Skip to main navigation Skip to search Skip to main content

Refined Metamodel Disturbance Observer-Based Control for Coarse Pointing Assembly under Constraints

  • Beihang University

Research output: Contribution to journalArticlepeer-review

Abstract

The target tracking performance of the coarse pointing assembly (CPA) is a critical factor in determining the data transmission efficiency of the inter-satellite laser communication. However, under the influence of multiple disturbances, such as gimbal coupling torque, friction, and satellite-Transmitted vibration, the angle tracking performance of the CPA can be decreased, which then affects the safety of the system in terms of angular velocity. To address the performance and safety requirements of the CPA, this paper proposes a refined metamodel disturbance observer-based state constraints controller. First, a refined metamodel disturbance observer is proposed to achieve accurate disturbance estimation by combining the data-driven state-space Kriging metamodeling method with a refined disturbance observer. Both disturbance numerical data and partially known information (e.g. vibration frequency and structure) are fully utilized to reduce estimation conservatism. Second, based on the observer, a state constraint controller is designed to ensure the angle tracking performance and angular velocity constraints. Finally, the effectiveness and robustness of the proposed control method are validated through numerical simulations, indicating an enhanced angle tracking performance compared to the traditional disturbance observer-based control method.

Original languageEnglish
Article number2450017
JournalGuidance, Navigation and Control
Volume4
Issue number4
DOIs
StatePublished - 30 Nov 2024

Keywords

  • Laser communication
  • coarse pointing assembly
  • disturbance observer-based control
  • metamodeling
  • state constraints

Fingerprint

Dive into the research topics of 'Refined Metamodel Disturbance Observer-Based Control for Coarse Pointing Assembly under Constraints'. Together they form a unique fingerprint.

Cite this