An Intelligent Control Method for Urban Traffic Signal Based on Fuzzy Neural Network

The paper presents a traffic signal control method using a layer-structured fuzzy neural network (FNN) for learning rules of fuzzy logic control system. The FNN has advantages of both fuzzy expert system (fuzzy reasoning) and artificial neural network (self-study). The system is not needed to build the model of traffic flow for signal control approach at an intersection, it can be successfully trained to adapt different traffic flow and different conditions at the intersection based on the real-time data, this significantly reduces a lot of effort of extracting traffic expert's knowledge into fuzzy if-then rules. In order to get better dynamic response and reduce the computing capacity, the weights of FNN are optimized and the step length for self-study is modified based on fuzzy logic. Compared with traditional fuzzy control plan for traffic signal, the proposed FNN algorithm shows better performances and adaptability.

I. Introduction

Urban traffic has a great effect on the social development. Traffic signal is an essential element to manage the transportation network. Along with the movement of Intelligent Transportation System (ITS) in the United States, traffic signal control remains one of the most heavily funded research and development items. The research on traffic signal control is by no means complete.