Research On Control And Optimization Problems In Urban Intelligent Transportation System

Urban Intelligent Transportation System (UITS) is a kind of traffic solution which results from the integration of urban traffic control system, advanced computer technology, auto control technology, communication technology and artificial intelligence. Intelligent traffic control and intelligent traffic guidance are the essential parts of UITS. Research in these fields is very important considering its practical and theoretical value.This dissertation makes a comprehensive and deep research on urban traffic control methods, traffic route guidance system and advanced traffic management system, and presents some new methods after summarizing a lot of previous research reports in these fields.The main contents in this dissertation include:1. A kind of flexible hierarchical fuzzy controller for urban single traffic intersection is presented. This kind of fuzzy controller can decrease the number of fuzzy rules greatly, and is able to track the change of traffic condition around intersection to improve the efficiency of controller by modifying the fuzzy member function of controller through evolution algorithm.2. A kind of area traffic control method based on distributed Q-Leaming is presented. This method does not need an upper supervision level which is widely used in traditional area traffic control. It is a potentially instructive urban area traffic control method which needs no traffic flow model and only limited communication volume among different traffic intersection controllers.3. A kind of new intelligent collective optimization method, Bacterial Colony Chemotaxis (BCC) algorithm based on the description of bacterial colony chemotaxis, is presented. BCC algorithm takes advantage of both a single bacterium's reaction to chemoattractants and the exchange of position information among bacteria to find the optimum. BCC algorithm much improves the performance of BC algorithm while possessing the searching ability of a singlebacterium, is a kind of potentially powerful optimization method worth of much more research. This method can be used in dynamic route guidance system.4. A kind of model for urban traffic management system based on multi-agent technology is presented. This distributed management system can integrate traffic light control system and route guidance system effectively, and can also strengthen the real-time property and robustness of traffic control system.At the end of this dissertation, the author also gives some suggestion to the further research in these fields.