Vehicle dispatching method and system for two-way two-lane intersection without signal

Abstract

This invention provides a vehicle scheduling method and system for a two-way, two-lane unsignalized intersection, comprising: Step S1: Preprocessing vehicles based on given real-world traffic conditions at a two-way, two-lane intersection to determine the likelihood of collisions and analyze conflict relationships between vehicles; Step S2: Constructing a vehicle scheduling model based on deep search of conflict relationships; Step S3: Utilizing nonlinear programming to make vehicle action decisions, determining the optimal passage time and vehicle acceleration; and based on the vehicle scheduling model, performing lower-level planning to design a vehicle scheduling strategy for the unsignalized intersection; Step S4: Building a simulation scenario of the intersection based on a given real-world intersection scenario and initializing vehicle information for each vehicle; Step S5: Inputting the vehicle information into the vehicle scheduling model to make behavioral decisions and obtain the action output of each vehicle at the next moment. This invention can improve vehicle traffic efficiency and reduce the probability of accidents.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, specifically to a vehicle scheduling method and system for a two-way, two-lane, unsignaled intersection based on a queue model, and more particularly to a vehicle scheduling method and system for a two-way, two-lane, unsignaled intersection. Background Technology

[0002] The development of intelligent vehicles is an important indicator of a country's scientific research strength and industrial level. Intelligent transportation technology utilizes advanced electronic and information technologies to control the driving of intelligent vehicles, automating routine, sustained, and fatigue-inducing operations during driving. In the field of intelligent transportation systems, unsignaled intersections are an important research area. Traditional intersections typically rely on traffic lights to control vehicle traffic, but traffic light scheduling easily leads to traffic congestion and delays. Therefore, researching vehicle scheduling methods for unsignaled intersections is of great significance for improving traffic efficiency.

[0003] Previous studies have used methods to manage traffic by building vehicle conflict graphs, but these methods often fail to adequately consider the actual conditions at intersections. Furthermore, other methods employ optimization algorithms such as genetic algorithms for traffic management, but these suffer from high computational complexity and poor real-time performance.

[0004] Therefore, it is necessary to propose a vehicle scheduling method for unsignalized intersections that can take into account both vehicle priority and the actual conditions of the intersection, so as to improve traffic efficiency and reduce the probability of accidents. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a vehicle dispatching method and system for two-way, two-lane unsignalized intersections.

[0006] According to the present invention, a vehicle dispatching method and system for a two-way, two-lane, unsignaled intersection is provided, the scheme of which is as follows:

[0007] Firstly, a vehicle dispatching method is provided for a two-way, two-lane, unsignaled intersection, the method comprising:

[0008] Step S1: Based on the given real two-way two-lane intersection traffic situation, preprocess the vehicles to determine whether a collision will occur between the vehicles and analyze the conflict relationship between the vehicles.

[0009] Step S2: Construct a vehicle scheduling model based on deep search of conflict relationships;

[0010] Step S3: Use nonlinear programming to make vehicle action decisions, and find the optimal passage time and vehicle acceleration; based on the vehicle scheduling model, perform lower-level planning and design a vehicle scheduling strategy for unsignalized intersections.

[0011] Step S4: Build a simulation scenario of the intersection based on the given real intersection scenario, and initialize the vehicle information of each vehicle;

[0012] Step S5: Input the vehicle information into the vehicle scheduling model, make behavioral decisions, and obtain the action output of each vehicle at the next moment.

[0013] Preferably, in step S1, vehicle preprocessing includes setting vehicle constraints and designing a queue sorted according to the distance of vehicles from the center point of the intersection.

[0014] In a two-way, two-lane scenario model, the current vehicle conflict relationships at the intersection include: Intersecting relationships: the conflict between straight-ahead traffic to the east and left-turn traffic to the north; Diverging relationships: the conflict between straight-ahead traffic to the east and left-turn traffic to the east; Converging relationships: the conflict between straight-ahead traffic to the west and left-turn traffic to the south; the remaining conflict points in the east, west, and north directions are obtained by rotating the above conflict routes by 90°, 180°, and 270° respectively based on the center of the intersection; the conflict set is generated by the union of the intersecting, diverging, and convergent relationships.

[0015] In a two-way, two-lane road scenario model, the vehicle constraints include: a minimum speed of 0 km / h, a maximum speed of 30 km / h, and a minimum acceleration of -1.5 m / s². 2 Maximum acceleration 1.0 m / s² 2 The minimum vehicle spacing is 15m.

[0016] Preferably, in step S2, the vehicle scheduling model based on deep search of conflict relationships specifically includes: first, using matrix theory to construct the conflict relationships between vehicles; second, using the deep search tree algorithm to determine the order in which vehicles enter the intersection, so that vehicles in the queue do not have conflict relationships and travel within a safe range in terms of speed and spacing.

[0017] Preferably, in step S2, the matrix structure is as follows:

[0018] In this matrix, the number of vehicles in the queue at the intersection is n, and each element is a. ij , when a ij =1 indicates that vehicle j has a conflict relationship with vehicle i, and when a ij =0 indicates that there is no conflict between vehicle j and vehicle i;

[0019] Since the conflict relationship is directed, the adjacency matrix is ​​not a symmetric matrix; where i, j = 0, 1, 2, ..., n, when i, j = 0, the car is a virtual lead car that only serves as a guide, and the rest are follow cars.

[0020] Preferably, step S2 specifically includes the following steps:

[0021] Step S201: Sort the vehicles within the intersection area in order of their distance from the center point of the intersection, with the vehicles closer to the center point at the front; based on the sorting result, form an ordered vehicle queue, and number the vehicles according to their position in the sorting to form a queue model from vehicle number 1 to vehicle number N; set a virtual lead vehicle with the number 0 to represent the center point of the intersection.

[0022] Step S202: When vehicle Ci enters the intersection control area, determine whether the planning of the vehicles ahead is completed. If not, proceed to step S203; if completed, determine the conflict relationship of the vehicles ahead of vehicle Ci and select the vehicle with the largest number as the leading vehicle of Ci.

[0023] Step S203: Starting from the virtual lead vehicle, select an unvisited vehicle as the root node. Based on the elements in the conflict relationship matrix of the current node, determine the vehicles that have conflict relationships with the root node vehicle as child nodes through the depth-first search theory. Form a priority intersection passage queue with vehicles at the same depth that do not have conflict relationships. The closer the vehicle queue is to the root node, the higher its priority.

