Active Fault-Tolerant Sliding Mode Control of Flexible Spacecraft Using Adaptive Observer Based on Radial Basis Functions Neural Network

Document Type : Dynamics, Vibrations, and Control

Authors

1 Master's degree, Aerospace Research Institute, Tehran, Iran

2 Associate Professor, Aerospace Research Institute, Tehran, Iran

Abstract

This paper is focused on the design and analysis of a fault-tolerant sliding mode control algorithm together with an adaptive observer for applications to a flexible spacecraft in order to attenuate system uncertainties, actuator faults, and external disturbances. For estimation of the actuator faults, an adaptive observer is designed using a radial basis function neural network, whose performance is compared with an iterative learning observer. The proposed fault-tolerant control adopts a PID sliding surface for high performance, robustness, and fast response. Additionally, the vibration suppression control algorithm based on strain rate feedback was designed for active suppression of structural vibrations using piezoelectric actuators and sensors. Stability analysis of the closed-loop system is performed using the Lyapunov theorem to ensure its robust performance. A key feature of the proposed approach is its simplicity and its ability to stabilize the system under fault conditions while providing accurate actuator fault estimation with minimal computational burden. Simulations, presented as a comparative study, demonstrate the superior performance, robustness, and fault-tolerance of the proposed approach for a system with fully coupled rigid-flexible dynamics.

Graphical Abstract

Active Fault-Tolerant Sliding Mode Control of Flexible Spacecraft Using Adaptive Observer Based on Radial Basis Functions Neural Network

Keywords

Main Subjects


Smiley face

Volume 21, Issue 2 - Serial Number 80
Serial No. 80, Summer Quarterly
July 2025
  • Receive Date: 24 April 2025
  • Revise Date: 02 June 2025
  • Accept Date: 11 June 2025
  • Publish Date: 09 July 2025