International Workshop on Autonomous Agents and Multi-Agent Systems for Space Applications
MASSpace'24
at AAMAS 2024, the 23rd International Conference on Autonomous Agents and Multi-Agent Systems

The workshop on Autonomous Agents and Multi-Agent Systems for Space Applications (MASSpace) aims to be a multidisciplinary meeting place to discuss the contributions of autonomous agents and multi-agent systems to the space domain. In deed, the Space domain is moving fast, and recent evolution tends to consider more and more complex and composite systems (e.g. larger constellations, multiple-mission federations, multi-user systems, heterogenous robotic systems), with stronger expectations, notably to perform more and more accurate environmental monitoring, complex requests, or richer exploration scenarios. In such context, agent-based and multi-agent systems appear to provide relevant paradigms to answer to these expectations.

In recent years, several papers applying AAMAS models and techniques to Space domain have been published in AAMAS and other venues which strongly advocates for organizing an event on the cross-fertilization of AAMAS and Space, as (i) to provide Space domain experts with the means to use these multi-agent models and techniques, and (ii) to challenge multi-agents models and techniques with novel problems coming from the Space domain. Thus, any AAMAS attendee could find interest in participating to the workshop, due to the broad scope of relevant AAMAS-related topics.

Since the early years of AAMAS, Space have been identified as a very relevant and challenging application domain, and today, with the ever growing size and complexity of Space missions and their environment (e.g. NewSpace), all AAMAS topics and techniques (see AAMAS call for papers) have becoming even more relevant to address this innovative and challenging topics, such as (but not limited to):

• Organizations and institutions to model Space Systems
• Policy, regulation, sanctions, accountability and legislation for Space Systems, especially New Space Applications
• Trust and reputation in Space Systems
• Architecture and modelling for Space Applications
• Formal verification and validation of agent-based Space Systems
• Programming models and languages to develop agent-based Space Applications
• Human-agent interaction especially in Space environment and constraints
• Distributed problem solving to efficiently coordinate decisions made by space assets and actors
• Coalition formation to coordinate multiple missions and systems
• Single-agent and multi-agent planning and scheduling to determine plans to be performed by missions
• Reasoning and learning under uncertainty to devise robust plans and behaviors
• Machine learning and deep learning to adapt systems and agents behaviors
• Auctions and Mechanism Design to coordinate resource allocation in Space Systems
• Interactive simulation to assess Space Systems in realistic but simulated settings
• Simulation of complex systems such as Space Systems
• Fair Allocation of Space assets between multiple stakeholders
• Single- and Multi-agent Reinforcement Learning to learn collective behaviors
• RL in partially observable settings to handle uncontrolled environments
• Safe, Robust, Explainable RL to provide strong guarantees on learning agents
• Multi-robot coordination and collaboration for Observation and Exploration missions

The workshop welcomes submissions addressing any such topics applied to any Space-related application or used case, ranging from ground operations to deep space observation and exploration.