[0024] Step S204: For each optional next vehicle / child node, check the conflict relationship matrix and the conflict situation of the selected vehicles under the same priority queue. If there is no conflict, add the vehicle to the vehicle driving queue at the same depth, set the vehicle with the largest number in the queue as the root node, and execute step S203.

[0025] Step S205: During the access process, if a root node vehicle has no child nodes, then set other vehicles in the same priority queue as root nodes and execute step S203.

[0026] Step S206: During the access process, if a vehicle of a certain root node has no child nodes and there are no other vehicles in the same priority queue, the access will move to the previous priority queue until all vehicles are grouped together.

[0027] Step S207: The priority queue obtained by the depth-first search algorithm is the order in which vehicles enter the intersection. The higher priority queue enters the intersection first, and vehicles of the same priority must ensure that they pass through the center point of the intersection at the same time.

[0028] Preferably, in step S3, the vehicle's action decision is made using nonlinear programming, specifically including the following steps:

[0029] Step S301: Calculate the optimal acceleration for each vehicle using nonlinear programming;

[0030] Step S302: Calculate the time and distance each vehicle needs to accelerate from its initial speed to its maximum speed; to solve for the optimal time for vehicles to pass through an intersection under the same depth matrix, based on the intersection constraints and the displacement-time formula under uniform acceleration:

[0031]

[0032] Where d represents the distance of the vehicle from the intersection, v0 represents the initial speed of the vehicle, a represents the acceleration to be solved, and t is the vehicle travel time;

[0033] Step S303: Check if the vehicle can reach the intersection before reaching the maximum speed: If the distance traveled by the vehicle is greater than or equal to the initial distance, directly use the quadratic equation solution formula to calculate the time it takes for the vehicle to reach the intersection; if the distance traveled by the vehicle is less than the target distance, calculate the remaining distance traveled at the maximum speed, and calculate the time required to travel the remaining distance at the maximum speed.

[0034] Step S304: Call the depth matrix to compare the travel times of vehicles in the same priority queue and solve for the maximum time;

[0035] Step S305: Using the time as the target, calculate the optimal acceleration value for each vehicle in the same queue as it passes through the intersection using the quadratic equation solution formula.

[0036] Preferably, the vehicle information in the simulation scenario of step S4 includes vehicle speed, acceleration, expected travel route, and location information;

[0037] The simulation scenario also includes lane geographic information, specifically the approach lanes, exit lanes, and conflict behavior relationships in each direction; the approach lanes are divided into the intersection priority queue control area and the intersection center area; the exit lanes include the priority queue disbanding area.

[0038] Preferably, step S5 includes:

[0039] Define a unique identifier for each vehicle and specify its travel route, including the roads and intersections it passes through; simulate vehicle arrival patterns by setting vehicle traffic flow to a Poisson distribution; and process the scheduling cycle to achieve vehicle scheduling and traffic flow simulation. The formula for calculating the scheduling cycle is as follows:

[0040]

[0041]

[0042]

[0043] Where T represents the scheduling time; di S1 represents the distance of the vehicle from the center of the intersection; S2 represents the displacement distance at the vehicle's maximum speed; v represents the displacement distance at the vehicle's minimum speed. max Indicates the maximum speed of the vehicle; v min This represents the minimum speed at which a vehicle travels. Through the above processing, the vehicle scheduling model accurately simulates vehicle movement and traffic flow based on the vehicle's unique identifier and route, combined with the simulation of the Poisson distribution of vehicle arrivals and the calculation of the scheduling cycle.

[0044] Preferably, in step S5, the actions of each vehicle at the next moment are output, specifically including: calling relevant information such as vehicle waiting time, number of waiting vehicles, and number of passing vehicles at a certain time step; installing environmental sensors and traffic flow sensors in the four entrance lanes (east, west, south, and north) to read vehicle occupancy, traffic flow information, CO2 emissions, and fuel emissions; and interfaceing the vehicle action decision model implemented in step S3 to schedule vehicles every scheduling time step, allowing convoys to pass according to priority, setting the time interval between high-priority convoys and low-priority vehicles at intersections to resolve conflicts and reduce parking waiting time.

[0045] Secondly, a vehicle dispatching system for a two-way, two-lane, unsignaled intersection is provided. The system includes an intelligent vehicle dispatching system and a vehicle speed and acceleration calculation and control unit, and has V2X communication function for all vehicles in the queue.

[0046] The intelligent vehicle scheduling system calculates and controls the vehicles by the control unit, as well as the formation and movement control of the formation based on the depth-first search tree.

[0047] The vehicle speed and acceleration calculation and control unit acquires vehicle queue information and performs real-time scheduling planning for vehicles based on basic vehicle information and queue priority.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. Integration of Vehicle Queuing and Intelligent Scheduling System: This invention combines the consistency and efficiency of vehicle queuing with an intelligent scheduling system. By analyzing the conflict relationships between vehicle travel directions in lanes and utilizing a depth-first search method, vehicles are grouped into priority queues. This integration enables vehicles to pass through intersections in an orderly manner, significantly improving road space utilization and vehicle traffic efficiency. Furthermore, this method features low complexity and strong real-time performance; subsequent vehicles entering the intersection control area do not require recalculation, reducing the system's communication and computational burden.

[0050] 2. Real-time computation to obtain the optimal solution: This invention is based on vehicle depth-first search and obtains the optimal solution during vehicle passage through real-time computation and real-time data. Considering the dynamic nature of vehicle driving status and trajectory changes, this algorithm can adjust the priority queue in real time to optimize vehicle passage efficiency.

[0051] 3. Priority platooning model-based traffic sequence calculation: This invention employs a priority platooning model to calculate the traffic sequence for each priority platoon, making the calculation more reasonable and accurate. By acquiring real-time vehicle platooning information in the intersection area and using this model for conflict prediction and traffic priority determination, the system possesses stronger real-time performance.

