Navigable voronoi diagram : a local path planner for mobile robots using sonar sensors

Abstract

As robotics research is directed towards intelligent robot behavior, mobile robots need a capability to explore known / unknown environments autonomously. The ability of the autonomous navigation begins with a path planner local and global. In this paper, we address a new local path planner using a Navigable Voronoi Diagram(NVD) which comes from the Voronoi Diagram(VD). The characteristics of the NVD are safety, smoothness, fast reaction and so on. The NVD is constructed from sonar sensor readings and is integrated with a cost function. Another important advantages of the NVD is that it can drastically reduce the uncertainty of cheap sonar sensor readings due to its nature. The proposed local path planner based on the NVD has been successfully implemented on a Pioneer 3 equipped with 12 sonar sensors and its advantages were verified by 2 experiments in different indoor environments.