Execution-Level Internal-Load-Regularized Wrench Control for Position-Servo Space Robot in Closed-Chain Truss Climbing
IEEE Robotics and Automation Letters · 2026-09 · 11 (9) · 10258-10265
DOI: 10.1109/lra.2026.3711367 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Robotics and Automation Letters, 2026-09. Verfasst von Shutao Wu und 8 weiteren Autor:innen. Die Arbeit nennt 22 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Postdoctoral Fellowship Program of CPSF, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
Multi-contact truss climbing for on-orbit assembly requires repeated rigid support switching, which changes the active contact set, feasible wrench set, and grasp-map null space, and can amplify closed-chain internal loading on lightly damped structures. This paper proposes a switching-oriented internal-load-regularized wrench-control execution framework for closed-chain truss-climbing robots driven by high-reduction position servos. A center-of-inertia-based net-wrench command is distributed by a feasibility-constrained quadratic program (QP) that enforces contact limits and penalizes grasp-map null-space components. The allocated branch wrenches are executed through transition-conditioned variable-impedance admittance with damping injection, online disturbance compensation, and joint micro-increment actuation, making the method compatible with position-servo hardware. Hardware experiments on a truss-climbing prototype with six-axis wrist sensing and ground-based microgravity unloading validate execution-level load transfer and wrench regulation. Across full-support, walk, and trot gaits, the controller maintains bounded wrench tracking and post-switch recovery. Trot-gait ablations show that removing switching-aware reference conditioning or disturbance compensation increases the transition-window vertical-wrench RMSE by 31–33%. Allocation-layer ablations further show that removing the null-space penalty increases the current-based effort proxy under comparable $F_{z}$ tracking, while a conservative damped pseudoinverse allocation baseline shows higher effort than the proposed constrained allocation.
Förderung
- National Natural Science Foundation of China
- Postdoctoral Fellowship Program of CPSF
- China Postdoctoral Science Foundation
Zitieren
APA
Wu, S., Xu, K., Zhuang, X., Liu, H., Song, B., Chen, J., Zi, P., Zhang, T., & Ding, X. (2026). Execution-Level Internal-Load-Regularized Wrench Control for Position-Servo Space Robot in Closed-Chain Truss Climbing. IEEE Robotics and Automation Letters, 11(9), 10258-10265. https://doi.org/10.1109/lra.2026.3711367
Vancouver
Wu S, Xu K, Zhuang X, Liu H, Song B, Chen J, et al. Execution-Level Internal-Load-Regularized Wrench Control for Position-Servo Space Robot in Closed-Chain Truss Climbing. IEEE Robotics and Automation Letters. 2026;11(9):10258-10265. doi:10.1109/lra.2026.3711367
BibTeX
@article{Wu2026,
author = {Wu, Shutao and Xu, Kun and Zhuang, Xinghan and Liu, Haoyuan and Song, Bochao and Chen, Jiawei and Zi, Peijin and Zhang, Tao and Ding, Xilun},
title = {Execution-Level Internal-Load-Regularized Wrench Control for Position-Servo Space Robot in Closed-Chain Truss Climbing},
journal = {IEEE Robotics and Automation Letters},
year = {2026},
volume = {11},
number = {9},
pages = {10258--10265},
doi = {10.1109/lra.2026.3711367}
}
EUVIMED