Collaborative Robot Exploration and Rendezvous

Abstract

We consider the problem of how two heterogeneous robots can arrange to meet in an unknown environment from unknown starting locations: that is, the problem of arranging a robot rendezvous. We are interested, in particular, in allow- ing two robots to rendezvous so that they can collaboratively explore an unknown environment. Specically , we address the problem of how a pair of exploring agents that cannot communicate with one another over long distances can meet if they start exploring at dieren t unknown locations in an unknown environment. We propose several alternative algorithms that robots could use in attempting to rendezvous quickly while continuing to explore. These algorithms exemplify dieren t classes of strategy whose relative suitability depends on characteristics of the prob- lem denition. We consider the performance of our proposed algorithms analytically with respect to both expected- and worst-case behaviour. We then examine their behaviour under a wider set of conditions using both numerical analysis and also a simulation of multi-agent exploration and rendezvous. We examine the exploration speed, and show that a multi-robot system can explore an unknown environment faster than a single-agent system, even with the constraint of performing rendezvous to allow communication. We conclude with a demonstration of rendezvous implemented on a pair of actual robots.