A transport vehicle dynamic planning method, device, electronic device and storage medium

By using traffic network diagrams and real-time traffic data in path planning, local optimization and lane selection are solved, and the problem of existing technology being unable to deal with traffic congestion in a timely manner is improved, and transportation efficiency and system operation efficiency are improved.

CN118411826BActive Publication Date: 2024-11-15SHAANXI SIWEI SHUBANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410663985.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-27
Publication Date
2024-11-15
Estimated Expiration
2044-05-27

AI Technical Summary

Technical Problem

Existing path planning technologies are unable to respond to sudden traffic jams in a timely manner, resulting in increased costs of transport vehicles and reduced transportation efficiency.

Method used

By obtaining the traffic network graph, establishing a weighted directed graph, determining key nodes, generating global road planning, and using real-time traffic data to judge blocked nodes, performing local optimization to update the lane selection of vehicles, using long-term and short-term memory networks and ant colony algorithms Optimize lane selection and adjust driving strategies to avoid blockages.

Benefits of technology

It improves the flexibility of path planning and transportation efficiency, reduces delays caused by traffic congestion, and improves the operation efficiency of the traffic system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118411826B_ABST
    Figure CN118411826B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device, electronic device and storage medium for dynamic planning of transport vehicles, and relates to the technical field of path planning; a traffic network diagram of a target area is obtained, and a weighted directed graph is established according to the traffic network diagram; multiple path plans are generated through multiple path planning algorithms, and key nodes are determined according to repeated nodes of multiple path plans; the key nodes are marked in the weighted directed graph and a global road plan is generated; real-time traffic flow data from the target node to the next node is obtained, and it is determined whether the next node is a congested node; if there is a congested node, local optimization is performed to determine the lane selection of the target vehicle, and the completed route is updated to make the target vehicle return to the global road plan. A weighted directed graph is constructed, and a global path plan is generated using key nodes, potential traffic congestion points are predicted, and driving strategies are adjusted in a timely manner, so as to improve the flexibility of planning and reduce delays caused by traffic congestion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of path planning, and in particular relates to a method, device, electronic equipment and storage medium for dynamic planning of a transport vehicle. Background Art

[0002] In modern society, transport vehicles, with their core position, connect all links of economic development. They not only serve as key nodes in the logistics chain to ensure the smooth flow of global trade and supply chains, but also as a link between urban and rural areas, promoting the optimal allocation of resources and the coordinated development of the economy. More importantly, transport vehicles play an irreplaceable role in daily life. They carry people's food, clothing, housing and transportation, ensuring social stability and improving people's livelihood. With the development of society, urban roads are becoming more and more congested, so road planning is needed to reduce the time cost during transportation.

[0003] Existing technologies use multi-dimensional features to capture road traffic status by real-time analysis and prediction of dynamic loads in large-scale transportation networks. Adaptive multi-feature fusion technology is used to analyze the data of the traffic road network, which can map and generate a balanced distribution map that takes into account the dynamic traffic status. This technical method not only improves the diversity of path selection for evacuation vehicles in traffic congestion, but also uses a data-driven approach to allocate potential path requirements, enhances the ability of the transportation system to quickly resume normal operations, and completes path planning.

[0004] The above method solves some problems, but there are still some problems. This technical method cannot cope with sudden congestion when planning the route, and cannot make predictions or give timely response measures, resulting in increased transport vehicle costs and reduced transport efficiency. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that sudden congestion cannot be faced during route planning, and timely predictions and countermeasures cannot be given, resulting in increased transport vehicle costs and reduced transport efficiency, and to propose a transport vehicle dynamic planning method, device, electronic device and storage medium.

[0006] In a first aspect of the present invention, a transport vehicle dynamic programming method is first proposed, the method comprising:

[0007] Obtaining a traffic network diagram of a target area, and establishing a weighted directed graph according to the traffic network diagram;

[0008] Generate multiple path plans through multiple path planning algorithms, and determine key nodes based on repeated nodes of multiple path plans;

[0009] Marking the key nodes in the weighted directed graph and generating a global road plan;

[0010] Obtain real-time traffic flow data from the target node to the next node and determine whether the next node is a congested node;

[0011] If there is a blocked node, local optimization is performed to determine the lane selection of the target vehicle, and the completed route is updated to make the target vehicle return to the global road plan.