[0052] 4. Nonlinear Programming and Speed ​​Guidance: This invention employs nonlinear programming and simultaneous vehicle speed planning algorithms to provide suggested speed guidance for vehicles in the intersection control area. Combining real-time intersection conditions and platooning information in each direction, as well as the location and timing of collisions, this algorithm effectively resolves conflicts while reducing vehicle waiting time and improving traffic comfort.

[0053] Other beneficial effects of the present invention will be explained in detail through the introduction of specific technical features and technical solutions in specific embodiments. Those skilled in the art should be able to understand the beneficial technical effects brought about by these technical features and technical solutions through the introduction of these technical features and technical solutions. Attached Figure Description

[0054] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0055] Figure 1 This is a flowchart of intelligent vehicle dispatching for a queue-based two-lane unsignaled intersection.

[0056] Figure 2 This is a diagram showing the vehicle conflict relationships at a queue-based, two-lane, unsignaled intersection.

[0057] Figure 3 This is a diagram of the intersection lane area in an embodiment of the present invention;

[0058] Figure 4 This is a diagram showing the intersection conflict relationships in an embodiment of the present invention;

[0059] Figure 5 This is a queuing sequence diagram of the intersection model in an embodiment of the present invention;

[0060] Figure 6 This is an example of the queuing order of ten vehicles at an intersection in this embodiment of the invention.

[0061] Figure 7This is a diagram showing the depth-first search tree result in an embodiment of the present invention;

[0062] Figure 8 This is a flowchart of the intersection model priority queue model in an embodiment of the present invention. Detailed Implementation

[0063] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0064] This invention provides a vehicle scheduling method for a two-lane, two-way unsignalized intersection to improve traffic efficiency and reduce the probability of accidents. First, a depth-first search tree based on conflict relationships is constructed by drawing a directed graph of vehicle conflict relationships, realizing a vehicle scheduling queue model for the unsignalized intersection. This model improves traffic efficiency by avoiding vehicle collisions at the intersection. Then, a vehicle scheduling implementation strategy is proposed. By evaluating the shortest passage time for vehicles without conflict relationships, the optimal time for each intelligent vehicle to cross the intersection under the scheduling strategy is planned, and the optimal input (acceleration and speed) for each vehicle is calculated. Finally, simulation experiments are conducted to analyze the feasibility of the strategy, and the vehicle scheduling model for the unsignalized intersection based on the queue model is optimized. (Refer to...) Figure 1 As shown, the method specifically includes the following:

[0065] Step S1: Based on the given real-world traffic conditions at a two-lane, two-way intersection, preprocess the vehicles to determine if collisions will occur and analyze the conflict relationships between them. Vehicle preprocessing includes setting vehicle constraints and designing a queue sorted by distance from the intersection center point.

[0066] In a two-lane, two-way intersection scenario, the current vehicle conflict relationships at the intersection include: Intersecting relationships: conflict between straight-ahead traffic to the east and left-turn traffic to the north; Diverging relationships: conflict between straight-ahead traffic to the east and left-turn traffic to the east; Converging relationships: conflict between straight-ahead traffic to the west and left-turn traffic to the south. The remaining conflict points in the east, west, and north directions are obtained by rotating the above conflict routes by 90°, 180°, and 270° respectively, based on the intersection center. The conflict set is generated by the union of the intersecting, diverging, and convergent relationships.

[0067] In a two-way, two-lane road scenario model, the vehicle constraints include: a minimum speed of 0 km / h, a maximum speed of 30 km / h, and a minimum acceleration of -1.5 m / s². 2Maximum acceleration 1.0 m / s² 2 The minimum vehicle spacing is 15m.

[0068] Step S2: Construct a vehicle scheduling model based on deep search for conflict relationships. The scheduling model involves using a deep search tree algorithm to traverse the directed graph of vehicle conflicts to determine the optimal path for vehicles to pass through the intersection. By traversing and searching, the optimal scheduling strategy can be found that ensures no conflict between vehicles passing through the intersection simultaneously, thus preventing collisions.

[0069] The vehicle scheduling model based on deep search for conflict relationships specifically includes: First, using matrix theory to construct conflict relationships between vehicles. Second, using a deep search tree algorithm to determine the order in which vehicles enter the intersection, ensuring that vehicles in the queue do not have conflict relationships and that their speed and spacing remain within safe ranges.

[0070] The matrix structure is In this matrix, the number of vehicles in the queue at the intersection is n, and each element is a. ij , when a ij =1 indicates that vehicle j has a conflict relationship with vehicle i, and when a ij = 0 indicates that vehicle j has no conflict relationship with vehicle i. Since the conflict relationship is directed, this adjacency matrix is ​​not a symmetric matrix. Where i, j = 0, 1, 2...n, when i, j = 0, the vehicle is a virtual lead vehicle that only acts as a guide, and the rest are follow vehicles.

[0071] The specific steps of step S2 are as follows:

[0072] S201: Order the vehicles within the intersection area according to their distance from the center point of the intersection, with vehicles closer to the center point appearing first. Based on the order, form an ordered vehicle queue, numbering each vehicle according to its position in the queue, creating a queue model from vehicle number 1 to vehicle number N. To ensure safe driving, a virtual lead vehicle (numbered 0) is designated to represent the center point of the intersection.

[0073] S202: When vehicle Ci enters the intersection control area, determine whether the planning for the vehicles ahead is complete. If not, proceed to step S203. If complete, determine the conflict relationship of vehicles ahead of vehicle Ci and select the vehicle with the largest number as Ci's leading vehicle.

[0074] S203: Starting with the virtual lead vehicle, select an unvisited vehicle as the root node. Based on the elements in the conflict relationship matrix of the current node, use depth-first search to determine vehicles with conflict relationships to the root node as child nodes. Form priority queues for vehicles at the same depth that do not have conflict relationships. Queues of vehicles closer to the root node have higher priority.

[0075] S204: For each optional next vehicle / child node, check the conflict relationship matrix of its conflict relationship matrix and the conflict of selected vehicles under the same priority queue. If there is no conflict, add the vehicle to the vehicle driving queue at the same depth, set the vehicle with the largest number in the queue as the root node, and execute step S203.

[0076] S205: During the access process, if a root node vehicle has no child nodes, then set other vehicles in the same priority queue as root nodes and execute step S203.

[0077] S206: During the access process, if a vehicle of a certain root node has no child nodes and there are no other vehicles in the same priority queue, the access will go to the previous priority queue until all vehicles are grouped together.

