Research/ MULTI-ROBOT SYSTEMS
Multi-Agent Autonomy for Space Exploration on the CADRE Lunar Technology Demonstration
Jean-Pierre de la Croix, Federico Rossi, Roland Brockers, Dustin Aguilar, Keenan Albee, Elizabeth R. Boroson, Abhishek Cauligi, Jeff H. Delaune, Robert A. Hewitt, Dima Kogan, Grace Lim, Benjamin J. Morrell, Yashwanth Kumar Nakka, Viet Linh Nguyen, Pedro F. Proença, Gregg Rabideau, Joseph A. Russino, Maíra Saboia da Silva, Guy Zohar, Subha Comandur
Multi-Robot & SwarmsOpenAlex keyword: multi-agent coordination · Learning-based MethodsOpenAlex topic: Reinforcement Learning in Robotics · Autonomy, Task Planning & OperationsOpenAlex topic: AI-based Problem Solving and Planning
CADREnamed in title, abstract · CLPSnamed in abstract, keywords · Intuitive Machines (IM-1, IM-2)named in abstract
CADRE (Cooperative Autonomous Distributed Robotic Exploration) is a lunar technology demonstration mission of multi-agent autonomy on a team of three rovers and a base station. The mission is slated land at the Moon’s Reiner Gamma region as a CLPS (Commercial Lunar Payload Services) pay-load on the IM-3 mission in 2024. The goal of CADRE is to demonstrate how a team of autonomous rovers, receiving only high-level tasks from Earth, can autonomously explore a region of the Lunar surface, as well as perform a distributed measurement in coordination with a multi-static ground-penetrating radar. We envision that multi-agent autonomy will enable future missions to address hitherto-unanswered questions in planetary science on the Moon, Mars, and beyond. In this paper, we describe the autonomy architecture developed for CADRE, both for multi-agent coordination, and for single-agent driving surface mobility, and discuss the requirements and constraints that led to the selection of this architecture.