[0012] Optionally, the key node includes a first key node and a second key node; and determining the key node according to repeated nodes of multiple paths includes:

[0013] Extract the selection times of each node in multiple path planning to generate a node selection times table, and take the node whose selection times of any node in multiple path planning algorithms exceeds the selection threshold as the first key node;

[0014] If there is a discontinuous connection between any first key nodes, the traffic efficiency of any node between any first key nodes is calculated, the traffic efficiency threshold is dynamically adjusted according to the traffic efficiency between any nodes, and the node with a traffic efficiency higher than the traffic efficiency threshold is determined as the second key node.

[0015] Optionally, obtaining real-time traffic flow data from the target node to the next node and determining whether the next node is a congested node includes:

[0016] Real-time acquisition of real-time traffic flow data from the target node to the next node; the real-time traffic flow data includes inflow traffic flow data and stranded traffic flow data of the next node;

[0017] Extract historical inflow traffic data to calculate an average speed, match the speed of the target vehicle in the inflow traffic data with the average speed, and if the speed of the target vehicle within a preset time is lower than the average speed, determine that the target vehicle is a slow vehicle;

[0018] It is determined whether the next node of the target node is a blocked node according to the penalty function, and if the penalty score is greater than the penalty threshold, the next node is determined to be a blocked node.

[0019] Optionally, if there is a congested node, local optimization is performed to determine the lane selection of the target vehicle including:

[0020] Inputting historical traffic flow data into a long short-term memory network model for training to obtain target model parameters, updating model parameters of the long short-term memory network model according to the target model parameters to obtain a prediction model, and inputting real-time traffic flow data of any next node into the prediction model to obtain a prediction result;

[0021] The target node is taken as the starting node, and the node with the shortest distance from the blocked node to the global road planning is taken as the ending node, and it is determined whether the ending node is blocked. If the ending node is blocked, the key node adjacent to the ending node in the global road planning is taken as the ending node;

[0022] Initialize the prediction result, and traverse the nodes in the area that meet the preset conditions through a simulated annealing algorithm to generate a node table; the preset condition is that the radius between the starting node and the ending node is a restricted area;

[0023] Update the pheromone of the ant colony algorithm, iteratively select the node table according to the pheromone through the ant colony algorithm, until the maximum number of iterations is met, the loop is terminated and the optimal result is output, and the lane selection is determined according to the optimal result; the pheromone is a constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants, and the constraint range includes the node area smaller than the preset condition and the node area larger than the blocked node.

[0024] In a second aspect of the present invention, a transport vehicle dynamic planning device is provided, comprising: the device comprises: a directed graph establishment module, a key node determination module, a global planning generation module, a blocked node judgment module and a lane selection module:

[0025] The directed graph establishment module is used to obtain a traffic network diagram of a target area and establish a weighted directed graph according to the traffic network diagram;

[0026] The key node determination module is used to generate multiple path plans through multiple path planning algorithms, and determine the key nodes according to the repeated nodes of the multiple path plans;

[0027] The global planning generation module is used to mark the key nodes in the weighted directed graph and generate a global road plan;

[0028] The congested node judgment module is used to obtain real-time traffic flow data from the target node to the next node and judge whether the next node is a congested node;

[0029] The lane selection module is used to perform local optimization to determine the lane selection of the target vehicle if there is a congested node, and update the completed route to make the target vehicle return to the global road plan.

[0030] Optionally, the key node determination module includes: a first node determination module and a second node determination module:

[0031] The first node determination module is used to extract the selection times of each node in the multiple path planning to generate a node selection times table, and take the node whose number of selections of any node in the multiple path planning algorithms exceeds the selection threshold as the first key node;

[0032] The second node determination module is used to calculate the traffic efficiency of any node between any first key nodes if there is a discontinuous connection between any first key nodes, dynamically adjust the traffic efficiency threshold according to the traffic efficiency between any nodes, and determine the node with a traffic efficiency higher than the traffic efficiency threshold as the second key node.

