Research Project
Era-Agent: Embodied Robotics with Aerospace Agentic AI
Research Team
Lead Researchers:
- Dr. Enno Huang, Computer Science
Collaborators:
- Jean-Pierre de la Croix, JPL
- Michel Ingham, JPL
Student Team:
- Rowel Espejo
Funding
- Funding Organization:
- Funding Program:
SYNOPSIS
- Agent-swarms planning and operation.
- Embodied Agents use Action Prediction for Robot manipulation
- Rovers plan trajectory for Rover Navigation
- Agents use reasoning to stay aligned with trustworthy mission goals
Motivation/Research Problem
-
Many lunar missions, such as CADRE, are moving toward multi-agent systems.
-
Teams of heterogeneous robots can combine complementary strengths
-
Collaboration across different types of agent-swarms
Research Questions & Research Objectives
Research Methods
- We adopt a simulation-first research strategy, using lunar environments to develop, train, and validate ERA before deployment on physical robotic systems.
- We aim to enable sim-to-real transfer by learning coordination and reasoning
- ERA across diverse rover types in multiple environments to test how well multi-agent collaboration generalizes across different scenarios
Research Deliverables and Products
- Action prediction
- Multi-agent auto-coordination
- Rover navigation
- Trustworthy value alignment
Commercialization Opportunities
- Lunar Robot Coordination Platform: helps teams of robots work together on lunar missions.
- Trustworthy Space Autonomy: supports safe and reliable AI for space robotics.
- Sim-to-Real Robotics Framework: helps move robot systems from simulation to real deployment.
Research Timeline
Start Date:
End Date: TBD
Research Team
Lead Researchers:
- Dr. Enno Huang, Computer Science
Collaborators:
- Jean-Pierre de la Croix, JPL
- Michel Ingham, JPL
Student Team:
- Rowel Espejo
Funding
- Funding Organization:
- Funding Program: