Loading...

Actuator failure-tolerant control of an all-thruster satellite in coupled translational and rotational motion using neural networks

Tavakoli, M. M ; Sharif University of Technology | 2018

1202 Viewed
  1. Type of Document: Article
  2. DOI: 10.1002/acs.2942
  3. Publisher: John Wiley and Sons Ltd , 2018
  4. Abstract:
  5. The nonlinear model predictive control (MPC) approach is used to control the coupled translational-rotational motion of an all-thruster spacecraft when one of the actuators fails. In order to model the dynamical response of the spacecraft in MPC, instead of direct integration, a neural network (NN) model is utilized. This model is built of a static NN, followed by a dynamic NN. The static NN is used to find the changes of the mapping of “the demanded forces to the thrusters” and “the real torques/forces produced by the remaining thrusters” after the failure occurrence through online training. In this manner, the effect of failed thruster on the dynamics can be found and the need for conventional and separate “fault detection and isolation” in previous works is eliminated. The dynamic NN is used to replicate the dynamical equations of the spacecraft excluding the effects of the mapping of thrusters demand to the real generated force and torque. Using updated model of the spacecraft through the learning capability of the NNs, the nonlinear MPC is able to compensate the failure of the thruster with the help of the remaining active thrusters. Through numerical simulations, it is shown that in the presence of thruster failure(s) this approach can effectively control the spacecraft with the remaining actuators. © 2018 John Wiley & Sons, Ltd
  6. Keywords:
  7. All-thruster satellite ; Coupled translational-rotational satellite control ; Dynamic neural network ; Thruster failure-tolerant satellite control ; Actuators ; Fault detection ; Mapping ; Neural networks ; Nonlinear systems ; Predictive control systems ; Rotational flow ; Satellites ; Direct integration ; Dynamic neural networks ; Fault detection and isolation ; Learning capabilities ; Neural network model ; Nonlinear model predictive control ; Satellite control ; Translational-rotational motion ; Model predictive control
  8. Source: International Journal of Adaptive Control and Signal Processing ; 2018 ; 08906327 (ISSN)
  9. URL: https://onlinelibrary.wiley.com/doi/full/10.1002/acs.2942