[0033] Optionally, the congestion node judgment module includes: a vehicle flow acquisition module, a vehicle speed judgment module and a node judgment module:

[0034] The vehicle flow acquisition module is used to acquire real-time vehicle flow data from the target node to the next node; the real-time vehicle flow data includes the inflow vehicle flow data and the stranded vehicle flow data of the next node;

[0035] The vehicle speed judgment module is used to extract historical inflow vehicle flow data to calculate an average speed, match the target vehicle speed in the inflow vehicle flow data with the average speed, and if the speed of the target vehicle within a preset time is lower than the average speed, the target vehicle is judged to be a slow vehicle;

[0036] The node judgment module is used to judge whether the next node of the target node is a blocked node according to a penalty function, and if the penalty score is greater than a penalty threshold, the next node is judged to be a blocked node.

[0037] Optionally, the lane selection module includes: a traffic flow prediction module, an end node determination module, a node traversal module and a lane determination module:

[0038] The vehicle flow prediction module is used to input historical vehicle flow data into the long short-term memory network model for training to obtain target model parameters, update the model parameters of the long short-term memory network model according to the target model parameters to obtain a prediction model, and input the real-time vehicle flow data of any next node into the prediction model to obtain a prediction result;

[0039] The end node determination module is used to take the target node as the starting node, take the node with the shortest distance from the blocked node to the global road planning as the end node, judge whether the end node is blocked, and if the end node is blocked, take the key node adjacent to the end node in the global road planning as the end node;

[0040] The node traversal module is used to initialize the prediction result and generate a node table by traversing the nodes in the area that meet the preset conditions through a simulated annealing algorithm; the preset condition is that the radius between the starting node and the ending node is a restricted area;

[0041] The lane determination module is used to update the pheromone of the ant colony algorithm, and iteratively select the node table according to the pheromone through the ant colony algorithm until the maximum number of iterations is met, the loop is terminated and the optimal result is output, and the lane selection is determined according to the optimal result; the pheromone is a constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants, and the constraint range includes the node area smaller than the preset condition and the node area larger than the blocked node.

[0042] In a third aspect of the present invention, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0043] Memory, used to store computer programs;

[0044] The processor is used to implement any of the above-mentioned steps of a method for dynamic planning of a transport vehicle when executing a program stored in the memory.

[0045] In a fourth aspect of the implementation of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any of the steps of a method for dynamic planning of a transport vehicle described above is implemented.

[0046] Beneficial effects of the present invention:

[0047] The present invention proposes a method for dynamic planning of transport vehicles, which obtains a traffic network diagram of a target area, establishes a weighted directed graph according to the traffic network diagram; generates multiple path plans through multiple path planning algorithms, and determines key nodes according to repeated nodes of multiple path plans; marks the key nodes in the weighted directed graph and generates a global road plan; obtains real-time traffic flow data from the target node to the next node, and determines whether the next node is a congested node; if there is a congested node, performs local optimization to determine the lane selection of the target vehicle, and updates the completed route to make the target vehicle return to the global road plan. Construct a weighted directed graph, use key nodes to generate a global path plan, predict potential traffic congestion points, adjust driving strategies in a timely manner, improve planning flexibility, and reduce delays caused by traffic congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below in conjunction with the accompanying drawings.

[0049] Figure 1 A flow chart of a method for dynamic planning of a transport vehicle is provided for an embodiment of the present invention;

[0050] Figure 2 A flow chart of another method for dynamic planning of transport vehicles is provided for an embodiment of the present invention;

[0051] Figure 3 A structural diagram of an electronic device is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the present invention, the description of "first", "second", etc. is only used for descriptive purposes, and cannot be understood as indicating or implying its relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0053] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0054] The embodiment of the present invention provides a method for dynamic planning of transport vehicles. Figure 1 , Figure 1 A flow chart of a transport vehicle dynamic planning method provided by an embodiment of the present invention. The method comprises the following steps:

[0055] S101, obtaining a traffic network diagram of a target area, and establishing a weighted directed graph according to the traffic network diagram.

[0056] S102, generating multiple path plans through multiple path planning algorithms, and determining key nodes according to repeated nodes of the multiple path plans.

[0057] S103, marking key nodes in a weighted directed graph and generating a global road plan.

