Incremental Human Motion Map system and human walking behavior representation in indoor environment

Abstract

For robots working in human living space, both environmental information acquisition and human existence detection and estimation are necessary. In this study, by using human existence estimation function and mapping function of SLAM, a concept and architecture of Human Motion Map by representing extracted human behavior in the human living space as a map are proposed. Human Motion Map is a high-dimensional map, which incorporating human states information into results of conventional SLAM. On building Human Motion Map incrementally, human walking area interpolating algorithm and map update algorithm are developed. Finally, we implemented human motion map generating system on a mobile robot and demonstrated a Human Motion Map result for a floor area of our research lab in CIT, generated from individual observations of 80 experiments. Human Motion Map could be used for robots' motion planning of robots and further estimation of human motion in dynamic environment.