FPGA-based Genetic Algorithm And Its Application In Traffic Lights

You can see traffic jam everywhere in a city. It causees not only economy losses, air pollution but also a bad impact on passenger’s health.Single traffic signal control method is often hard to adapt to complex traffic situations,which causes cars waited in vain.These problems reduce traffic capacity of the road as a result. So the intelligent traffic light is imperative.Intelligent traffic light is an important part of intelligent transportation system. It can effectively increase road capacity and improve the traffic situation.In this paper, make the result of the difference that vehicles arrived and vehicles left in each road phase as a recognition criterion, with the road saturation level, put a method that includes sensor control and adaptive control which is based on genetic algorithm.Sensor control also includes single point control and fixed-cycle control. The whole traffic light system is based on the SOPC that belongs to the Altera company’s Nios II embbed processor.Genetic algorithm is widely used in intelligent control, but most of the applications are operated by a computer with Von Neumann or Harvard architecture. Since these two architectures are the result of serial running, the running time of genetic algorithm is often not very good. When faced with a number of large-scale, complex problems, the calculation of the applications can not be achieved sometimes.Use the FPGA to achive the hardware implementation of the algorithm because of the advantages of parallel computing on the FPGA. FPGA-based genetic algorithm has a good improvement in running time.This can improve the performance of the system and increase the usefulness of genetic algorithm.The experiment results show that the design of the traffic light system can meet the desired design goals. FPGA-based genetic algorithm is turned out to be faster than the GA that bases on traditional computer, which means it is possible to running successfully for some applications which are complicated.