This paper proposes a real-time traffic signal intelligent control method with transit-priority. The objective of this control method is to reduce the delays of passengers and special vehicles. Transit-priority is divided into the special transit-priority which is an absolute priority and the normal transit-priority which is a relative priority. When the detectors in the red phase detect special vehicles arrival, the phase will become a special phase, the current green phase must be

more »

... and the special phase will be run. After the special vehicles pass through, the next running phase selection will be done using the phase selection method with normal-transit-priority, by this time, the phase with more urgency will be selected. It embodies transit-priority idea. The green increase time of current phase is inferred by a fuzzy controller of which the inputs are the vehicles number of current phase and next phase. Multi-layer neural network is used to realize this fuzzy controller. Compared with fixed-time control method and the fuzzy control method, simulation research shows that this method obtains a good performance in decreasing the delays of passengers and special vehicles.