[0058] S104, obtaining real-time traffic flow data from the target node to the next node, and determining whether the next node is a congested node.

[0059] S105: If there is a congested node, local optimization is performed to determine the lane selection of the target vehicle, and the completed route is updated to make the target vehicle return to the global road plan.

[0060] A dynamic planning method for transport vehicles provided in an embodiment of the present invention constructs a weighted directed graph, uses key nodes to generate global path planning, predicts potential traffic jams, adjusts driving strategies in a timely manner, improves planning flexibility, and reduces delays caused by traffic jams.

[0061] In one implementation, historical traffic data in the target area is obtained, pre-processed, and outliers are screened out to ensure data accuracy. A variety of path planning algorithms are used, such as Dijkstra algorithm, A* algorithm, Floyd-Warshall algorithm, etc., including but not limited to the above algorithms; the route between the starting point and the destination is planned, and the most intersections are selected as key nodes to ensure subsequent node positioning, forming a complete global road plan.

[0062] In one implementation, by acquiring the traffic flow data from the target node to the next node in real time and determining whether the next node is a congested node, a key prerequisite can be laid for subsequent local road optimization work. The acquisition and analysis of such real-time data allows the traffic management system to quickly identify potential or existing traffic congestion problems, so that local optimization can be carried out in a targeted manner. Guiding the target vehicle to a less congested route not only improves the pertinence and effectiveness of the optimization measures, but also helps to improve the overall driving efficiency of the target vehicle.

[0063] In one implementation, when a congested node appears in the global road planning, the starting node is determined by updating the completed route, that is, the target node is used as the starting node, and the next node of the congested node in the global road planning, that is, the adjacent node of the congested node, is used as the end node. The traffic efficiency around the congested node is predicted by the local optimization algorithm, and the local path planning is generated according to the starting node and the end node, and the lane selection of the target vehicle is determined before the target node to avoid the congested node. The target vehicle can return to the global road planning and continue to ensure an efficient driving path. The advantage of this method is that it combines local flexibility and global coherence, which not only solves the immediate traffic problem, but also maintains the long-term optimality of the driving route. In this way, the operating efficiency of the entire transportation system can be significantly improved and resource waste can be reduced.

[0064] In one embodiment, step S102 includes:

[0065] The selection times of each node in various path planning are extracted to generate a node selection times table, and the node whose selection times in various path planning algorithms exceeds the selection threshold is taken as the first key node.

[0066] If there is a discontinuous connection between any first key nodes, the traffic efficiency of any node between any first key nodes is calculated, the traffic efficiency threshold is dynamically adjusted according to the traffic efficiency between any nodes, and the node with a traffic efficiency higher than the traffic efficiency threshold is determined as the second key node.

[0067] In one implementation, path planning is performed using multiple algorithms. Comprehensive consideration of multiple path planning algorithms for each node and statistics of the number of selections for each node help to deeply analyze the node flow in the traffic network; accurately grasp the operating rules of the traffic network, and thus develop a more reasonable and efficient global path planning.

[0068] In one implementation, adjacent key nodes are determined to be continuous. When there is a discontinuous connection between the key nodes, the traffic efficiency of any node between the key nodes is calculated. The traffic efficiency is calculated by the number of vehicles passing through the intersection within the green light time.

[0069] In one implementation, the traffic efficiency between nodes in the region is updated in real time, and the node with the highest traffic efficiency is selected as the candidate node. If all candidate nodes cannot form a continuous connection, the traffic efficiency threshold is lowered and the candidate node is reselected; if there is at least one candidate node that can form a continuous connection, but the threshold is relatively low, resulting in too many candidate nodes, the threshold can be fine-tuned to find the optimal candidate node; the selected candidate node is used as the second node. By dynamically adjusting the traffic efficiency threshold, it is ensured that there is a node connection between adjacent nodes, and the selection of the necessary route is determined by the setting of key nodes, ensuring that the entire route of the global road planning is the optimal continuous route, which can significantly improve the operating efficiency of the entire transportation network and reduce congestion and delays.

[0070] In one embodiment, step S104 includes:

[0071] Real-time traffic flow data from the target node to the next node is obtained in real time; the real-time traffic flow data includes the inflow traffic flow data and the stranded traffic flow data of the next node.

