Title :
Adaptive critic design of a control augmentation system for an unmanned aerial vehicle
Author :
Santiago, Roberto A. ; Lendaris, George
Author_Institution :
NW Computational Intelligence Lab, Portland State Univ., OR, USA
Abstract :
Command augmentation systems (CAS) are a common part of modern airplane control and are best characterized as a form of tracking control. Pilot commands communicated through stick, pedal and throttle commands are translated into desired states for the airplane and the CAS through a close loop configuration with airplane work to minimize the difference between the actual state of the plane and the desired state as communicated from the pilot. In the research reported, simulated stick x commands were translated into desired roll rates for a simulated airplane. Using dual heuristic programming, a form of neurodynamic programming (a.k.a. adaptive critic methods and reinforcement learning), a CAS was designed. The resultant CAS shows effective tracking response between the desired and actual roll rate of the airplane.
Keywords :
aerospace simulation; aircraft control; closed loop systems; control system synthesis; dynamic programming; learning (artificial intelligence); neurocontrollers; actual roll rate; adaptive critic design; control augmentation system; dual heuristic programming; modern airplane control; neurodynamic programming; pilot commands; reinforcement learning; simulated stick commands; throttle commands; tracking control; unmanned aerial vehicle; Adaptive control; Airplanes; Content addressable storage; Control nonlinearities; Control systems; Learning; Nonlinear control systems; Optimal control; Programmable control; Unmanned aerial vehicles;
Conference_Titel :
Systems, Man and Cybernetics, 2003. IEEE International Conference on
Print_ISBN :
0-7803-7952-7
DOI :
10.1109/ICSMC.2003.1244364