Learning Safe Multi-UAV Coordination with Temporal-Spatial Constraints
J. Pierre, X. Sun, R. Fierro · AIAA Science and Technology Forum and Exposition.
Safe autonomy · multi-agent systems · controls
Research & Publications
Research on safety-constrained multi-agent reinforcement learning, UAV coordination, AI/ML systems, and AI-enabled control.
Dissertation
Securing The Skies: Safety-Constrained Decentralized Multi-UAV Coordination With Deep Reinforcement Learning
Ph.D. Dissertation, University of New Mexico, 2024.
My doctoral work studied how decentralized multi-agent systems can learn coordinated behavior while respecting safety and temporal-spatial constraints. The research sits at the seam between control theory and reinforcement learning: policy learning supplies adaptability, while explicit safety structure keeps the learned behavior connected to physical constraints and real-world deployment needs.
Selected publications
Multi-Agent Partial Observable Safe Reinforcement Learning for Counter Uncrewed Aerial Systems
J. Pierre, X. Sun, R. Fierro · IEEE Access, Vol. 11.
Multi-Agent Deep Reinforcement Learning for Countering Uncrewed Aerial Systems
J. Pierre, X. Sun, R. Fierro · Distributed Autonomous Robotic Systems.
Orthogonal Vector Field-Based Control for Multi-Robot 3D Moving-Target Circumnavigation
Control and coordination research presented at IEEE CDC.
Hardware Implementation of Model Predictive Control for Relative Motion Maneuvering
Controls research presented at the American Control Conference.
Research themes
- Safe and constrained reinforcement learning
- Decentralized multi-agent coordination
- Guidance, navigation, and control
- Autonomous systems and robotics
- Computer vision and perception
- Embedded AI and neural architecture search