Abstract
In environments where a robot lives together with human, it is necessary for the robot to work objective tasks safely. Furthermore, the human activity should be respected more than the robot behavior. Therefore, we propose a human frequency map (HFM), which is a grid map using the observed human position and frequency in each term. The HFM can construct pathway information available for the robot because the human walked path seems to be safe in high possibility. A main pathway network based on Voronoi algorithm is automatically generated from the HFM. Then, the path of respecting human movements by using the HFM information can be planned so as to consider the difficulty of encounter between robot and human or consider the ease of passing each other. Experimental results applied to a path planning of a mobile robot are demonstrated to show the feasibility of the proposed HFM.
