and each group corresponds to the shortest distance route one by one.

Images