Multi-robot Exploration Using Self-Biddings under Constraints on Communication Range

Abstract

This paper proposes a completely decentralized algorithm for multi-robot coordination to explore and map terrain, under constraints on communication range. This algorithm is based on a market bidding process of unrevealed frontiers, where each robot calculates the bids independently. When it reaches its target position, the robot makes a decision by itself, which involves every one of the team members and their ranges of communication, under a decentralized scheme and without the necessity of a central module. Since the approach maintains a more stable communication network among the robots and, consequently, more similar individual maps, the result is an efficient and fault tolerant approach. The bidding function includes the distance of the considered robot towards the frontier, the distances to the others robots and their respective objectives and include the constraint on communication range. This function provides the robots with a complex behavior, so the robot tries to explore distant frontiers, while trying to keep itself within communication range of the other robots.