Ant Colony Algorithm Basedonchaos And It’s Application

Ants is a humble creature in nature, although single ant has only a very limited intelligence, the ant colony has formed a highly organized society, this phenomenon attract many researcher’s attention. The original ant colony algorithm was born in Milan Polytechnic laboratory, Italy. Encouraged by its success to solve the NP problem, researchers from different country dedicated to the research on ant colony algorithm. Up to now, ant colony algorithm has been considerable development in theory, besides, application areas has extended from the traveling salesman problem and quadratic assignment problem.In this paper, we utilize the character of chaos and chaos optimization, proposed two kind of improved ant colony algorithm. And then, a further study on traveling salesman problem and robot path planning problem were undertaken. The main achievements of this paper are shown as follows.1. We proposed an improved MMAS, meanwhile, applied to traveling salesman problem. Firstly, improved algorithm use chaos optimization method based on Tent map, generate initial pheromones, and then exploit it to guide the early research process. When it comes to local optimum, we add the strategy of chaotic disturbance to enhance the capability of escaping from it. Finally, we used several standard TSP case for simulation to demonstrate the effectiveness of this method.2. We presented an improved ACS, and then applied to robot path planning problem. In the initial stage, we use second wave method generate initial pheromones to enhance search efficiency. In addition, we add pheromone diffusion model in pheromone update phase. What’s more, when algorithm stuck in local optimum, we introduce chaotic disturbance. Finally, simulation experiments illustrate the improved algorithm has better performance.3. We implemented a virtual reality project about Tanzhesi, using chaos ant colony system developed path finding module and auto roaming module.