Title: Optimal robust control approaches for a geostationary satellite attitude control

Authors: Naeimeh Najafizadeh Sari; Hadi Jahanshahi; Mahdi Fakoor; Christos Volos; Peyman Nikpey

Addresses: Faculty of New Sciences and Technologies, University of Tehran, Iran ' Faculty of New Sciences and Technologies, University of Tehran, Iran ' Faculty of New Sciences and Technologies, University of Tehran, Iran ' Department of Physics, Aristotle University of Thessaloniki, Thessaloniki, GR, 54124, Greece ' Faculty of Satellite Research Institute, Iranian Space Research Center, Tehran, Iran

Abstract: In this paper, two-optimal robust fuzzy proportional-integralderivative (FPID) and linear-quadratic regulator (LQR) controllers have been implemented for attitude control of a geostationary satellite, utilising momentum wheels. In the designed FPID controller, two fuzzy inference engines have been used, from which the second engine, is accounted to control the satellite attitude in severe deviations and preventing the system from instability. The designed FPID controller is optimised using the multi-objective genetic algorithm (MOGA) based on desired objective functions, which are deviation error from equilibrium states and control efforts. The optimal FPID controller is designed in such a way that while making extremum the desired objective functions, it also provides an appropriate controlling performance. Throughout designing robust LQR controller, design matrices of R and Q are selected in such a way to form a balance between the made control efforts and the settling time of the system.

Keywords: fuzzy PID controller; robust LQR controller; genetic algorithm; multi-objective optimisation; geostationary satellite attitude control.

DOI: 10.1504/IJAAC.2020.107090

International Journal of Automation and Control, 2020 Vol.14 No.3, pp.333 - 354

Received: 12 Jul 2018
Accepted: 26 Sep 2018
Published online: 04 May 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article