Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion

Alexander Luis Mitchell, Wolfgang Merkt, Aristotelis Papatheodorou, Ioannis Havoutis, Ingmar Posner
Proceedings of The 8th Conference on Robot Learning, PMLR 270:780-793, 2025.

Abstract

The current state-of-the-art in quadruped locomotion is able to produce a variety of complex motions. These methods either rely on switching between a discrete set of skills or learn a distribution across gaits using complex black-box models. Alternatively, we present Gaitor, which learns a disentangled and 2D representation across locomotion gaits. This learnt representation forms a planning space for closed-loop control delivering continuous gait transitions and perceptive terrain traversal. Gaitor’s latent space is readily interpretable and we discover that during gait transitions, novel unseen gaits emerge. The latent space is disentangled with respect to footswing heights and lengths. This means that these gait characteristics can be varied independently in the 2D latent representation. Together with a simple terrain encoding and a learnt planner operating in the latent space, Gaitor can take motion commands including desired gait type and swing characteristics all while reacting to uneven terrain. We evaluate Gaitor in both simulation and the real world on the ANYmal C platform. To the best of our knowledge, this is the first work learning a unified and interpretable latent space for multiple gaits, resulting in continuous blending between different locomotion modes on a real quadruped robot. An overview of the methods and results in this paper is found at https://youtu.be/eVFQbRyilCA.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-mitchell25a, title = {Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion}, author = {Mitchell, Alexander Luis and Merkt, Wolfgang and Papatheodorou, Aristotelis and Havoutis, Ioannis and Posner, Ingmar}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {780--793}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/mitchell25a/mitchell25a.pdf}, url = {https://proceedings.mlr.press/v270/mitchell25a.html}, abstract = {The current state-of-the-art in quadruped locomotion is able to produce a variety of complex motions. These methods either rely on switching between a discrete set of skills or learn a distribution across gaits using complex black-box models. Alternatively, we present Gaitor, which learns a disentangled and 2D representation across locomotion gaits. This learnt representation forms a planning space for closed-loop control delivering continuous gait transitions and perceptive terrain traversal. Gaitor’s latent space is readily interpretable and we discover that during gait transitions, novel unseen gaits emerge. The latent space is disentangled with respect to footswing heights and lengths. This means that these gait characteristics can be varied independently in the 2D latent representation. Together with a simple terrain encoding and a learnt planner operating in the latent space, Gaitor can take motion commands including desired gait type and swing characteristics all while reacting to uneven terrain. We evaluate Gaitor in both simulation and the real world on the ANYmal C platform. To the best of our knowledge, this is the first work learning a unified and interpretable latent space for multiple gaits, resulting in continuous blending between different locomotion modes on a real quadruped robot. An overview of the methods and results in this paper is found at https://youtu.be/eVFQbRyilCA.} }
Endnote
%0 Conference Paper %T Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion %A Alexander Luis Mitchell %A Wolfgang Merkt %A Aristotelis Papatheodorou %A Ioannis Havoutis %A Ingmar Posner %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-mitchell25a %I PMLR %P 780--793 %U https://proceedings.mlr.press/v270/mitchell25a.html %V 270 %X The current state-of-the-art in quadruped locomotion is able to produce a variety of complex motions. These methods either rely on switching between a discrete set of skills or learn a distribution across gaits using complex black-box models. Alternatively, we present Gaitor, which learns a disentangled and 2D representation across locomotion gaits. This learnt representation forms a planning space for closed-loop control delivering continuous gait transitions and perceptive terrain traversal. Gaitor’s latent space is readily interpretable and we discover that during gait transitions, novel unseen gaits emerge. The latent space is disentangled with respect to footswing heights and lengths. This means that these gait characteristics can be varied independently in the 2D latent representation. Together with a simple terrain encoding and a learnt planner operating in the latent space, Gaitor can take motion commands including desired gait type and swing characteristics all while reacting to uneven terrain. We evaluate Gaitor in both simulation and the real world on the ANYmal C platform. To the best of our knowledge, this is the first work learning a unified and interpretable latent space for multiple gaits, resulting in continuous blending between different locomotion modes on a real quadruped robot. An overview of the methods and results in this paper is found at https://youtu.be/eVFQbRyilCA.
APA
Mitchell, A.L., Merkt, W., Papatheodorou, A., Havoutis, I. & Posner, I.. (2025). Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:780-793 Available from https://proceedings.mlr.press/v270/mitchell25a.html.

Related Material