An artificial neural network control system for spacecraft attitude stabilization

Abstract

This document reports the results of research into the application of artificial neural networks to controlling dynamic systems. The network used is a feed-forward, fully-connected, 3-layer perception. Two methods of training neural networks via error back-propagation were used. Pattern matching training is a direct method that teaches the basic response. Performance index training is a new technique that refines the response. Performance index training is based on the concept of enforced performance. A neural network will learn to meet a specific performance goal if the performance standard is the only solution to a problem. Performance index training is devised to teach the neural network the time-optimal control law for the system. Real-time adaptation of a neural network in closed loop control of the CrewEquipment Retriever was demonstrated in computer simulations.