Dynamic Route Optimization Based On Improved Ant Colony Algorithm

Dynamic route guidance is an important part of the in-car navigation and intelligent transport systems, especially in today’s era of telematics its significance is very valuable. With the development of society, road traffic situation is more complex, sometimes navigation and traffic systems do not get the real-time traffic information.In the coming era of car networking, to forecast traffic flow and travel time is the basis of dynamic path optimization.Combines the advantages of ant colony algorithm can greatly improve the efficiency of dynamic path optimization.First,In this paper introduce several core issues of dynamic route guidance briefly, including:dynamic network model, real-time traffic information for dynamic road network model is given, and proposed a way to reduce storage space and can express road turn restrictions;dynamic traffic access to information,gives a brief traffic flow forecasting methods based less real-time traffic information; travel time prediction, dynamic route guidance is based on the travel time of road.To predict the travel time based on the traffic information and traffic flow theory.Then,According to the ant colony algorithm deficiency,this paper presents an adaptive ant colony optimization algorithm with direction factor. Mainly in three aspects of the improvements, including join orientation factor, fast convergence to the optimal solution;improved pheromone left, legacy factor of pheromone dynamically updated based on the difference of pheromone between the maximum and minimum; join the global optimal incentives, according to the local optimal path searching sizes,weights are updated and search times, avoiding trapped in local optima.