关键词: 记忆原理, 蚁群算法, 蚁群智能体, 记忆库, 路径选优

Abstract: To introduce classical ant colony algorithm into memory mechanisms，regards ant as the agent with memory，and by analyzing the principles of storage，update and forget，a model of ant colony agent based on biological memory principles is established.In the model，ant agents can constantly search and update its own and ant colony memory banks and adjust current optimal path immediately by virtue of memory.The combination of memory principles and classical ant colony algorithm have taken the place of latter’s optimal pattern of iterative repetition，which can realize path optima much better，shorten the searching time and make the algorithm more efficient.The application shows that the model can track the latest development in real time，adjust and judge the current optimal path immediately，thus guide it and other ant agents to find the optimum path.

Key words: memory principle, ant colony algorithm, ant colony agent, memory bank, path optima