[0072] The historical inflow traffic data is extracted to calculate the average speed, and the speed of the target vehicle in the inflow traffic data is matched with the average speed. If the speed of the target vehicle is lower than the average speed within a preset time, the target vehicle is determined to be a slow vehicle.

[0073] The penalty function is used to determine whether the next node of the target node is a blocked node. If the penalty score is greater than the penalty threshold, the next node is determined to be a blocked node.

[0074] In one implementation, in global road planning, the lane type of the target vehicle at the next node is the original lane, the incoming traffic volume is the traffic volume flowing into the next node, and the penalty function is:

[0075]

[0076] Among them, F is the penalty score, A is the stranded traffic flow of the original lane, B is the inflow traffic flow, e is a natural constant, α is the preset lane selection probability, the lane types include left-turn lanes, right-turn lanes and through lanes, m is the number of original lanes, n is the outflow traffic flow of the original lane, is the proportion of slow vehicles in B, that is, the proportion of inflowing traffic.

[0077] In one implementation, a target vehicle traveling on a global road plan needs to make a congestion judgment on a key node in advance; for example, before entering a target node, the target vehicle needs to make a congestion judgment on the next key node, which is conducive to the target vehicle making plans in advance and reducing unnecessary losses caused by special circumstances.

[0078] In one implementation, by acquiring the traffic flow data from the target node to the next node in real time, including the inflow and stranded traffic flow, the traffic flow can be accurately monitored. The average speed calculated using historical data can effectively identify slow vehicles and provide decision conditions for the current lane selection.

[0079] In one embodiment, step S105 includes:

[0080] The historical traffic flow data is input into the long short-term memory network model for training to obtain the target model parameters, the model parameters of the long short-term memory network model are updated according to the target model parameters to obtain the prediction model, and the real-time traffic flow data of any next node is input into the prediction model to obtain the prediction result.

[0081] The target node is taken as the starting node, and the node with the shortest distance from the blocked node to the global road planning is taken as the ending node. It is judged whether the ending node is blocked. If the ending node is blocked, the adjacent key node of the ending node in the global road planning is taken as the ending node.

[0082] The prediction results are initialized, and the nodes in the area that meet the preset conditions are traversed through the simulated annealing algorithm to generate a node table; the preset condition is that the radius between the starting node and the end node is a restricted area.

[0083] Update the pheromone of the ant colony algorithm, and iteratively select the node table according to the pheromone through the ant colony algorithm until the maximum number of iterations is met, end the loop and output the optimal result, and determine the lane selection based on the optimal result; pheromone is the constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants. The constraint range includes the node area that is smaller than the preset condition and larger than the node area of ​​the blocked node.

[0084] In one implementation, the training data is input into the long short-term memory network model, and the output and cell state of each time step are calculated through the input gate, forget gate and output gate of the long short-term memory network model. The value of the loss function is calculated according to the output of the model and the true label. The gradient of the loss function with respect to the model parameters is calculated by the back propagation algorithm of time. The weights and biases of the model are updated according to the gradient, and the weights and biases are the model parameters. Repeat the above process of forward propagation, loss calculation, back propagation and parameter update until a certain stop condition is met, such as: loss convergence, reaching the maximum number of iterations, etc.; use the gradient descent method to update the model parameters. Update the weights and biases of the model according to the calculated gradient and learning rate.

[0085] In one implementation, the global road planning is connected by continuous key nodes; in the global road planning, the next node of the blocked node, that is, the adjacent node of the blocked node is used as the end node. When the end node is also blocked, the system can automatically find the adjacent key node in the global planning as a replacement. For example: when the end node is also blocked, the end node is updated to the second blocked node, and the next key node of the second blocked node in the global road planning is used as the end node; ensuring that the route planning is always in the optimal state, further improving the intelligence and practicality of the route planning.

[0086] In one implementation, the restricted area is a closed area with an edge trend that is approximately circular, not a standard circle. Its edge trend is limited by the connection of edge nodes, that is, edge nodes should be continuously connected. If there is no continuous connection, it will shrink inward until the nodes are continuously connected. Therefore, there may be irregular depressions on the edge of the restricted area; wherein, the restricted area is a restricted area with the blocked node as the center and the radius as the distance between the starting node and the ending node; and the blocked node is used as a restricted area to keep ants in the ant algorithm away from the area; for example, ants will search in an area that is approximately donut-shaped.

