Evolution of humanoid locomotion control
Science Robotics · 2026-08-19 · 11 (117)
DOI: 10.1126/scirobotics.aed3973 ↗At a glance
Übersichtsarbeit in Science Robotics, published on 2026-08-19. Written by Yan Gu and 12 co-authors. Institution: School of Mechanical Engineering, Purdue University, West Lafayette, IN, USA. Major MEDLINE topics: Robotics, Locomotion. 291 references. Funded by: National Science Foundation, Office of Naval Research, NIH Office of the Director.
Abstract (original language)
Humanoid robots stand at the forefront of robotics, aiming to capture the agility, robustness, and expressivity of human movement in an anthropomorphic form. The locomotion control of humanoids has evolved from classical model–based methods to reinforcement learning powered by large-scale simulation and now to generative models that produce adaptive, whole-body behaviors, propelling humanoids toward operation in real-world environments. This survey positions humanoid control at a turning point, converging toward a unified paradigm of physics-guided generative intelligence that integrates optimization, learning, and predictive reasoning. We identify three core principles linking these paradigms: physics-based modeling, constrained decision-making, and adaptation to uncertainty. Building on these connections, we provide recommendations for researchers and outline open challenges in safety, accessibility, and human-level capability. These directions represent a transformation from engineered stability to intelligent autonomy, laying the groundwork for humanoid generalists capable of operating safely, collaborating naturally, and extending human capability in the open world.
Subjects
MeSH terms (MEDLINE)
Authors and institutions
- School of Mechanical Engineering, Purdue University, West Lafayette, IN, USA.
- Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA.
- Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore.
- Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA, USA.
- Tandon School of Engineering, New York University, New York, NY, USA.
- Department of Mechanical and Civil Engineering, California Institute of Technology, Pasadena, CA, USA.
ORCID profiles: Yan Gu, Guanya Shi, Fan Shi, I-Chia Chang, Yen-Jen Wang, Qilong Cheng, Ivan Lopez-Sanchez, Yunchu Feng, Aaron D Ames, Hao Su, Koushil Sreenath
Funding
- National Science Foundation
- Office of Naval Research
- NIH Office of the Director
Cite
APA
Gu, Y., Shi, G., Shi, F., Chang, I. C., Wang, Y. J., Cheng, Q., Olkin, Z., Lopez-Sanchez, I., Feng, Y., Zhang, J., Ames, A. D., Su, H., & Sreenath, K. (2026). Evolution of humanoid locomotion control. Science Robotics, 11(117). https://doi.org/10.1126/scirobotics.aed3973
Vancouver
Gu Y, Shi G, Shi F, Chang IC, Wang YJ, Cheng Q, et al. Evolution of humanoid locomotion control. Science Robotics. 2026;11(117). doi:10.1126/scirobotics.aed3973
BibTeX
@article{Gu2026,
author = {Gu, Yan and Shi, Guanya and Shi, Fan and Chang, I-Chia and Wang, Yen-Jen and Cheng, Qilong and Olkin, Zachary and Lopez-Sanchez, Ivan and Feng, Yunchu and Zhang, Jian and Ames, Aaron D and Su, Hao and Sreenath, Koushil},
title = {Evolution of humanoid locomotion control},
journal = {Science Robotics},
year = {2026},
volume = {11},
number = {117},
doi = {10.1126/scirobotics.aed3973}
}
EUVIMED