Precision Autonomous Virtual Flight Control

Flight simulation provides a cost effective tool for testing aircraft systems. Testing avionics systems across a particular flight profile requires driving the simulation through the flight profile to generate the required simulated sensor data. Testing avionics systems with a human pilot is costly and it cannot provide precise repeatability, which is essential to evaluating the effect of mission parameters on aircraft survivability. However, a"virtual pilot"is capable of generating a repeatable control, which guarantees an identical flight every time. Development of a virtual pilot that can address a range of maneuvers and which can be adapted to various aircraft configurations is a demanding task. Most current methods are not feasible for use by application Test Engineers since those methods require manual tuning of the virtual pilot for each configuration and for each maneuver. Under this SBIR, ART proposes to develop a virtual pilot control generation tool that can accomplish the task effectively. The problems of accuracy, robustness, and efficiency will be addressed using the advanced inverse simulation technique, the mathematical optimization method, a feedback compensator, and parallel computing. A graphical user interface will be developed in order to facilitate the tool usage. The innovation involved is the integration of the diverse technologies of rotorcraft modeling, inverse simulation, and optimization into a unified tool that is efficient and effective in providing a repeatable control history to drive nonlinear rotorcraft models through user defined flight profiles.