Home research People General Info Seminars Resources Intranet
Abstract

James Motes, Read Sandstrom, William Adams, Tobi Ogunyale, Shawna Thomas, Nancy M. Amato, "Interaction Templates for Multi-Agent Systems," Technical Report, TR18-002, Parasol Laboratory, Department of Computer Science, Texas A&M University, Jul 2018.
Technical Report(pdf, abstract)

This paper describes a framework for multi-robot motion planning problems that require or utilize interactions between robots. Solutions need to consider interactions between agents on a motion planning level to determine the feasibility and cost of each agent's portion of the task. Modeling these problems with current task and motion planning approaches typically requires reasoning about the possible interactions and checking many of the possible robots combinations when searching for a solution. We present a multi-robot motion planning method called Interaction Templates (ITs) which offloads certain types of robot interactions from the task planner to the motion planner. ITs encode iterations between a set of robots and are tiled and connected to robots' individual roadmaps. The resulting combined roadmap allows interactions to be considered by the motion planner. We apply ITs to homogeneous and heterogenous robot teams under both required and optional cooperation scenarios which previously required a task planning method. We show improved performance over a current TMP approach.