Intelligent Control And Simulation Research On Urban Traffic Flow

In this paper, we have done some research in intelligent control and simulation of urban traffic flow. To intelligent control, we use the theory of fuzzy control, because fuzzy control is a manner of intelligent control based on rules which needs not to build precise mathematical model, and can control traffic quickly in real time. In order to describe the congested status of roads effectively, we define the congest parameter and take it as the input of our fuzzy controller. Congest parameter is the unity of the two parameters, car density and average velocity. It can reflect the degree of congestion on roadway truly, and avoid the unilateralism of using car density or average velocity singly. The use of congest parameter cuts the number of input, and lows the complexity of design, and accelerates the response to the traffic congestion. In the paper, Two projects of fuzzy control are put forward, that is variable phase order project and const phase order project, and analyze the advantage and disadvantage of the two projects. When the cross be controlled by the former project, the phase order of the cross will vary with time elapse, however, when be controlled by the latter, the phase order will be const, only the time of phases changes with the congest degree.In order to analyze and validate the fuzzy control projects the paper puts forward. We build a simulation model using the theory of Cellular Automata, then design and realize the ECTS (ECNU Traffic Simulation) traffic flow simulation system. The system has the following performance, (1)intuitional: Users can see the moving status of cars on the road and in the cross, (2)applicability: The system can get the data of car flow from the system itself, and also can get data from real-life. It can adjust the parameter value automatically based on the analysis of intelligent control module,(3)controllable: Users can change the parameter value of lane and each cross, such as the length of lane, the cycle time of cross, the number of phase of cross. These performance make user watch and analyze the status of traffic easily, and can also help user work out the effective decision, ensure the traffic flow fluently.Finally, we analyze the performance of the ECTS, the result of analysis indicate that the ECTS work s steadily and can reflect the real traffic more precisely. At the same time, we also validate the fuzzy control project using the ECTS. The data of the experiment indicate, the two projects can control traffic flow effectively.