Multi-Phase Fuzzy Neural Control with Transit-Priority for Signalized Intersection

This paper proposes a transit-priority signal control method for urban intersection using fuzzy neural techniques. In order to reduce the delay of buses and passengers, a changeable-phase-order method is proposed. The phase with more passengers is preferential to be selected as the next phase by the end of current phase. The green increase time of current phase is inferred by a fuzzy controller which contains two inputs: the vehicles number of current phase and next phase. The fuzzy controller is implemented by a multi-layer neural network. Compared with the traditional fuzzy control method and fixed-time control method, Simulation investigations demonstrate the efficiency of the proposed approach.

I. Introduction

In the urban transportation system, a large portion of traffic congestion and delay occur at the intersections. Traffic signal control plays an important role in mitigating traffic congestion. Fuzzy logic, which models control based on human expert experience and knowledge, has been successfully applied to many automatic control tasks. The application of fuzzy logic in the time-variant traffic control system is firstly started by Pappis[1] in 1977, and many researchers [2]–[6] use fuzzy control technology for the traffic control system.