Resilient Autonomous Subsystems for Unmanned Air Systems (UAS)

Distributed battle management of autonomous agents requires controlling a group of semi-isolated small teams, operating under extensive uncertainty about the enemy situation and also about the status and plans of some of our own assets. To minimize waste of resources, isolated agents make predictions about what other agents are planning to do. If these predictions are good, the vehicles will be able to coordinate effectively even during communication blackout periods.However, in cases were the communication black-outs are long as compared to the natural time constants of the mission, the knowledge on the state of the remote agent may become stale. As a consequence, the number of predicted possible paths for the remote agent will grow, increasing the computation load and uncertainty for the local agents lowering the team’s ability to coordinate.Barnstorm Research has developed RTREVE, a set of algorithms to support autonomous collaboration in communication challenged environments. RTREVE equipped teams are able to perform collaborative behaviors robustly, even when facing attrition and a challenging communications environment.For SBIR N193-141 we propose to implement RTREVE’s autonomous behaviors and supporting algorithms as RAIDER/FACE compliant Units of Portability (UoPs) and to verify their performance in simulation.