Adaptive actor-critic learning-based robust appointed-time attitude tracking control for uncertain rigid spacecrafts with performance and input constraints
Advances in Space Research(2023)
摘要
This paper addresses the prescribed performance attitude tracking control problem for a rigid spacecraft with input constraint and bounded disturbance on the special orthogonal group (SO (3 )). By introducing a proper pseudo-Morse function as the attitude error function, the tracking problem on matrix manifold SO (3 ) is reformulated as an equivalent stabilization problem in the associated Lie algebra (so (3 )). To on-line compensate the system uncertainties, an actor-critic neural network (NN) architecture is incorporated into the controller design process. The critic NN intends to reconstruct the long-term integral cost function and evaluate the prescribed per-formance attitude tracking quality. Based on the output of the critic NN, the actor NN is utilized to reconstruct the unknown system uncertainties and generate the auxiliary compensation control. By introducing the actor-critic architecture into the logarithm barrier Lyapunov function method, a model-free attitude tracking controller is proposed. Furthermore the input saturation limit is considered to support the normal operation of the rigid spacecraft. Besides, an adaptive robustifying term is designed to enhance the disturbance attenuation performance. By incorporating the Lyapunov stability analysis, we prove that the designed adaptive robust controller can ensure that all signals are uniformly ultimately bounded (UUB), and prescribed performance constraints are not destroyed. Finally, numerical simulations are included to verify the performance of the proposed adaptive robust tracking control strategy with intelligent actor-critic learning mechanism.(c) 2022 COSPAR. Published by Elsevier B.V. All rights reserved.
更多查看译文
关键词
Actor-critic neural network,Attitude tracking control,Barrier Lyapunov function,Bounded disturbance,Input constraint,Prescribed performance
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要