A Fuzzy Control Strategy Of Traffic Signal Based On The Neural Network

Traffic Signal Control is the key is ensure the traffic safety and the smooth road. Study of Traffic Signal Automatic Control is very important to reduce the traffic delay, the accidents and pollution. this article takes guangzhou's tianhe northern junction for example, and introduce a fuzzy control strategy of traffic signal based on the Neural Network and Fuzzy Control Technic in order to decrease the vehicles average delay.View of the city intersection signal control issues, this paper analysis the advantange and disadvantange of the traffic control model and strategy, with the requirement of the Traffic Flow Forceasting and Control,introduce the design and implementation of such fuzzy control strategy, then elaborated on the designing thoughts, systems modeling, the function realization and algorithms improvement in details.Then we evaluate the effect of the fuzzy control strategy through the Simulation Experiment written in the VC++6.0 platform. In the field of Neuro Forecasting, Improving Neuro Network algorithm decrease about 50% running time and 50% computing iterations.In the field of Fuzzy Control, this strategy decrease about 30% average delay compared by the fix-time control strategy,and decrease about 8% average delay compared by the inductive control strategy, The result shows that the strategy is actually much more effective.Finally, we summarize the work of the paper, the existence of deficiencies and give several suggestions to future research.