[0087] In one implementation, the simulated annealing algorithm and the ant colony algorithm are combined to intelligently traverse and select nodes within a preset area to determine the optimal lane selection. This method not only takes into account real-time traffic data, but also optimizes the lane selection process through algorithms, allowing target vehicles to make more reasonable driving decisions in complex traffic networks. The updating of pheromones and the application of constraints further improve the efficiency and accuracy of the search.

[0088] In one implementation, in the ant colony algorithm, pheromones are updated so that ants can exchange information with each other when transmitting pheromones, thereby constraining the direction of the ants and obtaining the best result as quickly as possible within the range of activity.

[0089] Based on the same inventive concept, the embodiment of the present invention also provides a transport vehicle dynamic planning device. Figure 2 , Figure 2 A schematic diagram of the structure of a transport vehicle dynamic planning device provided by an embodiment of the present invention includes: a directed graph establishment module, a key node determination module, a global planning generation module, a blocked node judgment module and a lane selection module:

[0090] The directed graph building module is used to obtain the traffic network diagram of the target area and build a weighted directed graph based on the traffic network diagram.

[0091] The key node determination module is used to generate multiple path plans through multiple path planning algorithms and determine the key nodes according to the repeated nodes of the multiple path plans.

[0092] The global planning generation module is used to mark key nodes in a weighted directed graph and generate a global road plan.

[0093] The congestion node judgment module is used to obtain the real-time traffic flow data from the target node to the next node and judge whether the next node is a congested node.

[0094] The lane selection module is used to perform local optimization to determine the lane selection of the target vehicle if there is a congested node, and update the completed route to make the target vehicle return to the global road plan.

[0095] A transport vehicle dynamic planning device provided based on an embodiment of the present invention constructs a weighted directed graph, uses key nodes to generate global path planning, predicts potential traffic jams, adjusts driving strategies in a timely manner, improves planning flexibility, and reduces delays caused by traffic jams.

[0096] In one embodiment, the key node determination module includes: a first node determination module and a second node determination module:

[0097] The first node determination module is used to extract the selection times of each node in multiple path planning to generate a node selection times table, and take the node whose selection times in multiple path planning algorithms exceeds the selection threshold as the first key node.

[0098] The second node determination module is used to calculate the traffic efficiency of any node between any first key nodes if there is a discontinuous connection between any first key nodes, dynamically adjust the traffic efficiency threshold according to the traffic efficiency between any nodes, and determine the node with a traffic efficiency higher than the traffic efficiency threshold as the second key node.

[0099] In one embodiment, the congestion node judgment module includes: a vehicle flow acquisition module, a vehicle speed judgment module and a node judgment module:

[0100] The vehicle flow acquisition module is used to acquire the real-time vehicle flow data from the target node to the next node; the real-time vehicle flow data includes the inflow vehicle flow data and the stranded vehicle flow data of the next node.

[0101] The vehicle speed judgment module is used to extract the historical inflow traffic data to calculate the average speed, match the target vehicle speed in the inflow traffic data with the average speed, and if the speed of the target vehicle within a preset time is lower than the average speed, the target vehicle is judged to be a slow vehicle.

[0102] The node judgment module is used to judge whether the next node of the target node is a blocked node according to the penalty function. If the penalty score is greater than the penalty threshold, the next node is judged to be a blocked node.

[0103] In one embodiment, the lane selection module includes: a traffic flow prediction module, an end node determination module, a node traversal module and a lane determination module:

[0104] The traffic flow prediction module is used to input historical traffic flow data into the long short-term memory network model for training to obtain target model parameters, update the model parameters of the long short-term memory network model according to the target model parameters to obtain a prediction model, and input the real-time traffic flow data of any next node into the prediction model to obtain a prediction result.

[0105] The end node determination module is used to take the target node as the starting node, and the node with the shortest distance from the blocked node to the global road planning as the end node, and judge whether the end node is blocked. If the end node is blocked, the adjacent key node of the end node in the global road planning is taken as the end node.

