A Campus Traffic Congestion Detecting Method Based on BP Neural Network

This paper presents a novel method for detecting the campus traffic congestion by combining BP neural network with campus traffic congestion descriptor. In this method, road occupancy rate is proposed and proved to be the most effective descriptor among other descriptors of traffic congestion level in campus. The campus traffic congestion levels are divided into three phases based on three-phase traffic theory. Experimental results show that the proposed method is capable of detecting campus traffic congestion.