الگوریتم کلونی مورچه ها بر اساس برنامه ریزی مسیر برای مهاجرت عامل موبایل

Analysis of the travel agent questions (Travelling Agent Problem, TAP), and that it is a class of complex combinatorial optimization problems, mobile agent migration path planning is the most classic problems; Second, for ant colony algorithm to solve such problems in need long search time and ease into a local minimum shortcomings, the introduction of genetic algorithms and ant colony algorithm for global and local updating rules to improve, greatly reducing the travel agent ant colony algorithm to solve problems caused by the system into a local minimum stagnation phenomenon may be; Finally, through simulation experiments verify the validity of the proposed algorithm.

Research travel agent issue is important because the travel agent to solve the problem, the general sense can be given from the mobile agent moves between hosts in different migration path planning, making the mobile agent to concentrate the main effort priority access to those most likely to complete their task of the host to ensure that mobile agent Can be completed within the shortest possible time users of distributed computing tasks assigned to it, which can greatly improve the operating efficiency of mobile agent system. Application of the basic ant algorithm travel agent problems, can make ants on behalf of a mobile agent. Ants in addition to a higher probability of selection tend to spend a short time, high concentrations of pheromone path, we should also give priority to complete the task the probability is high, the short latency of the host, because ants have visited a number of tasks in a high probability , a short delay after the host is likely to have completed the task, you can return to the original host without direct access to the rest of the host. When the mobile agent on all hosts are unable to complete their task and eventually return to the initial host, the travel agent on the degradation of the famous traveling salesman problem (Travelling SalesmanProblem, TSP), so the traveling salesman problem is atravel agent issues special case. Travel agent is to promote the traveling salesman problem. Thus, the travel agent is a very complex combinatorial optimization problems, Moizumi in his doctoral thesis in the travel agent has been proved theoretically that the problem is NP-complete, its time and space complexity is very high, which requires to solve Travel agent problems generally must have the adap tive, self-learning, distributed, parallel intelligent features, which can be within an acceptable time frame arrive at the optimal solution or near optimal solution

Travel agent is a class of complex combinatorial optimization problem, aimed at addressing the mobile agent to move between different hosts how to plan the optimal migration routes. In the ant colony algorithm based on the introduction of genetic algorithms and ant colony algorithm for global and local updating rules to improve, greatly reducing the ant colony algorithm into a local minimum and cause the system to a standstill phenomenon possible. The simulation results show that the improved ant colony algorithm makes mobile agent can better efficiency and shorter time to complete the task, robustness of the algorithm to be strengthened