基于规则的移动机器人实时运动规划

研究移动机器人在动态环境中的导航与避障问题。为提高规划的实时性,提出了基于规则的规划方法,将多移动障碍环境机器人的运动规划分解为相对简单的单移动障碍运动规划,利用最优控制来实现单障碍的最优避障,并用智能搜索方法解决了移动机器人在多移动障碍环境中的实时运动规划问题。仿真实例表明了该方法的有效性。

The problem on navigation and collision avoidance for autonomous mobile robots in a dynamic en-vironment is studied. A new control scheme based on rule is presented to solve the problem. First, the scheme decomposing multiple moving obstacles along global path into a single obstacle on a local path is pro-posed. Then collision avoidance for a single obstacle is solved by means of optimal control strategy . In addion to collision avoidance,the proposed scheme is concerned with performance in terms of time consistency with norminal plan. The simulation results based on the scheme are presented to demonstrate its efficiency.

清华大学智能技术与系统国家重点实验室;;中科院机器人学开放研究实验室资助课题