TY - JOUR
T1 - Spacecraft anti-unwinding attitude tracking with guaranteed performance
T2 - A DREM-based adaptive control approach
AU - Chen, Bin
AU - Shao, Xiaodong
AU - Yang, Haoyang
AU - Li, Dongyu
AU - Hu, Qinglei
N1 - Publisher Copyright:
© 2025 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
PY - 2026/1/15
Y1 - 2026/1/15
N2 - This article proposes an indirect adaptive control scheme for spacecraft attitude tracking with guaranteed performance and unwinding avoidance. Performance funnels are imposed on a filtered tracking error to secure prescribed behavioral bounds for both attitude and angular velocity errors. Within this setting, if an adaptive controller is designed, the estimation-error-dependent perturbation term in the Lyapunov function derivative inevitably involves the unknown inertia matrix, which hinders the use of conventional direct adaptive control methods that typically rely on canceling the perturbation term. To address this issue, a parameter estimator is developed using the dynamic regressor extension and mixing (DREM) technique. Under a strictly weak assumption of interval excitation, the resulting estimation error dynamics can effectively dominate the perturbation term without depending on a fragile cancellation operation. By leveraging the DREM estimator and barrier functions, the designed adaptive controller achieves exponential attitude tracking with guaranteed performance bounds and unwinding avoidance. Finally, simulation and experimental results validate the efficacy of the proposed adaptive control approach.
AB - This article proposes an indirect adaptive control scheme for spacecraft attitude tracking with guaranteed performance and unwinding avoidance. Performance funnels are imposed on a filtered tracking error to secure prescribed behavioral bounds for both attitude and angular velocity errors. Within this setting, if an adaptive controller is designed, the estimation-error-dependent perturbation term in the Lyapunov function derivative inevitably involves the unknown inertia matrix, which hinders the use of conventional direct adaptive control methods that typically rely on canceling the perturbation term. To address this issue, a parameter estimator is developed using the dynamic regressor extension and mixing (DREM) technique. Under a strictly weak assumption of interval excitation, the resulting estimation error dynamics can effectively dominate the perturbation term without depending on a fragile cancellation operation. By leveraging the DREM estimator and barrier functions, the designed adaptive controller achieves exponential attitude tracking with guaranteed performance bounds and unwinding avoidance. Finally, simulation and experimental results validate the efficacy of the proposed adaptive control approach.
KW - Adaptive control
KW - Attitude tracking
KW - Dynamic regressor extension and mixing (DREM)
KW - Prescribed performance
UR - https://www.scopus.com/pages/publications/105024723010
U2 - 10.1016/j.asr.2025.10.066
DO - 10.1016/j.asr.2025.10.066
M3 - 文章
AN - SCOPUS:105024723010
SN - 0273-1177
VL - 77
SP - 2323
EP - 2338
JO - Advances in Space Research
JF - Advances in Space Research
IS - 2
ER -