[0078] S207: The priority queue obtained by the depth-first search algorithm is the order in which vehicles enter the intersection. The higher priority queue enters the intersection first, and vehicles of the same priority must ensure that they pass through the center point of the intersection at the same time.

[0079] Through the above steps, the vehicle scheduling model based on conflict relationship deep search can effectively determine the order in which vehicles enter the intersection, enabling vehicles in the queue to travel within a safe range and avoid conflict relationships.

[0080] Step S3: Use nonlinear programming to make vehicle action decisions and find the optimal passage time and vehicle acceleration. Based on the above vehicle scheduling strategy, design the lower-level architecture. Through calculation and optimization, the optimal passage time and vehicle acceleration can be solved to achieve a conflict-free relationship between vehicles and enable vehicles to pass through the intersection smoothly.

[0081] This step utilizes nonlinear programming to make vehicle action decisions, specifically including the following steps:

[0082] S301: Calculate the optimal acceleration for each vehicle using nonlinear programming;

[0083] S302: Calculate the time and distance each vehicle needs to accelerate from its initial speed to its maximum speed; to solve for the optimal time for vehicles to pass through an intersection under the same depth matrix, the intersection constraints and the displacement-time formula under uniform acceleration can be used:

[0084]

[0085] Where d represents the distance of the vehicle from the intersection, v0 represents the initial speed of the vehicle, a represents the acceleration to be solved, and t is the vehicle travel time.

[0086] S303: Check if the vehicle can reach the intersection before reaching maximum speed: If the distance traveled by the vehicle is greater than or equal to the initial distance, the time to reach the intersection is directly calculated using the quadratic equation formula. This quadratic equation formula is calculated using distance, initial speed, and acceleration. If the distance traveled by the vehicle is less than the target distance, the remaining distance at maximum speed is calculated, along with the time required to travel that remaining distance at maximum speed.

[0087] S304: Call the depth matrix to compare the travel times of vehicles in the same priority queue and solve for the maximum time.

[0088] S305: Using this time as the objective, the optimal acceleration value for each vehicle in the same queue passing through the intersection is calculated using a quadratic equation solution formula.

[0089] Step S4: Simulation scenario. Based on the given real intersection scenario, a simulation scenario of the intersection is built, and the vehicle information of each vehicle in the simulation scenario is obtained. The vehicle information includes the status information of each vehicle, the action information of each vehicle at the previous moment, and the ID information of each vehicle.

[0090] This step involves building a simulated intersection scenario based on a given real intersection scene. Specifically, the simulation scenario includes vehicle information such as vehicle speed, acceleration, expected travel route, and location information; lane geographic information such as approach lanes, exit lanes, and conflict behavior relationships in each direction; approach lanes are divided into an intersection priority queue control area and an intersection center area; and exit lanes include a priority queue disbanding area.

[0091] Step S5: Input the status information of all vehicles into the vehicle scheduling model and the lower-level architecture model to obtain the output of vehicle behavior decision; output the action evaluation of the next action of each vehicle.

[0092] Specifically, the input of the status information of all vehicles into the vehicle scheduling model and the lower-level architecture model includes:

[0093] Define a unique identifier (vehicle ID) for each vehicle and specify its travel route, including the roads and intersections it passes through. To simulate vehicle arrival, vehicle traffic is set to a Poisson distribution. The simulator parameters are set as follows: vehicle arrival time is 2 seconds, vehicle length is 3 meters, and lane width is 4 meters. To achieve vehicle scheduling and traffic flow simulation, the scheduling cycle is processed. The formula for calculating the scheduling cycle is as follows:

[0094]

[0095]

[0096]

[0097] Where T represents the scheduling time; d i S1 represents the distance of the vehicle from the center of the intersection; S2 represents the displacement distance at the vehicle's maximum speed; v represents the displacement distance at the vehicle's minimum speed. max Indicates the maximum speed of the vehicle; v min Indicates the minimum speed at which the vehicle can travel;

[0098] Through the above processing, the vehicle scheduling model can accurately simulate vehicle movement and traffic flow based on the vehicle's unique identifier and route, combined with the simulation of the Poisson distribution of vehicle arrivals and the calculation of the scheduling cycle.

[0099] In this step, the model is invoked to output the action evaluation for the next action of each vehicle, specifically including: retrieving information such as vehicle waiting time, number of waiting vehicles, and number of passing vehicles at a certain time step. Environmental sensors and traffic flow sensors are installed in the four entrance lanes (east, west, south, and north) to read vehicle occupancy, traffic flow information, CO2 emissions, and fuel emissions. The vehicle action decision model implemented in interface step S3 schedules vehicles at every scheduling time step, with the convoy passing according to priority. The time interval between high-priority convoys and low-priority vehicles at intersections is set. Conflicts are resolved and stopping waiting time is reduced.

[0100] This invention also provides a vehicle dispatching system for a two-way, two-lane, unsignaled intersection. The system includes an intelligent vehicle dispatching system and a vehicle speed and acceleration calculation and control unit. All vehicles in the queue have V2X communication capabilities. The intelligent vehicle dispatching system's calculation and control unit controls the vehicles and performs queuing and queuing motion control based on a depth-first search tree. The vehicle speed and acceleration calculation and control unit acquires vehicle queue information and performs real-time dispatching planning based on basic vehicle information and queue priority.

[0101] The present invention will now be described in more detail.

[0102] This invention provides a vehicle dispatching method for a two-way, two-lane, unsignaled intersection, the method specifically comprising:

[0103] Step S1: Based on the given real two-way two-lane intersection traffic conditions, preprocess the vehicles to determine whether a collision will occur between them and analyze the conflict relationship between the vehicles.

[0104] Vehicle information preprocessing includes initializing vehicle speed, acceleration, distance from the intersection center point, and vehicle travel routes. A convergence pattern in vehicle conflict relationships indicates a conflict occurring in the exit lane, where multiple vehicles attempt to enter the same exit lane, leading to a conflict. A divergence pattern indicates a conflict occurring in the entrance lane, where multiple vehicles attempt to enter the intersection from different entrance lanes, resulting in a conflict. For example... Figure 2 As shown, in a two-way, two-lane scenario model, the current vehicle conflict relationships at the intersection include: intersection relationships: conflict between straight-ahead traffic to the east and left-turn traffic to the north; divergent relationships: conflict between straight-ahead traffic to the east and left-turn traffic to the east; and convergent relationships: conflict between straight-ahead traffic to the west and left-turn traffic to the south. The remaining conflict points in the east, west, and north directions are obtained by rotating the above conflict routes by 90°, 180°, and 270° respectively, based on the intersection center. The conflict set is generated by the union of the intersection, divergent, and convergent relationships.

