Research Of Dynamic Route Optimization Based On Primary And Secondary Population ACO

In recent years, China’s urbanization is developing rapidly. Urban road network is an important part of urban development, which carries with the development of the entire city. The development and efficient use of the transport network is one of the key issues of urban development. Intelligent Transport System is not only an effective method to improve the efficiency of road use and solve traffic congestion, but also the future direction of urban transport development. Dynamic route optimization is the core content of Dynamic Route Guidance System in Intelligent Transport System. Ant colony algorithm has a natural advantage in dealing with the dynamic route optimization problem with parallel computing, strong robustness, etc..First, the research status of intelligent transportation, dynamic path optimization and optimization algorithm are described in this article, and the basic ant colony algorithm and its improved algorithm is researched and analysed at the same time. Then the article gives an improved ant colony algorithm-ant colony algorithm based on the strategy of primary and secondary groups(PASPAS for short) which is conmbined with path crossover mutation strategy and removal of crossover strategy and have a good advantage in continue search ability and convergence.Secondly, the traveling salesman problem model is used to verify the effectiveness of the PASPAS algorithm. Three different characteristic and scale model is used for simulation, ant colony algorithm and max-min ant colony algorithm is used to be compared with PASPAS. Then this article simulates and discusses the new parameter of PASPAS algorithm, and find out appropriate combination of parameters. The experiments show that the algorithm has better search performance, faster convergence and high accuracy of the solution.Finally, according to the actual road network features, a simple dynamic road network simulation model is established in this article and the PASPAS algorithm and multi-threading technology is combined to make PASPAS algorithm with a real parallel computing capability to reduce the running time. Then, according to the characteristics of the road network,small window strategy and normalization process is used in PASPAS algorithm. After several simulation,the experiments show that the computing speed of the algorithm is further improved in the same solution quality.