From insect behavior to transferable robot locomotion: inferring embodied locomotor principles from limited data via adversarial inverse reinforcement learning
Bioinspiration & Biomimetics · 31.08.2026 · 21 (4) · 046034
DOI: 10.1088/1748-3190/ae901f ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinspiration & Biomimetics, online erschienen am 11.08.2026. Verfasst von Yuchen Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Robotics , Graduate School of Engineering, Tohoku University, Sendai 980-8579, Japan. Hauptthemen laut MEDLINE: Robotics, Insecta, Locomotion, Biomimetics. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Ministry of Higher Education, Science, Research, and Innovation of Thailand under the Re-inventing University program, a Grant-in-Aid for Scientific Research, JSPS KAKENHI a Grant-in-Aid for JSPS Fellows.
Abstract (Originalsprache)
Insect locomotion exhibits remarkable adaptability and flexibility despite the limited scale of its nervous system. However, the underlying principles that govern leg coordination remain difficult to extract and model computationally. Understanding how insects achieve stable and adaptive locomotion has long provided important inspiration for the development of control strategies in bio-inspired robotics. Nevertheless, many existing approaches rely on predefined coordination rules, manually tuned parameters, or hand-crafted reward functions, which limit the flexibility and transferability of the resulting control strategies. To address this limitation, this study proposes a data-driven framework based on adversarial inverse reinforcement learning , which directly learns continuous locomotion control policies from stick insect walking data and infers latent reward structures and control strategies from biological behavioral demonstrations. Experimental results show that even when trained using only a short segment of flat-terrain demonstration data, the learned policy can still be extracted to learn adaptive and flexible leg coordination patterns under different environmental conditions. Furthermore, the learned reward network can be transferred across different dynamic systems to guide policy learning for robot models with different morphologies. Compared with methods relying solely on reward shaping, the proposed approach achieves faster convergence and produces more biologically consistent gait coordination. A preliminary deployment on a physical bio-inspired robot further demonstrates the potential of the learned policy for sim-to-real application. The proposed method provides a transferable data-driven framework for extracting and learning locomotion control strategies from biological behavior, with potential applications to bio-inspired robotic systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Robotics , Graduate School of Engineering, Tohoku University, Sendai 980-8579, Japan.
- Research Fellow of Japan Society for the Promotion of Science, Tokyo 102-0083, Japan.
- Bio-inspired Robotics and Neural Engineering Lab, School of Information Science and Technology, Vidyasirimedhi Institute of Science and Technology, Rayong 21210, Thailand.
- SDU Biorobotics, The Mærsk Mc-Kinney Møller Institute, University of Southern Denmark, Odense 5230, Denmark.
ORCID-Profile: Yuchen Wang, Thirawat Chuthong, Mitsuhiro Hayashibe, Poramate Manoonpong, Dai Owaki
Förderung
- the Ministry of Higher Education, Science, Research, and Innovation of Thailand under the Re-inventing University program
- a Grant-in-Aid for Scientific Research
- JSPS KAKENHI a Grant-in-Aid for JSPS Fellows
Zitieren
APA
Wang, Y., Chuthong, T., Hayashibe, M., Manoonpong, P., & Owaki, D. (2026). From insect behavior to transferable robot locomotion: inferring embodied locomotor principles from limited data via adversarial inverse reinforcement learning. Bioinspiration & Biomimetics, 21(4), 046034. https://doi.org/10.1088/1748-3190/ae901f
Vancouver
Wang Y, Chuthong T, Hayashibe M, Manoonpong P, Owaki D. From insect behavior to transferable robot locomotion: inferring embodied locomotor principles from limited data via adversarial inverse reinforcement learning. Bioinspiration & Biomimetics. 2026;21(4):046034. doi:10.1088/1748-3190/ae901f
BibTeX
@article{Wang2026,
author = {Wang, Yuchen and Chuthong, Thirawat and Hayashibe, Mitsuhiro and Manoonpong, Poramate and Owaki, Dai},
title = {From insect behavior to transferable robot locomotion: inferring embodied locomotor principles from limited data via adversarial inverse reinforcement learning},
journal = {Bioinspiration & Biomimetics},
year = {2026},
volume = {21},
number = {4},
pages = {046034},
doi = {10.1088/1748-3190/ae901f}
}
EUVIMED