Research On Dynamic Traffic Routing Guidance Based On Genetic Algorithm

With the progress of science and technology, the concept of routing guidance system brings a new idea to solve the traffic problem. Urban traffic guidance is routing guidance system is applied in urban traffic. As the core of intelligent transport system, urban traffic guidance system is one of the most effective methods to solve the existing traffic problems.The prototype of the routing guidance is the shortest path in graph theory problems. The current commonly used solving methods include the classic Dijkstra algorithm and all kinds of bionic algorithms. Dijkstra algorithm can only show that the single-objective shortest path for the path problem, Ant colony optimization and Particle Swarm Optimization in the bionic algorithms is more difficult to give the several optimal solutions, so, these algorithms are all not suitable to solve the K pieces of the optimal paths for dynamic traffic routing guidance. Genetic algorithm is better to other algorithms in time and result, when be used to solved the K pieces of the optimal paths problem. So, in this paper, a genetic algorithm to solve the K pieces of the optimal paths based on the dynamic traffic targets, is used to achieve the dynamic traffic routing guidance.Specific algorithm is described as follow. As a target value excepting length in using of the multi-objective optimization ideas, the traffic generated real-time traffic jams data is simulate by membership function and a timer, based on the algorithm to solved the K pieces of the optimal paths problem. The target value is used to change the path evaluation function to adjusting routing guidance momentarily. An interactive interface between man and computer is displayed on the map. Users can choose real-time path in the interactive interface, according to the travel require.In this paper, numerical experiments are used to validate the genetic algorithm, which solves K pieces of the optimal paths based on the dynamic traffic, is rational. The results of the specific algorithm are shown in the map. Algorithm is simple in structure, fast in convergence. This algorithm is particularly suitable for solving several real-time optimal paths between any two points in large-scale transportation network.