Research On The Road Network Based Path Selection Strategy

As the rapid development of urbanization in our country, the number of cars is advancing rapidly. As a result, the traffic congestion problems are highlighted and can not be ignored. In order to solve these urban traffic congestion problems, carry out the energy conservation and environmental protection policies; one of the ideas is building an complete technical framework system of traffic guidance based on the existing urban traffic system.The traffic guidance technical framework system is an path planning system set up for the traffic road network model. It optimizes the travel route choices relying on the existing electronic information technology and algorithm theory. In this way, the intelligent traffic guidance system can save our travel time and travel cost, enhance the overall efficiency of the road traffic as well.There are two aspects that we mainly consider in the establishment of an traffic guidance technical framework system: a high stability model of practical transportation network and an efficient path selection strategy under the road network model. In terms of the construction of a road network model, mostly road segment intersection as the basic elements of the network model, and according to the relationship between nodes in the graph model to represent the actual road network path in the existing road network model studies. The disadvantage of this model is not able to get an intuitive relationship between the road segments, besides it needs some auxiliary information just like turn restrictions,direction restrictions etc. to express the relationship between paths. In order to better express this predecessor and successor relationships between adjacent road segments, this paper constructs RNMBRS network models. This RNMBRS network model corresponding storage structure using two opposite adjacent tables and the content of each node is the information of the road segment. all subsequent nodes after the head node are road segments with the head node segment has up and is up relations, this storage structure to some extent can improve the efficiency of path searching. In terms of the optimal path selection strategy, in the study of existing best path selection strategies its strategic thoughts mostly inherited from graph theory OD path ideas, it is difficult have more breakthroughs in the search space and time, in addition to the problem of repeat counting did not get a good solution under the global state. In order to narrow the search space and time complexity of the path selection strategy, this paper proposed the two-way path selection strategy of sharing information of global condition. To some extent better using shared information,avoiding repeat counting,and reducing the search space and time space in the path planning.In order to verify the stability of the RNMBRS network models and the feasibility of the path selection strategy, all experiments are conducted using the Visual Studio 2010 and MapXtreme as development tools, about the storage of the digital map, utilizing the storage solutions that road segment as the basic elements. The results verified the effectiveness of the two-way path selection strategy based on the sharing information of global condition proposed in this paper. Compare with the traditional path planning algorithm, the proposed path selection strategy is efficient and practical. The feasibility of the idea proposed in this thesis is confirmed based on the simulated experiment as well.