人工势场引导信息素扩散的机器人导航规划

Robot Guidance Planning Based on Pheromone Diffusion Guided by Artificial Potential Field

为了保证机器人自动导航过程中沿着最优路径前进,提出了基于势场引导信息素扩散的导航规划方法。建立了工作环境的栅格模型;在分析传统蚁群算法缺陷基础上,提出使用人工势场引导信息素扩散,制定了栅格环境下信息素扩散方向查询表,建立了信息素扩散量的圆锥模型;在人工势场引导下,信息素扩散到隐含最优路径附近区域,将蚂蚁搜索范围限制在一个隐含最优路径的局部区域内,减小了蚁群搜索空间。经仿真验证,与传统蚁群算法相比,人工势场引导算法的信息素能够快速聚集在最优路径上;与另一改进蚁群算法相比,势场引导算法规划出导航路径的迭代次数减少了一半,消耗时间降低了约一个数量级,保证了导航规划的快速性和实时性。

To guarantee optimal path can be found by robot under global path planning,meanwhile decrease planning time,path planning method based on potential field guided ant colony algorithm is proposed.Grid model of work environment is built.On basis of analyzing shortcomings of traditional ant colony algorithm,phenomenon diffusion guided by potential field is put forward.Phenomenon diffusion direction lookup table under grid model is draw up,and circular cone model of phenomenon diffusion capacity is built.Guided by potential field,phenomenon diffuses to the area implying optimal path,which restricts searching range to local area,and breadth of ant searching space is reduced.Clarified by simulation,compared with traditional ant colony algorithm,phenomenon is aggregated to optimal path fatly when it is guided by potential field.Compared with another improved ant colony algorithm,iteration times of potential field guided ant colony algorithm decrease by half until optimal path is planned,besides,consuming time decreases by about one order.