Policy gradient reinforcement learning for fast quadrupedal locomotion


This paper presents a machine learning approach to optimizing a quadrupedal trot gait for forward speed. Given a parameterized walk designed for a specific robot, we propose using a form of policy gradient reinforcement learning to automatically search the set of possible parameters with the goal of finding the fastest possible walk. We implement and test… (More)
DOI: 10.1109/ROBOT.2004.1307456


7 Figures and Tables


Citations per Year

385 Citations

Semantic Scholar estimates that this publication has 385 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{Kohl2004PolicyGR, title={Policy gradient reinforcement learning for fast quadrupedal locomotion}, author={Nate Kohl and Peter Stone}, journal={IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004}, year={2004}, volume={3}, pages={2619-2624 Vol.3} }