Intelligent Transportation Systems Based On Artificial Neural Network Detection And Control

First, the paper illustrates the concept of intelligent transportation system. Intelligent transportation system is a synthetic research system of interdisciplinarity, information, systematization. Its main contents are to elaborate the technologies of artificial intelligence, automatic control, computer science, information and communication, electron transducer. These technologies are applied in traffic management system on the whole ground. Based on the characteristics of large scale and omni-direction, ITS is also an integrated transportation management system with real time, accurateness, high performance.Intelligent transportation system is one of the most focusing research fields in several years. Based on the background of intelligent transportation system, the paper discusses the development of ITS in America, Europe, Japan and other developed countries. The paper is to introduce present something in China. At the same time, some scientific terms on ITS are listed in the paper.Artificial neural network with specially nonlinear and adaptable information processing conquers disadvantages of traditionally artificial intelligence in intuition, such as pattern, speech recognition, non-structural information processing. Because of these, artificial neural network is successfully applied in neural expert system, pattern recognition, intelligent optimization, combinatorial optimization, prediction. Combined with other conventional ways, artificial neural network will promote continual progress in artificial intelligence and information processing. In recent years, artificial neural network is making further progress toward simulating human cognition. Combined with fuzzy system, genetic algorithm and evolution mechanism, artificial neural network forms computational intelligence and becomes an important aspect of artificialintelligence.In the paper, the basic principles of BP neural network, RBF neural network, genetic algorithm and fuzzy control are to be considered in detail. BP neural network, RBF neural network and genetic algorithm are applied in traffic volume forecasting of ITS. By means of recorded vehicle flow data of each crossing of Jingshi Road in Jinan, prediction models are feasibly constructed.Due to applicable extension of artificial neural network, it is a useful tool in solving nonlinear system, especially to complicated and ambiguous models. Constructed forecasting models depend on the theories of BP neural network and improved method, RBF neural network, genetic neural network. According to specific instances, using a strongly mathematical tool-MATLAB6.5 and programming, the paper expresses related simulation curves and forecasting data. Moreover, the data are analyzed and compared with initially recorded data. Aiming at the real situation of Jinan city, a simulation platform on ITS of Jinan city is established by using fuzzy theory and VC++6.0. The platform, including clear and understandable graphics, flexible parameters, is satisfying.By the constructed forecasting models and s imulation platform, dealing with offline prediction and online simulation, analyzing the gotten data, it is indicated that there is a significant function by using the models. In fact, it will provide a credible foundation for the application of these results in the future.