Planning and Optimization for Multi-Robot Planetary Cave Exploration under Intermittent Connectivity Constraints
Filip Klaesson, Petter Nilsson, Tiago Stegun Vaquero, Scott Tepsuporn, Aaron D. Ames and Richard M. Murray
Published in ICAPS Workshop on Planning and Robotics, 2020
Exploring subsurface structures with autonomous robots is of growing interest in the context of planetary caves studies. Communication between robots in these environments is severely degraded which complicates coordination and information distribution. In this paper we focus on planning for mobility and communication in a cave exploration scenario where the situational awareness of a static base station is critical. We propose a notion of information-consistency where a plan itself is part of the information to be shared between robots, and propose a method for generating information-consistent plans. We discuss in detail how the resulting plan can be robustly implemented with minimal communication through local mission executives that run on individual robots. We investigate the performance of the planning algorithm, and integrate the local mission executives in a high-fidelity simulation environment.
Planning and Optimization for Multi-Robot Planetary Cave Exploration under Intermittent Connectivity Constraints: Download
Presentation slides are available here.