Our project titled “Robust and Adaptive Autonomy for Multi-agent Maneuvers (RAAMM)” is selected for funding by the ARL A2I2 program! AI/ML systems for Army missions require the ability to adapt to changing dynamics and adversaries. This project aims to address this critical challenge through three research thrusts. Thrust one will develop machine vision systems that provide improved perception and prediction capabilities in situations with missing data, as would be the case when environmental conditions are unknown or insufficiently sensed. Thrust two will develop reinforcement learning algorithms and architectures for planning with multi-domain teams (including human supervisors) that are able to adapt to evolving dynamics and adversaries. Thrust three will evaluate these methods in a demonstration platform that seamlessly transitions from simulation-based prototyping to hardware-in-the-loop testing to real-world demonstrations on ground and aerial unmanned vehicles.
This project is a collaboration with Dr. Girish Chowdhary, Dr. RS Sreenivas, Dr. Jaggi Yedetore, and Dr. Heather Filippini from UIUC and Dr. John Rogers and Dr. Phil David from ARL.