0855-0900		Introductory remarks	Steve Chien, Gauthier Picard
0900-1000		Invited talk	Federico Rossi
1000-1030		Coffee break
1030-1230		Session 1
	1030-1055	Lunar Leader: Persistent, Optimal Leader Election for Multi-Agent Exploration Teams 📄	Keenan Albee (NASA Jet Propulsion Laboratory), Sriramya Bhamidipati (NASA Jet Propulsion Laboratory), Joshua Vander Hook (Outrider, Inc.), Federico Rossi (Jet Propulsion Laboratory - California Institute of Technology)
	1055-1120	CADRE MoonDB: Distributed Database for Multi-Robot Information-Sharing and Map-Merging for Lunar Exploration 📄	Maíra Saboia (Jet Propulsion Laboratory - California Institute of Technology), Federico Rossi (Jet Propulsion Laboratory - California Institute of Technology), Viet Nguyen (Jet Propulsion Laboratory - California Institute of Technology), Grace Lim (Jet Propulsion Laboratory - California Institute of Technology), Dustin Aguilar (Jet Propulsion Laboratory - California Institute of Technology), Jean-Pierre de la Croix (Jet Propulsion Laboratory - California Institute of Technology)
	1130-1155	DRIFT: Deep Reinforcement Learning for Intelligent Floating Platforms Trajectories 📄	Matteo El Hariry (University of Luxembourg), Antoine Richard (University of Luxembourg), Vivek Muralidharan (University of Luxembourg), Matthieu Geist (Cohere), Miguel Olivares (Universidad de Luxemburg)
	1155-1220	Integrated Modeling and Planning for On-Orbit Assembly of Large Space Structures with Mobile Crawling Robots 📄	Alexandre Albore (ONERA), Mathieu Rognant (ONERA)
1230-1400		Lunch break
1400-1515		Session 2
	1400-1425	Going Beyond Mono-Mission Earth Observation: Using the Multi-Agent Paradigm to Federate Multiple Missions 📄	Jean-Loup Farges (ONERA), Filipo S. Perotto (ONERA), Gauthier Picard (ONERA), Cedric Pralet (ONERA), Cyrille De Lucy (Airbus Defence and Space), Jonathan Guerra (Airbus Defence and Space), Philippe Pavero (Airbus Defence and Space), Fabrice Planchou (Airbus Defence and Space)
	1425-1450	Hierarchical Temporal Planning in an Earth Observation Satellite Software Architecture 📄	Alexandre Albore (ONERA), Rafael Bailon-Ruiz (ONERA)
	1450-1515	Understanding Drill Data for Autonomous Application 📄	Sarah E. Boelter (University of Minnesota), Ebasa G. Temesgen (UMN), Brian Glass (NASA Ames Research Center), Maria Gini (University of Minnesota)
1515-1530		Closing remarks	Gauthier Picard

Title: Multi-agent autonomy on the Moon: NASA' Coordinated Autonomous Distributed Robotics Explorers (CADRE) mission

Abstract: This talk will present the multi-agent autonomy architecture of NASA's Cooperative Autonomous Distributed Robotic Explorers (CADRE) mission, a technology demonstration that will deliver a team of autonomous rovers to the Moon's Reiner Gamma region in the coming year. Multi-robot systems hold great promise to address a number of key questions in planetary science. They can observe phenomena of interest from multiple, geographically-distributed locations at the same time; produced detailed three-dimensional images of the subsurface through seismic and RADAR surveys; and offer increased resilience compared to monolithic explorers, enabling bolder exploration. Autonomy is a key enabling technology for these multi-robot systems: it allows agents to operate together as a team, building on each other's abilities, with no humans in the loop, a critical capability when light-speed delays and low bandwidth make teleoperation infeasible or inefficient. But how do we design, build, test, and fly algorithms for a team of autonomous robots? How should the robots decide who (if any) should be in charge; when they should drive, and when they should recharge; how to explore an unknown region together; and how to collect measurements in formation? In this talk, we will walk through CADRE's autonomy stack, explore the trade-offs that informed the design of its multi-agent autonomy architecture, and end by speculating about the future of multi-robot systems for planetary exploration.

Bio: Dr. Federico Rossi is a Robotics Technologist with the Multi-Agent Autonomy Group within the Robotics section of NASA’s Jet Propulsion Laboratory. At JPL, he led the multi-agent autonomy team for NASA' Coordinated Autonomous Distributed Robotics Explorers (CADRE) mission, which will deliver a team of three autonomous rovers to the Moon's Reiner Gamma region in the coming year. Dr. Rossi received a Ph.D. in Aeronautics and Astronautics at Stanford University under the guidance of Prof. Marco Pavone in 2018, and M.Sc. in Space Engineering from Politecnico di Milano in 2013. His research focuses on decision-making under uncertainty in multi-agent robotic systems and operations of autonomous agents, with applications to planetary exploration and Earth science. At JPL, he has developed autonomy technologies for cooperative robotic exploration, joint orbit and observation optimization for exploration of small bodies, and under-ice navigation in Antarctica.