Research On Signal Light Of Intersection Base On Cellular Automaton

With the development of social economy, traffic system plays a more and more im-portant role in the development of national economy and becomes one of the national keyconstruction projects. In order to prevent a series of social problems and environmental prob-lems caused by traffic jam from restricting urban economy, traffic flow, as the main researchobject of the traffic problems, draws the extensive attention of researchers. The current trafficflow consists of motor vehicle, non-motor vehicle and pedestrian in China. Such unique traf-fic structure mode leads to the rapid increase of conflict points in the traffic flow and the dete-rioration of traffic jam, and then results in increasing difficulty in traffic flow organizationalplanning. Therefore, the study of road traffic flow is of great theoretical significance.On one hand, this paper devotes to researching on road traffic flow modeling based oncellular automaton. By means of the theory of cellular automaton, this paper simulates andanalyzes the crossing effect and signal light control of the road traffic flow, and analyses theimpact of different crossing types and structures on road traffic flow. Meanwhile, that themodel built is suitable for the study of traffic flow dynamics is proved in this paper.On the other hand, to make traffic flow model more similar with real traffic environment,cellular automaton model is improved with the method of combining with fuzzy control, con-sidering running vehicle’s speed and gap etc as perceived objects, building fuzzy perceptionand bringing this fuzzy perception into vehicle random deceleration process. In the rule offuzzy control, the distance from front vehicle and back vehicle, vehicle’s speed and driver’sresponse time are taken as fuzzy controller’s input variables. Through the analysis, randomi-zation deceleration probability of current vehicle at some point can be obtained.In the end, on the basis of cellular automaton model based on fuzzy control, by analyzingthe relationship among road traffic flow, density and speed under different conditions, thispaper bring the neural network’s self-learning and self-adapting features and ability into fuzzycontrol, building a crossing traffic flow model which is more suitable for current road situa-tion. Based on single crossing simulation model, in order to realize the effect of signal lightlinkage at crossing, phase difference is controlled reasonably for adaptive timing. In allusionto the linkage control method of road traffic crossing, this paper brings in relevant evidence totestify the model.