Application Of Improved Ant Colony Algorithm In Path Selection Of Mobile Agent

The traveling agent problem is a classic,representative and complex combinatorial optimization problem ,which in Path Selection of Mobile Agent Migration. Ant colony algorithm is a new evolutionary algorithm and extremely suit to solve the travelling agent problem,which has the characteristic of parallelism, positive feedback and heuristic search, but it has the limitation of stagnation,and is easy to fall into local optimums。Based on duty weight and by using the number of iterative algorithms to update the rules and information-volatile factor to enhance the ability of choosing the path, there were two new methods were proposed, which were for the combination of the existing Ant colony algorithm and Mobile Agent characteristics in this paper.The main work of this paper can be concluded as follows:1. First of all, introduced the basic principles of ant colony algorithm and its application on the TSP, gave the status of research and applications. And then introduced the architecture of Mobile Agent system and its key technologies,applications and advantages .2. In view of the problem of ant colony algorithm applied in Mobile Agent migration and combine with the character of Mobile Agent, there were two improved ant colony algorithms were proposed in this paper. The first method is to use the idea of weight task to update ant path information and information-volatile factor. According to its own weights, a mobile agent carrying the corresponding task,can adaptively increase or reduce the incremental pheromone and pheromone evaporation coefficient in different stages. The experiment result shows that the shortcomings of ant colony algorithm can be adaptively overcome and better to improve the ability to solve the problem.The other new method is to use the number of iterative algorithms to update the rules and information-volatile factor, the algorithm will show different defects in different process, so according to the number of iterative algorithm, the improved ant colony algorithm can use the corresponding strategies to improve the algorithm of global search ability and prevent the emergence of " Early lag " phenomenon. The experiments show that the algorithm in this paper can solve the problem with more advantages.3. At last, the Travel Agent problem (TAP) was solved by the two improved ant colony algorithms, carried out a large number of simulation experiments and carried out in-depth analysis and comparison on the results of experiments. The experimental results show that the two improved ant colony algorithms had improved the ability of optimization than the basic ant colony algorithm.