The Study On Route Optimization Technology For Intelligent Transportation System Based On PCNN Algorithm

Traffic Flow Guidance System is one of the core research fields in Intelligent Transportation Systems. Based on the Dynamic Traffic Distribution Modal and software research, it provides drivers optimal route choices and real-time road condition, which combines GPS, Communication, Control Theory and Computer technology. Therefore, drivers can avoid the heavy traffic road, contributed to the least time costs to destination. All of the systems alleviate the traffic jam and exhaust emission condition, promoting a better environment with less air pollution. Optimal Route Choice Algorithm is one of the key ingredients of Dynamic Traffic Distribution Theory. Its major goals is to find the optimal route to destination from road net as soon as possible with lowest computing error, according to the road structure, road weight and OD value. One of the most important evaluations standard for it is the real-time quality and global optimization quality. The research on optimal Route Selection algorithms is booming, which mainly consists of Dijkstra in Map theory, GA and Neural Networks algorithms. The algorithms cannot meet the requirements above, for its intrinsic quality or reality difficulties.At the beginning of this paper, the structure, workflow, and development condition of global Traffic Flow Guidance System are introduced. Then the Route Optimization Choice algorithms classification and structure are summarized. After analyzed the limitation of Dijkstra algorithm and Genie algorithm, a Pulse Coupled Neural Networks (PCNN) algorithm is proposed. PCNN provides Group Similarity, Synchronous exciting and Auto-wave generation characters, and is widely researched and applied in Image Processing, Target Identification and Optimization fields. The workflow is constructed and software is developed with VC++ based on PCNN algorithm. The simulation results proved the algorithm is feasible, which is verified by simulation Road Weight value. At the end of this paper, the different results of PCNN algorithm and Dijkstra algorithm is compared, the reason of the difference is analyzed under their internal mechanism, then the conclusion of PCNN has more advantages is drawn.