[0105] like Figure 3 As shown, the approach lanes are one-way. To illustrate the conflict relationships between vehicles at the intersection, the lanes are discretized within the intersection area. Simultaneously, the area within a 100m diameter of the approach lanes is designated as the intersection control zone, and the exit lanes are designated as the priority queue disbanding zone.

[0106] As described above, for vehicles in the intersection control area, vehicles are grouped into priority queues, and their speeds are dynamically adjusted to avoid collisions at the intersection and to prevent them from stopping and waiting. The center area of ​​the intersection is the main area where vehicle collisions may occur. The priority queue disbanding area is the area where the entire queue has passed through the intersection and is no longer under the algorithm's control.

[0107] like Figure 4 As shown, there are 12 possible routes for vehicles at a two-lane intersection. Starting from 1 for left turns from the south, the numbers increment counter-clockwise. For each numbered route, there exists a convergence set, an intersection set, and a divergence set. For route number 2, the intersection set is S2 = {4, 7, 11}, the convergence set is Q2 = {6, 10}, and the divergence set is R. k = {1, 2, 3}. The collision set of this vehicle is C2 = {1, 2, 3, 4, 6, 7, 10, 11}. For vehicles traveling in all twelve directions, their collision sets can be obtained, and then sorted according to their numbers to generate a collision matrix.

[0108] Step S2: Construct a vehicle scheduling model based on deep search for conflict relationships. The scheduling model involves using a deep search tree algorithm to traverse the directed graph of vehicle conflicts to determine the optimal path for vehicles to pass through the intersection. By traversing and searching, the optimal scheduling strategy can be found that ensures no conflict between vehicles passing through the intersection simultaneously, thus preventing collisions.

[0109] In the process of constructing the vehicle scheduling model, matrix theory is first used to establish the conflict relationships between vehicles. For example... Figure 5 As shown, all vehicles in the intersection control area are queued sequentially according to their distance from the intersection center point. The lead vehicle is a virtual vehicle located at the intersection center point, with a virtual guiding speed of 15 km / h and numbered 0. Subsequent vehicles are sorted by their distance from the intersection center point, numbered from 1 to N from closest to furthest. Vehicles follow the queue, and subsequent vehicles make decisions based on the vehicle in front, modulating their speed and acceleration. After numbering the vehicles, their driving direction codes are extracted; based on the driving direction conflict set, the conflict relationship between the vehicle and the vehicle in front is determined, and a conflict matrix is ​​output using matrix theory.

[0110] Furthermore, in step S2, the matrix structure is as follows: In this matrix, the number of vehicles in the queue at the intersection is n, and each element is a. ij , when a ij =1 indicates that vehicle j has a conflict relationship with vehicle i, and when a ij = 0 indicates that vehicle j has no conflict relationship with vehicle i. Since the conflict relationship is directed, this adjacency matrix is ​​not a symmetric matrix. Where i, j = 0, 1, 2...n, and when i, j = 0, the vehicle is a virtual lead vehicle that only acts as a guide, and the rest are follow vehicles.

[0111] Secondly, a depth-first search algorithm is used to prioritize vehicles, enabling real-time vehicle control. During actual driving, vehicles acquire information such as acceleration, speed, and distance between surrounding vehicles through V2V communication and onboard sensing devices, including onboard LiDAR. Stable driving is maintained by applying constraints. Simultaneously, in the intersection control area, the intelligent vehicle dispatching system and the vehicle speed and acceleration calculation control unit communicate with the vehicles via V2X. In this area, the vehicle communicates with the system in real time, and the system continuously updates the priority queue using a depth-first search algorithm, thereby modulating the vehicle's acceleration and speed. This keeps the vehicle's speed within a limited range, improving lane utilization and enabling efficient passage through intersection areas.

[0112] In depth-first search, an array is first used to set the state of all nodes to unvisited. Then, nodes that are being visited are marked as visited. By recursively calling the algorithm only on unvisited nodes, it ensures that the algorithm does not get stuck in an infinite loop. Secondly, during the execution of the depth-first search algorithm, vehicles closer to the root node are given higher priority, and during the traversal, each node with the same priority is checked for conflicts. Nodes with conflicts at the same priority but higher node numbers are made children of the nodes with lower node numbers. This ensures that there are no conflicts between vehicles in the same priority queue, allowing them to safely pass through the center point of the intersection. Figure 6 As shown, the vehicles are queued according to their distance from the center point of the intersection, with vehicles having higher numbers being farther from the intersection. Their routes are indicated above the vehicles. When this vehicle set passes through the intersection, the priority model implemented using the queue model and depth-first search tree is as follows: Figure 7 As shown: Vehicles C1 and C2 have the highest priority and pass through the intersection first, followed by C3, C4, and C5. Then, vehicles C6, C7, and C8 pass through the intersection simultaneously. Vehicles C9 and C10 pass through the intersection last. To ensure safe passage, vehicles of the same priority pass through the center point of the intersection simultaneously.

[0113] In step S2, the vehicle scheduling strategy based on conflict relationship depth search specifically includes: using a depth search tree algorithm to determine the order in which vehicles enter the intersection based on the constructed conflict relationship matrix, ensuring that vehicles in the queue do not generate conflicts during their journey and can travel within a safe range. The specific steps are as follows:

[0114] S201: Order the vehicles within the intersection area according to their distance from the center point of the intersection, with vehicles closer to the center point appearing first. Based on the order, form an ordered vehicle queue, numbering each vehicle according to its position in the queue, creating a queue model from vehicle number 1 to vehicle number N. To ensure safe driving, a virtual lead vehicle (numbered 0) is designated to represent the center point of the intersection.