[0106] The node traversal module is used to initialize the prediction results and generate a node table by traversing the nodes in the area that meet the preset conditions through the simulated annealing algorithm; the preset condition is that the radius between the starting node and the end node is a restricted area.

[0107] The lane determination module is used to update the pheromone of the ant colony algorithm. The ant colony algorithm iteratively selects the node table according to the pheromone until the maximum number of iterations is met, the loop ends and the optimal result is output, and the lane selection is determined based on the optimal result. The pheromone is a constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants. The constraint range includes the node area that is smaller than the preset condition and the node area that is larger than the blocked node.

[0108] The embodiment of the present invention further provides an electronic device, such as Figure 3As shown, it includes a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304, and the memory 303 is used to store computer programs; when the processor 301 is used to execute the program stored in the memory 303, the following steps are implemented:

[0109] Obtaining a traffic network diagram of a target area, and establishing a weighted directed graph according to the traffic network diagram;

[0110] Generate multiple path plans through multiple path planning algorithms, and determine key nodes based on repeated nodes of multiple path plans;

[0111] Marking the key nodes in the weighted directed graph and generating a global road plan;

[0112] Obtain real-time traffic flow data from the target node to the next node and determine whether the next node is a congested node;

[0113] If there is a blocked node, local optimization is performed to determine the lane selection of the target vehicle, and the completed route is updated to make the target vehicle return to the global road plan.

[0114] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0115] The communication interface is used for communication between the above terminal and other devices.

[0116] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0117] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0118] In another embodiment provided by the present invention, a computer-readable storage medium is also provided, in which a computer program is stored. When the computer program is executed by a processor, a dynamic planning method for a transport vehicle in any one of the above embodiments is implemented.

[0119] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A transport vehicle dynamic planning method, characterized in that: The method comprises: Obtaining a traffic network diagram of a target area, and establishing a weighted directed graph according to the traffic network diagram; Generate multiple path plans through multiple path planning algorithms, and determine key nodes based on repeated nodes of multiple path plans; Marking the key nodes in the weighted directed graph and generating a global road plan; Obtain real-time traffic flow data from the target node to the next node and determine whether the next node is a congested node; If there is a blocked node, local optimization is performed to determine the lane selection of the target vehicle, and the completed route is updated to make the target vehicle return to the global road plan; If there is a congested node, local optimization is performed to determine the lane selection of the target vehicle, including: Inputting historical traffic flow data into a long short-term memory network model for training to obtain target model parameters, updating model parameters of the long short-term memory network model according to the target model parameters to obtain a prediction model, and inputting real-time traffic flow data of any next node into the prediction model to obtain a prediction result; The target node is taken as the starting node, and the node with the shortest distance from the blocked node to the global road planning is taken as the ending node, and it is determined whether the ending node is blocked. If the ending node is blocked, the key node adjacent to the ending node in the global road planning is taken as the ending node; Initialize the prediction result, and traverse the nodes in the area that meet the preset conditions through a simulated annealing algorithm to generate a node table; the preset condition is that the radius between the starting node and the ending node is a restricted area; Update the pheromone of the ant colony algorithm, iteratively select the node table according to the pheromone through the ant colony algorithm, until the maximum number of iterations is met, the loop is terminated and the optimal result is output, and the lane selection is determined according to the optimal result; the pheromone is a constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants, and the constraint range includes the node area smaller than the preset condition and the node area larger than the blocked node.

2. A transport vehicle dynamic planning method according to claim 1, characterized in that: The key nodes include a first key node and a second key node; Determining key nodes based on repeated nodes of multiple paths includes: Extract the selection times of each node in multiple path planning to generate a node selection times table, and take the node whose selection times of any node in multiple path planning algorithms exceeds the selection threshold as the first key node; If there is a discontinuous connection between any first key nodes, the traffic efficiency of any node between any first key nodes is calculated, the traffic efficiency threshold is dynamically adjusted according to the traffic efficiency between any nodes, and the node with a traffic efficiency higher than the traffic efficiency threshold is determined as the second key node.

