Research On Collaborative Route Selecting Method For Operation Vehicles In Inland Port

A large number of freight vehicles participate in the process of collecting and distributing in inland port and if these vehicles are not properly guided,it is extremely prone to congestion,which leads to the increase of vehicle driving time and energy consumption in the port,the reduction of resource utilization,and impacts port throughput.How to plan a reasonable driving route for the operation vehicles in the port and improve the utilization efficiency of the road network is a momentous problem.Therefore,the research on the routing method for operation vehicles in inland ports is of great theoretical value and practical significance.According to the mentioned problems,the research content of this thesis is as follows:(1)A candidate paths searching method based on traffic flow short-term prediction was designed to obtain candidate paths that conform to the characteristics of operation vehicles in inland ports.Firstly,based on the prediction process of the K-Nearest Neighbor(KNN)algorithm,the factors affecting the traffic flow in the port was defined as the state vector,and the short-time traffic flow in single time step and multi-time step was predicted.Then,the optimal path searching algorithm,an improvement on the basis of A* algorithm,for operation vehicles was presented.In the path searching process,multiple indicators was used to evaluate the path,and the road segment impedance was calculated based on traffic flow,and the waiting time at the key nodes and operation flow of the operation vehicle was considered.Finally,based on the optimal path,the K Shortest Path(KSP)algorithm was used to search for the candidate paths of the operation vehicle.(2)A collaborative route selecting method was designed to alleviate the routing conflict caused by a large number of vehicles routing at the same time in the inland port.Firstly,the routing behavior of the vehicles that request routing at the same time was modeled as a game with incomplete information,and the notion of satisfaction equilibrium was applied to analyze the game.The collaborative route selecting model based on the game theory was established and the candidate paths was the strategy set of the vehicles participating in the game.Then,the collaborative route selecting algorithm(CoRS)was proposed to solve the model,and the convergence of the algorithm was proved theoretically.The execution process of the algorithm corresponded to the multiple stages of the game.In the initial stage of the game,each vehicle selected the path with the greatest utility from the candidate paths.In the subsequent game stage,the vehicles was grouped,and all groups performed adaptive learning according to the historical routing strategies.The game was repeated until the vehicle routing was balanced,and the road network utilization efficiency was maximized.(3)The methods proposed in this thesis were experimentally verified.In the experiments of the candidate paths searching method based on traffic flow short-term prediction,firstly,the nearest neighbor number of the short-time traffic flow multi-time step prediction algorithm based on KNN was selected as 8,and the accuracy of the prediction algorithm was verified.Then the effectiveness of the improved A* algorithm for the optimal path searching and the KSP algorithm for the optimal paths searching in the actual road network showed that the above algorithms can correctly obtain the corresponding results conforming to the characteristics of the operation vehicle.In the experiments of the collaborative routing method,the number of iterations required for the algorithm to reach the convergence was firstly identified.Compared with Dijkstra algorithm and self-adaptive learning algorithm,the CoRS can reduce the average driving time of the vehicles and the energy consumption of the road network when the number of vehicles was large.In summary,the work in this thesis is a preliminary attempt on the intelligent construction of inland ports.It has practical significance and practical value for reducing congestion in inland ports and improving the utilization efficiency of the road network.