[0115] S202: When vehicle Ci enters the intersection control area, determine whether the planning for the vehicles ahead is complete. If not, proceed to step S203. If complete, determine the conflict relationship of vehicles ahead of vehicle Ci and select the vehicle with the largest number as Ci's leading vehicle.

[0116] S203: Starting with the virtual lead vehicle, select an unvisited vehicle as the root node. Based on the elements in the conflict relationship matrix of the current node, use depth-first search to determine vehicles with conflict relationships to the root node as child nodes. Form priority queues for vehicles at the same depth that do not have conflict relationships. Queues of vehicles closer to the root node have higher priority.

[0117] S204: For each optional next vehicle / child node, check the conflict relationship matrix of its conflict relationship matrix and the conflict of selected vehicles under the same priority queue. If there is no conflict, add the vehicle to the vehicle driving queue at the same depth, set the vehicle with the largest number in the queue as the root node, and execute step S203.

[0118] S205: During the access process, if a root node vehicle has no child nodes, then set other vehicles in the same priority queue as root nodes and execute step S203.

[0119] S206: During the access process, if a vehicle of a certain root node has no child nodes and there are no other vehicles in the same priority queue, the access will go to the previous priority queue until all vehicles are grouped together.

[0120] S207: The priority queue obtained by the depth-first search algorithm is the order in which vehicles enter the intersection. The higher priority queue enters the intersection first, and vehicles of the same priority must ensure that they pass through the center point of the intersection at the same time.

[0121] Through the above steps, the vehicle scheduling model based on conflict relationship deep search can effectively determine the order in which vehicles enter the intersection, enabling vehicles in the queue to travel within a safe range and avoid conflict relationships.

[0122] Step S3: Use nonlinear programming to make vehicle action decisions and find the optimal passage time and vehicle acceleration. Based on the above vehicle scheduling strategy, design the lower-level architecture. Through calculation and optimization, the optimal passage time and vehicle acceleration can be solved to achieve a conflict-free relationship between vehicles and enable vehicles to pass through the intersection smoothly.

[0123] In step S3, nonlinear programming is used to make vehicle action decisions, specifically including the following steps:

[0124] S301: Calculate the optimal acceleration for each vehicle using nonlinear programming;

[0125] S302: Calculate the time and distance each vehicle needs to accelerate from its initial speed to its maximum speed; to solve for the optimal time for vehicles to pass through an intersection under the same depth matrix, the intersection constraints and the displacement-time formula under uniform acceleration can be used:

[0126]

[0127] Where d represents the distance of the vehicle from the intersection, v0 represents the initial speed of the vehicle, a represents the acceleration to be solved, and t is the vehicle travel time.

[0128] S303: Check if the vehicle can reach the intersection before reaching maximum speed: If the distance traveled by the vehicle is greater than or equal to the initial distance, the time to reach the intersection is directly calculated using the quadratic equation formula. This quadratic equation formula is calculated using distance, initial speed, and acceleration. If the distance traveled by the vehicle is less than the target distance, the remaining distance at maximum speed is calculated, along with the time required to travel that remaining distance at maximum speed.

[0129] S304: Call the depth matrix to compare the travel times of vehicles in the same priority queue and solve for the maximum time.

[0130] S305: Using this time as the objective, the optimal acceleration value for each vehicle in the same queue passing through the intersection is calculated using a quadratic equation solution formula.

[0131] In step S302 above, based on the constraints of the vehicle itself in the intersection, if the vehicle speed reaches the maximum speed of 30km / h before the intersection, the vehicle becomes a constant speed vehicle.

[0132] In step S303 above, the longest travel time among all vehicles of the same priority is first selected as the baseline time. Since the distance to the intersection and the initial velocity are known, the formula is used... It can calculate the acceleration values ​​of other vehicles with the same priority.

[0133] like Figure 8 The flowchart illustrates the model construction in steps S1 to S3. First, vehicle information within the intersection control area is queried and sorted by distance from the intersection center point, then numbered. Next, it's determined whether there are any conflicts among the vehicles within the intersection control area. If conflicts exist, a conflict matrix is ​​constructed according to step S202, and a virtual vehicle is added as the lead vehicle. Then, based on the depth-first search tree construction in step S2, a priority queue model is established. This includes: setting the root node; constructing the priority queue; and determining if any vehicle has a conflict with the currently visited node. If there are conflicting vehicles, the vehicle with the highest number is set as the root node, and the priority queue continues to be constructed; if there are no conflicting vehicles, other vehicles of the same priority are set as root nodes, and the previous priority is returned. After all vehicles have been traversed, step S3 is used to nonlinearly program and modulate the vehicle speed and acceleration.

[0134] In step S3, besides using nonlinear programming, intelligent algorithms such as genetic algorithms can also be used to achieve vehicle speed and acceleration modulation. For genetic algorithms, the basic genetic algorithm (SGA) takes all individuals in the population as its object and uses only basic genetic operators (selection, crossover, mutation) for population-based operations. The representation is as follows:

[0135]

[0136] Where: C represents the individual coding scheme, E represents the individual fitness evaluation function, P0 represents the initial population, and M represents the population size. Γ represents the selection operator, Γ represents the crossover operator, Ψ represents the mutation operator, and T represents the termination condition of the genetic algorithm.

[0137] Here, a genetic algorithm is used to find the shortest time required for multiple vehicles to reach their destination, modulating the acceleration and velocity of the vehicles. To simplify the algorithm's complexity, the vehicle trajectories are designed as standard uniformly accelerated motion. A fitness function is defined to calculate the total time, and crossover and mutation operations are used to evolve the individuals. Ultimately, the genetic algorithm finds the shortest time for multiple vehicles to reach their destination and calculates the optimal acceleration and velocity for each vehicle.

[0138] Step S4: Simulation scenario. Based on the given real intersection scenario, build a simulation scenario of the intersection and obtain the vehicle information of each vehicle in the simulation scenario. The vehicle information includes the status information of each vehicle, the action information of each vehicle at the previous moment, and the ID information of each vehicle.

[0139] In step S4, a simulation scenario of the intersection is built based on the given real intersection scene. Specifically, this includes: vehicle information in the simulation scenario, including vehicle speed, acceleration, expected travel route, and location information; lane geographic information, including approach lanes, exit lanes in each direction, and conflict behavior relationships; approach lanes are divided into an intersection priority queue control area and an intersection center area; exit lanes are divided into a priority queue disbanding area. In building the simulation scenario, to simulate vehicle arrival, vehicle traffic is set to a Poisson distribution. The vehicle spacing is calculated to be 20m using a formula.