3. A transport vehicle dynamic planning method according to claim 1, characterized in that: Obtaining real-time traffic flow data from the target node to the next node and determining whether the next node is a congested node includes: Real-time acquisition of real-time traffic flow data from the target node to the next node; the real-time traffic flow data includes inflow traffic flow data and stranded traffic flow data of the next node; Extract historical inflow traffic data to calculate an average speed, match the speed of the target vehicle in the inflow traffic data with the average speed, and if the speed of the target vehicle within a preset time is lower than the average speed, determine that the target vehicle is a slow vehicle; It is determined whether the next node of the target node is a blocked node according to the penalty function, and if the penalty score is greater than the penalty threshold, the next node is determined to be a blocked node.

4. A transport vehicle dynamic planning device, characterized in that: The device comprises: a directed graph building module, a key node determination module, a global planning generation module, a blocked node judgment module and a lane selection module: The directed graph establishment module is used to obtain a traffic network diagram of a target area and establish a weighted directed graph according to the traffic network diagram; The key node determination module is used to generate multiple path plans through multiple path planning algorithms, and determine the key nodes according to the repeated nodes of the multiple path plans; The global planning generation module is used to mark the key nodes in the weighted directed graph and generate a global road plan; The congestion node judgment module is used to obtain real-time traffic flow data from the target node to the next node and judge whether the next node is a congestion node; The lane selection module is used to perform local optimization to determine the lane selection of the target vehicle if there is a congested node, and update the completed route to make the target vehicle return to the global road plan; The lane selection module includes: a traffic flow prediction module, an end node determination module, a node traversal module and a lane determination module: The vehicle flow prediction module is used to input historical vehicle flow data into the long short-term memory network model for training to obtain target model parameters, update the model parameters of the long short-term memory network model according to the target model parameters to obtain a prediction model, and input the real-time vehicle flow data of any next node into the prediction model to obtain a prediction result; The end node determination module is used to take the target node as the starting node, take the node with the shortest distance from the blocked node to the global road planning as the end node, judge whether the end node is blocked, and if the end node is blocked, take the key node adjacent to the end node in the global road planning as the end node; The node traversal module is used to initialize the prediction result and generate a node table by traversing the nodes in the area that meet the preset conditions through a simulated annealing algorithm; the preset condition is that the radius between the starting node and the ending node is a restricted area; The lane determination module is used to update the pheromone of the ant colony algorithm, and iteratively select the node table according to the pheromone through the ant colony algorithm until the maximum number of iterations is met, the loop is terminated and the optimal result is output, and the lane selection is determined according to the optimal result; the pheromone is a constraint condition of the ant colony algorithm, which is used to constrain the activity area of ​​ants, and the constraint range includes the node area smaller than the preset condition and the node area larger than the blocked node.

5. A transport vehicle dynamic planning device according to claim 4, characterized in that: The key node determination module includes: a first node determination module and a second node determination module: The first node determination module is used to extract the selection times of each node in the multiple path planning to generate a node selection times table, and take the node whose number of selections of any node in the multiple path planning algorithms exceeds the selection threshold as the first key node; The second node determination module is used to calculate the traffic efficiency of any node between any first key nodes if there is a discontinuous connection between any first key nodes, dynamically adjust the traffic efficiency threshold according to the traffic efficiency between any nodes, and determine the node with a traffic efficiency higher than the traffic efficiency threshold as the second key node.

6. A transport vehicle dynamic planning device according to claim 4, characterized in that: The congestion node judgment module includes: a vehicle flow acquisition module, a vehicle speed judgment module and a node judgment module: The vehicle flow acquisition module is used to acquire real-time vehicle flow data from the target node to the next node; the real-time vehicle flow data includes the inflow vehicle flow data and the stranded vehicle flow data of the next node; The vehicle speed judgment module is used to extract historical inflow vehicle flow data to calculate an average speed, match the target vehicle speed in the inflow vehicle flow data with the average speed, and if the speed of the target vehicle within a preset time is lower than the average speed, the target vehicle is judged to be a slow vehicle; The node judgment module is used to judge whether the next node of the target node is a blocked node according to a penalty function, and if the penalty score is greater than a penalty threshold, the next node is judged to be a blocked node.

7. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 3 when executing a program stored in a memory.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in any one of claims 1 to 3 are implemented.