Design and Implementation of an Intelligent Combined Method for Urban Traffic Control at Intersections

Using intelligent methods is one of the efficient methods of scheduling the traffic lights at intersections to control traffic urban. In this paper used an clustering and genetic algorithms to control urban traffic at intersections based on multi-agent system. Proposed system get traffic load of each intersection. Next, uses clustering algorithm to find adjacent intersections. Then, it uses evolutionary computing algorithms to scheduling the traffic lights and also it has used exchange of messages between different agents to control possible fluctuations.The main capabilities and advantages of proposed method for simulating a multi-agent system for intelligent urban traffic control at the intersection are as follows: The number of intersection is considered infinite. Also, the distance between intersections is not limited and adjustable. Furthermore,this method is applicable to multi-ways. Due to use Multiagent technique and clustering algorithm for intersections, proposed method has been performed distributed processing which avoid excessive computational load processing in each cycle of traffic lights. For the design and implementation proposed method used Tropos methodology and JADE library. For evaluation, this proposed system tested with 1200 laboratory data in both low and heavy traffic areas. In comparison with fixed-time systems, the average run-time in sequential cycles of intersections in our combinational method is less than using the other single methods. In addition, for heavy traffic area and low traffic area, our system respectively has 18.5% and 30.8% (in average) improvement in delay time of vehicles, compared to the fixed-time methods.