[0140] D safe =D stop +t driver *v car

[0141] Among them, D safe For a safe following distance, D stop For emergency braking stopping distance, t driver For driver reaction speed, v car This refers to the vehicle's speed.

[0142] In addition, in the simulator constructed in step S4, the intersection and road radius are set to 200m, the intersection control area radius is 150m, and the average initial speed of vehicles is 10km / h. Traffic flow is tested multiple times using actual data.

[0143] Step S5: Input the status information of all vehicles into the vehicle scheduling model and the lower-level architecture model to obtain the output of vehicle behavior decision; output the action evaluation of the next action of each vehicle.

[0144] In step S5, the status information of all vehicles is input into the vehicle scheduling model and the lower-level architecture model, specifically including:

[0145] Define a unique identifier (vehicle ID) for each vehicle and specify the vehicle's route, including the roads and intersections it passes through. The simulator parameters are set as follows: vehicle arrival time is 2 seconds, vehicle length is 3 meters, and lane width is 4 meters. To achieve vehicle scheduling and traffic flow simulation, the scheduling cycle is processed. The formula for calculating the scheduling cycle is as follows:

[0146]

[0147]

[0148]

[0149] Where T represents the scheduling time; d i S1 represents the distance of the vehicle from the center of the intersection; S2 represents the displacement distance at the vehicle's maximum speed; v represents the displacement distance at the vehicle's minimum speed. max Indicates the maximum speed of the vehicle; v min This represents the minimum speed at which a vehicle travels. Through the above processing, the vehicle scheduling model can accurately simulate vehicle travel and traffic flow based on the vehicle's unique identifier and travel route, combined with the simulation of the Poisson distribution of vehicle arrivals and the calculation of the scheduling cycle.

[0150] In step S5, the model is invoked to output the action evaluation for the next action of each vehicle, specifically including: retrieving information such as vehicle waiting time, number of waiting vehicles, and number of passing vehicles at a certain time step. Environmental sensors and traffic flow sensors are installed in the four entrance lanes (east, west, south, and north) to read vehicle occupancy, traffic flow information, CO2 emissions, and fuel emissions. Vehicle occupancy is the ratio of the number of vehicles in the intersection to the available lane capacity within a given time period. CO2 emissions and fuel emissions reflect the environmental impact of vehicles. Average waiting time reflects traffic flow. A lower average waiting time means that vehicle queues at the intersection are relatively shorter, indicating higher traffic flow. Then, the vehicle action decision model implemented in step S3 is used to schedule vehicles at every scheduling time step, with convoys passing according to priority. The time interval between high-priority convoys and low-priority vehicles at the intersection is set to resolve conflicts and reduce parking waiting time.

[0151] In step S5, after a vehicle enters the priority queue disbandment area, the lead vehicle initiates a disbandment request, including vehicle information about the queue, such as the queue identifier and the number of vehicles. Upon receiving the disbandment request, the intelligent vehicle dispatching system instructs vehicles to leave the queue one by one according to their order. The system schedules vehicles to leave the disbandment area at regular intervals based on their position and priority within the queue to ensure traffic flow and safety. Finally, once all vehicles in the queue have left the disbandment area, the intelligent vehicle dispatching system deletes the vehicle information from the queue.

[0152] This invention provides a vehicle scheduling method and system for a two-lane, unsignaled intersection. By drawing a directed graph of conflict relationships and using a depth-first search tree, vehicle scheduling at the unsignaled intersection is achieved, thereby improving vehicle traffic efficiency and reducing the probability of accidents through optimized scheduling strategies. Simulation results demonstrate the feasibility of this scheduling model. Therefore, this invention has practical applications and economic benefits, and is suitable for urban traffic management and intelligent transportation systems.

[0153] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0154] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A vehicle dispatching method for a two-way, two-lane, unsignaled intersection, characterized in that, include: Step S1: Based on the given real two-way two-lane intersection traffic situation, preprocess the vehicles to determine whether a collision will occur between the vehicles and analyze the conflict relationship between the vehicles. Step S2: Construct a vehicle scheduling model based on deep search of conflict relationships; Step S3: Use nonlinear programming to make vehicle action decisions, and find the optimal passage time and vehicle acceleration; based on the vehicle scheduling model, perform lower-level planning and design a vehicle scheduling strategy for unsignalized intersections. Step S4: Build a simulation scenario of the intersection based on the given real intersection scenario, and initialize the vehicle information of each vehicle; Step S5: Input the vehicle information into the vehicle scheduling model, make behavioral decisions, and obtain the action output of each vehicle at the next moment; In step S3, nonlinear programming is used to make vehicle action decisions, specifically including the following steps: Step S301: Calculate the optimal acceleration for each vehicle using nonlinear programming; Step S302: Calculate the time and distance each vehicle needs to accelerate from its initial speed to its maximum speed; to solve for the optimal time for vehicles to pass through an intersection under the same depth matrix, based on the intersection constraints and the displacement-time formula under uniform acceleration: Where d represents the distance of the vehicle from the intersection. Let represent the initial velocity of the vehicle, 'a' represent the acceleration to be calculated, and 't' represent the vehicle's travel time. Step S303: Check if the vehicle can reach the intersection before reaching the maximum speed: If the distance traveled by the vehicle is greater than or equal to the initial distance, directly use the quadratic equation solution formula to calculate the time it takes for the vehicle to reach the intersection; if the distance traveled by the vehicle is less than the target distance, calculate the remaining distance traveled at the maximum speed, and calculate the time required to travel the remaining distance at the maximum speed. Step S304: Call the depth matrix to compare the travel times of vehicles in the same priority queue and solve for the maximum time; Step S305: Using the time as the target, calculate the optimal acceleration value for each vehicle in the same queue as it passes through the intersection using the quadratic equation solution formula.

2. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 1, characterized in that, In step S1, vehicle preprocessing includes setting vehicle constraints and designing a queue sorted by the distance of vehicles from the center point of the intersection. In a two-way, two-lane scenario model, the current vehicle conflict relationships at the intersection include: Intersecting relationships: the conflict between straight-ahead traffic in the east direction and left-turn traffic in the north direction; diverging relationships: the conflict between straight-ahead traffic in the east direction and left-turn traffic in the east direction; converging relationships: the conflict between straight-ahead traffic in the west direction and left-turn traffic in the south direction; the remaining conflict points in the east, west, and north directions are obtained by rotating the above conflict routes by 90°, 180°, and 270° respectively based on the center of the intersection. The conflict set is generated by the union of intersection, divergence, and convergence relationships; In a two-way, two-lane road scenario model, the vehicle constraints include: a minimum speed of 0 km / h, a maximum speed of 30 km / h, and a minimum acceleration of -1.5 m / s². 2 Maximum acceleration 1.0 m / s² 2 The minimum distance between vehicles is 15m.

3. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 1, characterized in that, In step S2, the vehicle scheduling model based on deep search of conflict relationships specifically includes: first, using matrix theory to construct the conflict relationships between vehicles; second, using the deep search tree algorithm to determine the order in which vehicles enter the intersection, so that vehicles in the queue do not have conflict relationships and travel within a safe range in terms of speed and spacing.

4. The vehicle dispatching method for a two-way, two-lane unsignalized intersection according to claim 3, characterized in that, In step S2, the matrix structure is as follows: ; In this matrix, the number of vehicles in the queue at the intersection is n, and each element is a. ij , when a ij =1 indicates that vehicle j has a conflict relationship with vehicle i, when a ij =0 indicates that there is no conflict between vehicle j and vehicle i; Since the conflict relationship is directed, the adjacency matrix is ​​not a symmetric matrix; where i, j = 0, 1, 2, ..., n, when i, j = 0, the car is a virtual lead car that only serves as a guide, and the rest are follow cars.

5. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 3, characterized in that, The specific steps of step S2 are as follows: Step S201: Sort the vehicles within the intersection area in order of their distance from the center point of the intersection, with the vehicles closer to the center point at the front; based on the sorting result, form an ordered vehicle queue, and number the vehicles according to their position in the sorting to form a queue model from vehicle number 1 to vehicle number N; set a virtual lead vehicle with the number 0 to represent the center point of the intersection. Step S202: When vehicle Ci enters the intersection control area, determine whether the planning of the vehicles ahead is completed. If not, proceed to step S203; if completed, determine the conflict relationship of the vehicles ahead of vehicle Ci and select the vehicle with the largest number as the leading vehicle of Ci. Step S203: Starting from the virtual lead vehicle, select an unvisited vehicle as the root node. Based on the elements in the conflict relationship matrix of the current node, determine the vehicles that have conflict relationships with the root node vehicle as child nodes through the depth-first search theory. Form a priority intersection passage queue with vehicles at the same depth that do not have conflict relationships. The closer the vehicle queue is to the root node, the higher its priority. Step S204: For each optional next vehicle / child node, check the conflict relationship matrix and the conflict situation of the selected vehicles under the same priority queue. If there is no conflict, add the vehicle to the vehicle driving queue at the same depth, set the vehicle with the largest number in the queue as the root node, and execute step S203. Step S205: During the access process, if a root node vehicle has no child nodes, then set other vehicles in the same priority queue as root nodes and execute step S203. Step S206: During the access process, if a vehicle of a certain root node has no child nodes and there are no other vehicles in the same priority queue, the access will move to the previous priority queue until all vehicles are grouped together. Step S207: The priority queue obtained by the depth-first search algorithm is the order in which vehicles enter the intersection. The higher priority queue enters the intersection first, and vehicles of the same priority must ensure that they pass through the center point of the intersection at the same time.

6. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 1, characterized in that, The vehicle information in the simulation scenario of step S4 includes vehicle speed, acceleration, expected driving route, and location information. The simulation scenario also includes lane geographic information, specifically the approach lanes, exit lanes, and conflict behavior relationships in each direction; the approach lanes are divided into the intersection priority queue control area and the intersection center area; the exit lanes include the priority queue disbanding area.

7. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 1, characterized in that, Step S5 includes: Define a unique identifier for each vehicle and specify its travel route, including the roads and intersections it passes through; simulate vehicle arrival patterns by setting vehicle traffic flow to a Poisson distribution; and process the scheduling cycle to achieve vehicle scheduling and traffic flow simulation. The formula for calculating the scheduling cycle is as follows: Where T represents the scheduling time; Indicates the distance of the vehicle from the center of the intersection; This indicates the distance the vehicle can travel at its maximum speed. This indicates the distance displaced at the vehicle's minimum speed. Indicates the maximum speed of the vehicle; This represents the minimum speed at which a vehicle travels. Through the above processing, the vehicle scheduling model accurately simulates vehicle movement and traffic flow based on the vehicle's unique identifier and route, combined with the simulation of the Poisson distribution of vehicle arrivals and the calculation of the scheduling cycle.

8. The vehicle dispatching method for a two-way, two-lane, unsignaled intersection according to claim 1, characterized in that, In step S5, the actions of each vehicle at the next moment are output, specifically including: calling relevant information such as vehicle waiting time, number of waiting vehicles, and number of passing vehicles at a certain time step; installing environmental sensors and traffic flow sensors in the four entrance lanes of the east, west, south, and north to read vehicle occupancy, traffic flow information, CO2 emissions, and fuel emissions; and implementing the vehicle action decision model in interface step S3 to schedule vehicles at every scheduling time step, allowing the convoy to pass according to priority, setting the time interval between high-priority convoys and low-priority vehicles at intersections to resolve conflicts and reduce parking waiting time.

9. A vehicle dispatching system for a two-way, two-lane, unsignaled intersection, based on the vehicle dispatching method for a two-way, two-lane, unsignaled intersection as described in any one of claims 1-8, characterized in that, The system includes an intelligent vehicle dispatching system and a vehicle speed and acceleration calculation and control unit, and has V2X communication capabilities for all vehicles in the queue. The intelligent vehicle scheduling system calculates and controls the vehicles by the control unit, as well as the formation and movement control of the formation based on the depth-first search tree. The vehicle speed and acceleration calculation and control unit acquires vehicle queue information and performs real-time scheduling planning for vehicles based on basic vehicle information and queue priority.

Images