The Study Of Intelligent Traffic Signal Based On Fuzzy Neural Network

Because most of the transports in the city are running in transportation trunk line,urban transport network smoothly or not has a very direct relationship with the main line of traffic conditions.So urban trunk road traffic signal coordination control is a effective measure to ease traffic congestion,improve the main road capacity and the entire capacity of urban traffic network.In this paper,based on fuzzy control and neural network theory,we propose urban intersection traffic signal intelligent control method.This paper provides a feasible control methods and strategies for the construction and development of Chinese urban intelligent transportation system.Firstly,we design a single intersection embedded software system.This system involves the design of the overall framework of the single intersection software system,the construction of embedded application software running environment(such as the transplantation of the Linux system and hardware drivers to write,etc.),the preparation of the signal embedded application.This single-point signal machine provides the foundation work for the intersection of intelligent control.Secondly,in order to solve the current traffic situation in the urban main roads,the paper uses the fuzzy control method based on neural network to adjust the phase difference between junctions when we design intersection traffic signal intelligent control method.Specifically,the design of the fuzzy controller uses two variables as the input,the one is the vehicle density between adjacent junctions,the other is the forthcoming number of vehicles through the intersection when the status of previous intersection is green phase.The output variable is the vehicle speed between the two junctions.Through the neural network algorithm,the weights of learning and the error correction makes the fuzzification,fuzzy inference and defuzzification process of the vehicle density and the number of vehicles can be achieved. So we can calculate the actual vehicle speed between the two junctions.Then,the phase difference between intersection can be obtained by this speed and the distance between junctions.Finally,this paper simulate and study intersection road fuzzy neural by the integrated use of MATLAB and VC6.0software.This method can shorten the delay time of the vehicles on the main road of the city,through the observation of the results and the use of the algorithm to the single-point signal,enabling traffic smoothly through the intersection.