An Approach to Fast Multi-Robot Exploration in Buildings with Inaccessible Spaces

Abstract

The rapid exploration of unknown environments is a common application of autonomous multi-robot teams. For some types of exploration missions, a mission designer may possess some rudimentary knowledge about the area to be explored. For example, the dimensions of a building may be known, but not its floor layout or the location of furniture and equipment inside. For this type of mission, the Space-Based Potential Field (SBPF) method is an approach to multi-robot exploration which leverages a priori knowledge of area bounds to determine robot motion. Explored areas and obstacles exert a repulsive force, and unexplored areas exert an attractive force. While SBPF has advantages over other methods of robot exploration in terms of simplicity and performance, inaccessible space poses a problem: it exerts a permanent attractive force, pulling robots away from useful exploration elsewhere and creating minima at its boundary. Prior research established a simple method of filling in inaccessible space as a solid obstacle once an enclosing boundary is discovered; however, this method requires the entire enclosing boundary to be discovered before it can be filled. In this paper, we propose a novel combined SBPF and frontier-based method of robot exploration called O-SBPF. Our method adds two new space classifications: open, areas known to be accessible; and occluded, areas which may be inaccessible. We describe a ray-casting approach to designate areas as open or occluded, and incorporate this designation into potential vector calculations. We then show the effectiveness of O-SBPF using ROS/Stage in worlds with inaccessible space. O-SBPF significantly outperforms SBPF in rooms with large obstacles, successfully reaching 95% coverage while SBPF becomes stuck in a minima. In less complex rooms, we show that O-SBPF generally reaches 95% coverage at the same time or before SBPF.