Abstract: There are some disadvantages when using traditional ant colony algorithm to solve path planning problems such as slow convergence speed，easily plunging into a local optimum and so on.Therefore，a new ant colony algorithm with characteristic of sensory adaptation for path planning which simulates the ants’ behavior according to the laws of sensation is presented.In this algorithm，the ants’ sensory capacity is reducing when walking on the high-intensity pheromone nodes，thus，the algorithm can not only keep the fast convergence speed but also ensure the diverse search.The results of simulations demonstrate that this algorithm has a strong search ability and a low possibility to plunge into a local optimum，the best path can also be found in a short time even in a complex environment.

Key words: robot, path planning, ant colony algorithm, sensory adaptation

摘要： 用传统蚁群算法进行机器人路径规划具有收敛速度慢，容易陷入局部最优等缺点。为此，通过模拟蚂蚁的感觉特征，提出了一种具有感觉适应功能的蚁群算法，并应用到机器人路径规划中。算法设置蚂蚁的感觉能力随着在高强度信息素节点上的行走而降低，在快速收敛的同时保证搜索多样性。大量仿真实验表明，该算法有较强的搜索能力，不易陷入局部最优，即使在障碍非常复杂的环境下使用该算法，也能快速收敛到一条全局优化路径。

关键词: 机器人, 路径规划, 蚁群算法, 感觉适应