Two-Stage Automatic Mass Balancing and Attitude Control for 3-DoF Spacecraft Attitude Simulators
DOI:
https://doi.org/10.31224/7342Keywords:
3-DoF spacecraft attitude simulator, automatic mass balancing, CR-CG offset estimation, gravity torque compensation, least-squares estimation, non-orthogonal sliding mass mechanismAbstract
Three-degree-of-freedom (3-DoF) air-bearing simulators are widely used for ground-based validation of spacecraft attitude determination and control systems. However, residual gravity torques caused by center-of-rotation to center-of-gravity (CR-CG) offsets can degrade simulation fidelity, particularly during coupled three-axis maneuvers. This paper presents a sequential coarse-to-fine automatic mass balancing framework, referred to as the Robust Automatic Mass Balancing Operator (RAMBO), designed to reduce residual CR-CG offsets under mechanical non-orthogonality and cross-axis coupling while maintaining a negative vertical offset for pendulum stability. The framework sequentially applies an active coarse stage (CRAMBO) for rapid offset reduction and a fine stage (FRAMBO) for precise CR-CG offset estimation using free-response motion data collected after platform reinitialization. To address the remaining offset-induced torque, a gravity-torque feedforward term is incorporated into a quaternion proportional-derivative (PD) controller. Hardware experiments during a coupled diagonal maneuver demonstrated that the estimated dynamic-equivalent CR-CG offset converged to approximately [0.08, 0.12, -27.62] μm. Based on the onboard VN-100 estimates, the integrated balancing and control framework maintained steady-state attitude errors within ±0.5° and angular velocity errors within ±0.4°/s for roll and ±0.2°/s for pitch and yaw. A secondary IMU comparison further provided a roll/pitch consistency check rather than an independent ground-truth validation. These results demonstrate the effectiveness of the proposed approach for safety-constrained three-axis attitude-control validation.
Downloads
Downloads
Posted
License
Copyright (c) 2026 Yong-Hwan Kim

This work is licensed under a Creative Commons Attribution 4.0 International License.