Abstract

Abstract: We address the multi-robot geometric task-and-motion planning (MR-GTAMP) problems in synchronous, monotone setups. The goal of the MR-GTAMP problem is to move with multiple robots objects to goal regions among other movable objects. To perform the tasks successfully and efficiently, the robots have to adopt intelligent collaboration strategies, i.e., deciding which objects to move, by which robot and to which position, and perform collaborative actions such as handovers. To enable robots with these collaboration capabilities we propose to first collect reachability and occlusion information for each robot, as well as information about whether two robots can perform a handover action by calling motion planning algorithms. The collected information is then used to build a graph structure which captures the precedence of the manipulations of different objects and supports to develop an optimal mixed-integer model to guide the search for highly efficient collaborative task-and-motion plans. Our search process is based on a Monte-Carlo Tree Search (MCTS) exploration strategy for an exploration-exploitation balance. We evaluate our framework in two challenging GTAMP domains and show our framework can generate high-quality task-and-motion plans regarding the average number of time steps, the average number of moved objects and the average planning time comparing with two state-of-the-art baselines.

Download the paper in pdf.

Many publishers do not want authors to make their papers available electronically after the papers have been published. Please use the electronic versions provided here only if hardcopies are not yet available. If you have comments on any of these papers, please send me an email! Also, please send me your papers if we have common interests.

This page was automatically created by a bibliography maintenance system that was developed as part of an undergraduate research project, advised by Sven Koenig.