Study On Methods And Utilizations In Automatically Selecting Path Based On GIS

Finding a shortest path between two given points over a surface while a certain condition prevails is a very significant researching field either in theory or in practice, and it can be appealed in many fields: highway alignments design, railway alignments design or pipe project, etc. But the decision has been involved in many constraints and it is controlled by the complexity. It is too difficult to build up unified decision model under those factors. If we introduce artificial intelligence into this field, it is hopeful to make rational decisions and to suggest reasonable projects.Optimizing highway alignment is a difficult combinational problem from highway engineering. The problem is to find an alignment incurring minimal total costs. From experience, it is an efficient and accurate way to calculate the total costs of the highway. The genetic algorithm (GA) is a good choice as a search algorithm. Its stochastic nature and global search characteristic enable the GA to find high quality solutions even for complicated problems. The solution for an accurate cost model may lie with the use of the Geographic Information System (GIS). GIS can spatially represent both the regions of the alignment. Furthermore, the spatial analysis capabilities of the GIS provide valuable inputs of the highway alignment optimization. This paper combines the capabilities of the GA and the GIS to optimize the highway alignments.After research, problem was reduced to a shortest path problem in math. In order to find the shortest path between two given points on integrative total cost surface, one genetic algorithm which could find the shortest path and another two GA ,one with slope constrain, another with collision-free, have been designed. The whole works have been summed up and the future researching directions have been discussed at the end of this paper.