PRVR: Learning Posture Recovery for a Legged Spherical Robot Using Vector Reward
IEEE Robotics and Automation Letters · 2026-09 · 11 (9) · 10489-10496
DOI: 10.1109/lra.2026.3712643 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Robotics and Automation Letters, 2026-09. Verfasst von Jinkai Wang und 12 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Legged spherical robots feature both walking and rolling modes and they are useful in explorations, but often end up in various body-inverted postures after rolling and accidental falls. Posture recovery in such cases where all legs are in the air and lose leg-ground contacts is an open and challenging problem. We propose a posture recovery learning method using vector reward called PRVR. A traditional scalar reward cannot evaluate each element of a vector action. Here, a vector reward method is proposed to evaluate each element of a vector action, such that the action of each joint closely conforms to the expected behavior. To reduce learning complexity, a rotation symmetry-based training and deployment method is proposed. Body-inverted postures can be divided into several symmetric classes, and then through training on one class of body-inverted postures, the robot can recover from various body-inverted postures. Simulations and experiments on a six-legged spherical robot are used to verify the effectiveness.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Wang, J., Xu, X., Qi, C., Zhang, C., Shou, X., Han, L., Wang, W., Ma, Y., Ma, B., Chen, J., Li, C., Ling, Z., & Zhang, X. (2026). PRVR: Learning Posture Recovery for a Legged Spherical Robot Using Vector Reward. IEEE Robotics and Automation Letters, 11(9), 10489-10496. https://doi.org/10.1109/lra.2026.3712643
Vancouver
Wang J, Xu X, Qi C, Zhang C, Shou X, Han L, et al. PRVR: Learning Posture Recovery for a Legged Spherical Robot Using Vector Reward. IEEE Robotics and Automation Letters. 2026;11(9):10489-10496. doi:10.1109/lra.2026.3712643
BibTeX
@article{Wang2026,
author = {Wang, Jinkai and Xu, Xin and Qi, Chenkun and Zhang, Chongfeng and Shou, Xing and Han, Liangliang and Wang, Weijun and Ma, Yiqin and Ma, Bida and Chen, Jiayi and Li, Chengjie and Ling, Zhenfang and Zhang, Xiaoyuan},
title = {PRVR: Learning Posture Recovery for a Legged Spherical Robot Using Vector Reward},
journal = {IEEE Robotics and Automation Letters},
year = {2026},
volume = {11},
number = {9},
pages = {10489--10496},
doi = {10.1109/lra.2026.3712643}
}
EUVIMED