Disturbance-Learning-Based Attitude Maneuvering Control for Spacecraft Under Strong Composite Disturbances

Attitude maneuvers of spacecraft inevitably introduce composite disturbances such as center of mass variations, structural flexible vibrations, and actuator uncertainties. These disturbances are strongly coupled with the system's states and control inputs, forming a composite disturbance profile that severely impacts attitude tracking and pointing accuracy. To address this challenge, a maneuvering control scheme based on composite disturbance learning is proposed. First, a deep coupled spacecraft attitude dynamics model is developed to capture the influence and transmission mechanisms of composite disturbances. Then, a composite controller incorporating a disturbance learning observer is designed to perform online modeling and real-time compensation of these disturbances. The effectiveness of the proposed approach is validated through experiment studies.