Study On Adjacent Traffic Intersection Coordination Control Algorithm And Microscopic Simulation

With the development of economy and rise in level of urbanization, the traffic problems become more and more serious. In traffic engineering field, more importance has been attached to how to make the best use of existing road, increase traffic capacity as well as reduce delay in signalized junctions as much as possible by means of scientific reasonable measures. City intersection is the joint of road network, which plays an important role in road capacity and service. Taking two adjacent traffic intersections as the subject investigated, the paper adopts technologies of fuzzy control and neural network in research and microscopic simulation of adjacent traffic intersection coordination control algorithm. Firstly, on the basis of analyzing basic theory of traffic system, carry out the traffic flow forecast to the middle link according to the correlativity of two adjacent traffic intersections. And then, put forward the two adjacent traffic intersection multi-phase fuzzy coordination control algorithm, taking the decrease of vehicles' average delay at adjacent traffic intersections as the target. Meanwhile, in order to solve the problems that it is difficult to artificially set more membership grade functions in signal timing of two-stage fuzzy control, and to verify its rationality as well as the selection and optimization of fuzzy control rules, etc., the paper employs the method of fuzzy neural network and simulates it with Matlab software, and the simulation result proves the good control effect. Whereafter, develop the software of intersection microscopic simulation that reproduces the traffic flow state at intersections by simulating the vehicles' running at intersections. It is intuitive to comprehend that the method of fuzzy neural network brought forward in the topic can make the traffic signal adjust accordingly in company with the change of traffic conditions, which relieves the traffic congestion. The simulation result indicates the effect from the algorithm is fairly better than that from traditional offset coordination control and the algorithm enhances the traffic efficiency at intersections.