Path Planning Using Pedestrian Information Map for Mobile Robots in a Human Environment

Abstract

A method is presented for planning the path of an autonomous mobile robot operating in a human environment. Traditional planning methods use a map showing only static obstacle information, so the path generated does not reflect movement rules in the environment. As a result, the path planning does not take into consideration such things as inaccessible area that there is not the obstacle and passage that is one side traffic. The proposed method overcomes this problem by using a pedestrian information map that shows the traffic frequency and velocity in each direction. The map shows the traffic areas and human movement manner. A pedestrian is defined as person moving toward a destination, and a pedestrian information map is generated using only the pedestrian trajectories for all logged human trajectories. Experimental results showed that the proposed method can effectively plan a path through traffic areas.