Urban Expressway Traffic Incident Detection Algorithm Based On Multi-Souce Traffic Data

Along with the rapid economic development and the acceleration of urbanization in China, urban expressway is developing rapidly in many cities. As a basis of main road network, the urban expressway plays a vital role in the urban road network. Under the accident condition, it's easy to result in the occurrence of secondary accidents, and the urban expressway congests will be heavier if the accidents can't be settled in time due to the high-speed and heavy-flow traffic.At the same time, AID is one of the key functions of traffic management and control system. How to detect and confirm the time, location and type of the incident accurately and timely are the basis to improve the reaction ability of traffic emergency management. The performance of AID algorithm is the core of traffic incident management, and also an important evaluation index of successful operation in intelligent transportation system. The AID algorithm based on only fixed detectors data or floating car data always performs unsatisfactorily because of excessive layout spacing, frequent communication failure and poor floating cars number. If both fixed detectors data and floating car data are used to detect traffic incidents, the detection results will be better, and when fixed detectors data or floating car data is missing, the AID algorithm can detect traffic incidents still.Based on above researches, an easily applicative AID algorithm base on floating car data and fixed detector data was proposed to improve the coverage of traffic incident detection. First, the algorithms based on fixed detector data and floating car data proposed separately integrated the change of traffic parameters in the temporal dimension and spatial dimension. Then the two algorithms were integrated effectively according to the D-S theory.Finally, the algorithm was verified using detector rate and false alarm rate with the traffic incident data, fixed detector data and floating car data collected from Beijing expressway, and the algorithm performed satisfactorily and can meet the needs for practical applications. The algorithm was an attempt based on multi-source data and could provide a reference for further study.