Study Of Traffic Control System Based On DSPs

The severe condition caused by urban traffic crowding has influenced the development of social economy and people's daily life. Crossroads are the main distributing centre of traffic and the cause for the urban crowding. Research on the traffic control of the crossroads is the key point to solve the urban traffic problem. Considering the current condition of the traffic problems, the thesis studies the dynamic optimizing control algorithm at the crossroads, and discusses the signal harmonized control system of intelligent traffic. Then the traffic controller is designed and accomplished based on DSP.The main work of the thesis is as follows:Under the consideration of the traffic control problem, the dynamic optimized control method is proposed aiming at minimizing the number of vehicles which are delayed at the crossroads in a period. According to the classic distribution of traffic flow at the crossroads, the corresponding dynamic optimizing control model is established with the controlling variables of time span and green light ratio, and a traffic signal optimizing algorithm is presented based on genetic algorithm and chaos algorithm. First, the method adopts genetic algorithm to carry on global optimizing searching, during which the flaws of the pre-mature and low ability in local searching of genetic algorithm are solved by ergodic random property of chaos algorithm, with the purpose of improving searching efficiency and the quality of solutions. The adaptive punishment strategy that is used to process constraint conditions is designed to apply the combined optimizing algorithm. The strategy introduces constraint measurement and the extent of application to describe the extent to which the decision variables comply with inequality restriction and equality restriction. Meanwhile, the adaptive punishment function is constructed to apply optimizing algorithm, which transforms the optimizing problem of inequality and equality constraints to the problem containing the upper and lower limitations. The simulation has been taken to research the dynamic optimizing algorithm at single crossroad, and proved the validity of the optimizing method in this paper.