Intelligent Control of Unmanned Air Vehicles: Program Summary and Representative Results

Abstract

In early 2001, AFRL and NAVAIR issued a PRDA requesting proposals to develop an intelligent controller (IC) for unmanned combat air vehicles. Two key requirements of the IC were (1) a learning approach that could go beyond current adaptive controllers and "remember" what it had learned across flight conditions and (2) a reconfigurable path planner that accounted for changes in the inner-loop behavior and generated near-optimal trajectories in real time. This paper presents a summary of the resulting IC program and some initial technical results.