Study of Modified Ant Colony Algorithm on Shortest Path Problem of Traffic System

Searching shortest path of the transportation network is one of the most important functions of ITS,in order to find the optimization path accurately and rapidly,disadvantages of using Dijkstra algorithm to deal with dynamic shortest path and a kind of modified ant colony algorithm are analysed and proposed,in which the mechanism of search direction and search hot section are introduced to improve searching performance.Through building modified ant colony algorithm model,using VC6.0 developing tool,for a example with actual traffic map,resolving the shortest path of the transportation network and contrasting the ant colony algorithm.Simulated experiments show that:average iterative times using the ant colony algorithm is 71.06,average iterative times using the modified ant colony algorithm is 55.82,the latter improved obviously.The modified ant colony algorithm can resolve the shortest path problem of traffic system of reference value and actual meaning.