Freeway Congestion Automatic Detection And Vehicle Guidance Based On Single-Loop Detectors

Traffic congestion has been a most common problem among big and middle cities and has badly disturbed people's daily life. It's almost based on real-time artificial observation to judge whether the road is congested now. In this dissertation, our main purpose is to establish an automatic system which can automatically and accurately detect the road's congestion using the traffic data collected from single-loop detectors.Firstly analyzed the situation of traffic congestion at present; List various kinds of methods to detect traffic data and their advantages and disadvantages, then explained the reason why single-loop data was selected as our research object. The main factors causing congestion was listed here and then discussed the index to evaluate congestion. This dissertation used the traffic quantity,density and speed to evaluate congestion; Since single-loop detectors can only get the quantity and occupancy and can't obtain speed directly, so we discussed this problem about using quantity and occupancy to calculate speed. In this part, domestic and foreign scholars' researches on this problem were reviewed and then particularly discussed the speed estimation method based on dynamic and adaptive mean vehicle length; After the congestion was detected, it's necessary to inform drivers and give them a guidance, so a new plan was presented called single vehicle guidance based on vehicle license plate recognition technology. This technology was introduced, as well as the main scheme of the single vehicle guidance and the communication form to promulgate the information. Then developed the whole structure of the system including every modules' main function and how do they work; Finally, an example about traffic congestion detection and guidance using Matlab6.5 with the real data collected was given to demonstrate the validity and feasibility of the integrated method proposed.