Leading the Way: An Efficient Multi-robot Guidance System (2015)
Recent advances in service robotics have made it possible to deploy a large number of mobile robots in indoor environments to perform tasks such as delivery, maintenance and eldercare. If a centrally connected multi-robot system is available, can it be effectively used to aid humans in other on-demand tasks? In this paper, we demonstrate how individual service robots in a multi-robot system can be temporarily reassigned from their original task to help guide a human from one location to another in the environment. We formulate this multi-robot treatment of the human guidance problem as a Markov Decision Process (MDP). Solving the MDP produces a policy to efficiently guide the human, but the state space size makes it infeasible to optimally solve it. Instead, we use the Upper Confidence bound for Trees (UCT) planner to obtain an approximate solution. We show that this solution outperforms an approach that uses a single robot to guide the human from start to finish.
In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Istanbul, Turkey, May 2015.

Samuel Barrett Ph.D. Alumni sbarrett [at] cs utexas edu
Piyush Khandelwal Ph.D. Alumni piyushk [at] cs utexas edu
Peter Stone Faculty pstone [at] cs utexas edu