| As modern society is increasingly dependent on transport, the traffic control system has attracted more and more widespread attention. Based on the highly advanced technology such as the information, communication and control technology, the high-tech development of intelligent transportation systems can significantly improve the efficiency of the transport network. Therefore, the intelligent traffic system has become one of the most effective and economic way to solve the traffic congestion problems.There are generally two approaches to solve the contradiction between road and car: one is to control demand, another is to increase traffic infrastructure. But both have its limitations. Limiting the increase of vehicle cannot fundamentally solve problems because economic development will inevitably bring travel increase and automobile industry in our country is on the initial stage. Moreover, resource and environment problems are becoming outstanding at present, thus the measure of increasing infrastructure does not work. Now it is imperative to find other ways except of demand control and road facilities to solving the increasing demand of traffic. Therefore, the use of sophisticated intelligent traffic transportation (ITS) with information, communication and controlling technology can dramatically improve traffic network operation efficiency; make it become the economic and effective one way to solve the problem of heavy traffic.The paper proceeds as follows:Chapter 2 is an introduction of the traffic control theory. Then the Chapter 3 puts forwards a design scheme of the intelligent traffic signal controller and gives hardware system a modularization design. Then it gives a detailed description of hardware design, including high-performance ARM9 chip as the processor, large capacity Flash memory chips, the network interface, CPLD expansion I/O port.Chapter 4 goes on the software development environment. The construction of QNX-based embedded systems is relied on the hardware platform of the traffic signal controller and achieves the basic hardware device drivers and the main control program.Chapter 5 examines approximate dynamic programming (ADP) in the field of traffic signal control applications and develops the adaptive controller which can be online operated. The controller can provide dynamic control of real-time operation and respond to changing traffic flows with online learning skills, and frequently updats signal timing plans.The last chapter makes simulation to verify that ADP is a practical choice independent of the real-time signal intersection control. This paper meets the needs of the modern traffic control network and intelligence consistent with the development trend of traffic signal controller and provides a useful reference. |