Research On Ant Colony Optimization And Its Applications

As the further studies of the essence of life,life sciences is develop at an unprecedented pace,Artificial intelligence research have begun to break away from the bondage of traditional Logical computation, and explore the new non-classical computation approach.The pioneer of Artificial intelligence Minsky had thought:"We should be enlightened from biology rather than physics."In background of this, ant colony algorithm emerged,in the year 1992, the Italian scholar M.Dorigo by studying the ant colony's foraging behavior, in his doctoral thesis propose a simulation evolutionary algorithm based on Colony——ant colony optimization (ACO).As the ant colony algorithm is steady, holistic, universal, distributed computing and many other advantages, attract a great attention of many scholars.compared with other more mature algorithm, ant colony algorithm has a more superior performance to solve some practical problems,In the past ten years,the development of ant colony algorithm is very fast.In combinatorial optimization, network routing, function optimization, data mining, robot path planning and even in the area of stock investment ant colony algorithm has been widely applied,and achieved good results.This paper focuses on the theory and application of ant colony algorithm, first, we elaborate on the biological mechanism of the algorithm, introduced the theory of basic ant colony algorithm, model, characteristics and the realization in TSP, and give a brief description about the application of ant colony algorithm in various fields.At the same time, pointed out some deficiencies about the basic ant colony algorithm, and then for the deficiencies we cited a number of typical improvement algorithms, such as ACS,MMAS,etc.;Then we through experiments and other scholars's research explain the parameter selection method in the basic ant colony algorithm models.Then this paper introduces the multi-robot system, and point out that the advantages of multi-robot systems and the necessity of the research.According to the theory of ACS,combination with multi-robot collaboration, this algorithm was applied to multi-robot path planning.Finally, we use the synthetical algorithm of ant colony algorithm and FCM algorithm, combination of edge detection techniques,and apply it to the colorized image segmentation, some typical images were used as experimental samples, we make a computer simulation, and the experiment achieved excellent results.It shows that this synthetical algorithm is an excellent approach for colorized image segmentation.