Learning-based optimization method for attitude control of jet-driven spacecraft

Desong Du; Naiming Qi; Haoda Cheng; Minghao Han; Rui Zhou; Yanfang Liu

doi:10.1117/12.3032458

16 August 2024 Learning-based optimization method for attitude control of jet-driven spacecraft

Desong Du, Naiming Qi, Haoda Cheng, Minghao Han, Rui Zhou, Yanfang Liu

Proceedings Volume 13218, First Aerospace Frontiers Conference (AFC 2024); 132180G (2024) https://doi.org/10.1117/12.3032458
Event: First Aerospace Frontiers Conference (AFC 2024), 2024, Xi’an, China

Abstract

For the attitude control problem of jet-driven spacecraft, this paper proposes a reinforcement learning (RL) based attitude control optimization algorithm (LAC), which optimizes the controller/policy's energy consumption while satisfying stability constraints. The algorithm incorporates three neural networks: the control policy network, the Lyapunov neural network, and the energy neural network. The control policy network directly outputs the forces of the jet thrusters, implicitly solving the thrust distribution optimization problem under the given control torque. The Lyapunov neural network and the energy neural network are respectively utilized to describe the stability constraints and energy consumption of the control policy network. The proposed control policy optimization algorithm employs the Lagrange multiplier method to optimize the control policy network, to minimize the energy consumption with the sample-based stability constraints. Numerical simulation results demonstrate that, compared with traditional RL algorithms and attitude control methods, the proposed attitude control algorithm exhibits significant advantages in terms of energy consumption.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Desong Du, Naiming Qi, Haoda Cheng, Minghao Han, Rui Zhou, and Yanfang Liu "Learning-based optimization method for attitude control of jet-driven spacecraft", Proc. SPIE 13218, First Aerospace Frontiers Conference (AFC 2024), 132180G (16 August 2024); https://doi.org/10.1117/12.3032458

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available