Design Of Intelligent Traffic Signal System Based On Embedded

With the rapid development of China’s economy in recent years,the process of urbanization is getting faster and faster.The ownership of motor vehicles is increasing day by day,which brings people a convenient way of life,but also causes a series of problems such as intersection congestion and exhaust pollution.More and more experts and scholars pay attention to this phenomenon,they are committed to easing the congestion at the intersection and improving the traffic efficiency.At present,China is making great efforts to build a smart city,one of the most important links is smart transportation,which needs to give full play to the urban traffic efficiency and improve the urban traffic congestion.In order to reduce congestion and improve the number of vehicles passing in a unit time,this paper designs a set of intelligent traffic signal system which can adjust the traffic time according to the traffic flow based on the theory of iterative learning control,combined with computer video image processing technology and embedded system application.First of all,because of the nonlinear dynamic characteristics of urban traffic flow,it is difficult to establish an accurate control model.At the same time,the traffic flow in the same time and in the same place every day has the characteristics of repeatability and obvious periodicity.Therefore,this paper uses iterative learning control algorithm to optimize the time period of traffic lights and the effective green time of each phase,so as to make the vehicle queuing length difference and system error tend to zero,and maximize the traffic efficiency.Secondly,the current traffic intersection monitoring system is particularly perfect,which can easily obtain the implementation monitoring video of the road intersection,without adding additional traffic flow detection equipment.Through image processing of the obtained monitoring video,the traffic flow data can be obtained.In this paper,the video image is preprocessed first,then the background difference method is used to extract the lane background and moving target,the improved virtual coil method is used,and opencv software is used to process the intersection monitoring video to obtain more accurate traffic statistics data.Then,according to the field survey of the intersection,the overall scheme design of the intelligent traffic signal system is carried out.Because the embedded system has the characteristics of strong portability,convenient customization and high cost performance.At the same time,based on the design requirements of intelligent traffic signal system,after comprehensive consideration,this paper analyzes the arm architecture system,combined with the characteristics of intelligent traffic signal system,determines ARM-V7 architecture as the instruction set of the chip,and finally selects S5PV210 chip based on Cortex-A8 as the core.Finally,the software platform of intelligent traffic light system is built and debugged.Under the operating system of Linux,the whole development environment takes S5PV210 as the core of the whole system,and builds software systems such as cross compilation tool chain migration,TFTP and root file system migration,bootloader migration and Linux kernel migration.Then,the intersection composed of Shenjian East Road and Guangxian road in Shanghai is photographed on the spot by mobile phone to obtain the monitoring video,and the algorithm is simulated by MATLAB software to obtain the best cycle time and the best green light time of each phase,to verify the effectiveness of the proposed iterative control algorithm for intelligent traffic lights,and to conduct relevant debugging,to verify the algorithm in the system,and to find the adjustment The later time can effectively reduce congestion and improve traffic efficiency.