Hierarchical action control technique based on prediction time for autonomous omni-directional mobile robots

Recently, various essential technologies of an autonomous mobile robot such as a self localization scheme, an environmental map formation and path planning, learning algorithm and communication are developed in the area of robot. In addition, a variety of service robots which offers service with the actual environment with other moving objects, including people are proposed and developed(B. Graf, 2004; Asoh,2001; DeSouza,2002; Thrun,1999; R. Bischoff). A variety of tasks are required for such a service robot, but here we will focus on problems related to moving, which is the most fundamental and important of tasks. In the environment include humans, safe and efficient movement should be required. To deal with this problem, there have been a variety of mobile control techniques for the autonomous mobile robot. The visibility graph-like method and the Voronoi diagram method are the technique to design the pass which can reach a goal without colliding with obstacles on the basis of the advance knowledge of environment. However, it is not easy to work effectively with an actual environment where the unknown static or moving obstacles exist, because the complete knowledge regarding environment is needed. On the one hand, sensor based approach and reactive technique are the technique where recognizes the information of environment with the sensor and then decides behaviour in real time. Therefore it is the technique which is suited for actual environment. This study proposes a hierarchical action control method for autonomous omni-directional mobile robot to achieve a smooth obstacle avoidance ensuring safety in the presence of moving obstacles including humans. The hierarchical control method considers various time scales for actions such as goal path planning, obstacle avoidance within the recognizable range, and emergency avoidance to deal with unexpected events. In the proposed method, several modules for each action are composed in parallel. The vertical axis is the time scale in the control system. In the lowest module, the robot can move to the goal safely and efficiently by planning from the environment information which is obtained in advance. In the higher module, the robot moves more safely by using the estimated information about the obstacles to avoid them. By integrating the output of each module comprehensively, it is possible to realize a safe and efficient movement according to 18