Human-Observation-Based Extraction of Path Patterns for Mobile Robot Navigation

In this paper, a mobile robot navigation method based on the observation of human walking is presented. The proposed method extracts paths that are frequently used by human and builds a topological map of the environment from the observed human walking paths. Unlike the conventional methods, the proposed method enables us to generate paths which are practical, have no obstacles, and are natural for humans since the paths reflect the motion of persons. For realizing the human observation in a large area, in this paper, multiple vision sensors are placed in space. By using distributed sensors, people can be observed even when the robot is not near them or if they are hidden behind obstacles. Mobile robot navigation based on the topological map is also performed with the support of the distributed sensors. The global position of the mobile robot can be directly measured by using external sensors, which makes the localization problem much easier. Based on the position information, the mobile robot can follow the generated paths and reach the goal point while avoiding obstacles.