FCM Algorithm for Identifying Urban Road Traffic Condition with Loop Sensor Data

An improved fuzzy c-means algorithm (FCM) for detecting urban road traffic conditions based on loop sensor data is presented. Two characteristic indices, occupancy rate and average occupancy rate per vehicle, are extracted from sensor data and FCM algorithm is designed to identify the traffic condition. In addition, the principle of determining the fuzziness index for valid cluster is presented in this paper. Taking one single intersection for instance, the presented algorithm is demonstrated by combining an external program with VISSIM emulation software. Results show that the algorithm can improve the identified effects of the urban road traffic conditions in real-time, the identified result is better when the detecting interval is synchronized with the signal cycle time.