The Design Of Intersection Signal Light Control System

With the continuous development of society and economy,people’s living standard has been higher and higher,and the number of private cars has been also increasing dramatically.The state leads to traffic network pressure has been more and more stressed,especially in the morning and evening rush hour.Traffic congestion not only causes lower efficiency of road network and contaminates more serious environment and increases the probability of traffic accident,but also endangers people’s life and property security and reduces travelling happiness.Although measures such as viaduct,underground pedestrian passageway,bus lane,motor vehicle restriction is taken have relieved road stress to some extent,However,These methods fail to solve the problem fundamentally.Moreover,Most current signal light control system uses switches that fixed time and preset order which cannot adjust with actual traffic information at the intersection automatically at home and abroad.Therefore,It is possible to cause problem that no vehicles go through the intersection when traffic light is green in one direction but long queues and waiting time in the other direction,which wastes road resources.To solve the problem,The dissertation takes an intersection signal light control system of LANZHOU as an example which uses video image detector to collect actual information on road to count vehicle flow by license plate recognition and calculate vehicle queue length by coordinate conversion.Inputting the two variables to fuzzy controller and fuzzy RBF neural network controller to optimize traffic light green time of next signal cycle.Comparing fixed time controller with fuzzy controller and fuzzy RBF neural network controller of same vehicle flow,drawing conclusion that average vehicle delay of fuzzy RBF neural network controller is less.The intelligent traffic control system takes Cyclone IV EP4CE10 FPGA as control chip,adopts modular design idea to design hardware such as camera,digital tube and Ethernet communication and software such as image acquisition module,image processing module,intelligent traffic control module and Ethernet communication module,Verilog Hardware Description Language to Program each controlling module and verify software feasibility.In the end of the dissertation,designs intersection signal light control system function interface on the Visual Basic 6.0 platform,and takes hardware lapping and program debugging to realize physical lap verification.The purpose of the dissertation is to use fuzzy RBF neural network controlling algorithm to optimize intersection signal light controlling parameters by collecting the actual information of vehicles on the road.Intelligent traffic control not only can alleviate congestion of urban traffic roads greatly and reduce vehicle delays but also improve the pass capacity of intersections and the efficiency of traffic management,which has great significance for intelligent management of cities.