An adaptive neural fuzzy control scheme for rotational maneuvering and vibration suppression of a flexible satellite system

Abstract

This paper presents an adaptive optimal control technique that addresses the problem of position tracking control in flexible mechanical systems. Proposed adaptive neural fuzzy control (ANFC) scheme has self-tuning mechanism and is based on the Takagi Sugeno Kang (TSK) model. ANFC parameters are optimized by the conjugate gradient (CG) method. The developed control scheme is then incorporated to a rotating satellite with flexible appendages (RSFA). Flexible satellites and many similar large flexible systems belong to the same class of multi-body distributed/discrete non-linear systems. These systems are highly coupled and infinite dimensional. The control scheme's objective is to perform rotational maneuvering of the satellite and to suppress the vibrations induced by the flexible appendages. ANFC based on steepest descent (SD) and adaptive proportional integral derivative (APID) control schemes have also been implemented, so to provide a comparison with the proposed control scheme. Results have shown that proposed ANFC scheme has faster convergence than other well established control schemes, and it performs satisfactorily well in suppressing appendage vibrations while maintaining position control.