Reference Free Platform Adaptive Locomotion for Quadrupedal Robots using a Dynamics Conditioned Policy
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913935539568640 |
|---|---|
| author | Rytz, David Choi, Suyoung Yu, Wanming Merkt, Wolfgang Hwangbo, Jemin Havoutis, Ioannis |
| author_facet | Rytz, David Choi, Suyoung Yu, Wanming Merkt, Wolfgang Hwangbo, Jemin Havoutis, Ioannis |
| contents | This article presents Platform Adaptive Locomotion (PAL), a unified control method for quadrupedal robots with different morphologies and dynamics. We leverage deep reinforcement learning to train a single locomotion policy on procedurally generated robots. The policy maps proprioceptive robot state information and base velocity commands into desired joint actuation targets, which are conditioned using a latent embedding of the temporally local system dynamics. We explore two conditioning strategies - one using a GRU-based dynamics encoder and another using a morphology-based property estimator - and show that morphology-aware conditioning outperforms temporal dynamics encoding regarding velocity task tracking for our hardware test on ANYmal C. Our results demonstrate that both approaches achieve robust zero-shot transfer across multiple unseen simulated quadrupeds. Furthermore, we demonstrate the need for careful robot reference modelling during training: exposing the policy to a diverse set of robot morphologies and dynamics leads to improved generalization, reducing the velocity tracking error by up to 30% compared to the baseline method. Despite PAL not surpassing the best-performing reference-free controller in all cases, our analysis uncovers critical design choices and informs improvements to the state of the art. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_16042 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Reference Free Platform Adaptive Locomotion for Quadrupedal Robots using a Dynamics Conditioned Policy Rytz, David Choi, Suyoung Yu, Wanming Merkt, Wolfgang Hwangbo, Jemin Havoutis, Ioannis Robotics This article presents Platform Adaptive Locomotion (PAL), a unified control method for quadrupedal robots with different morphologies and dynamics. We leverage deep reinforcement learning to train a single locomotion policy on procedurally generated robots. The policy maps proprioceptive robot state information and base velocity commands into desired joint actuation targets, which are conditioned using a latent embedding of the temporally local system dynamics. We explore two conditioning strategies - one using a GRU-based dynamics encoder and another using a morphology-based property estimator - and show that morphology-aware conditioning outperforms temporal dynamics encoding regarding velocity task tracking for our hardware test on ANYmal C. Our results demonstrate that both approaches achieve robust zero-shot transfer across multiple unseen simulated quadrupeds. Furthermore, we demonstrate the need for careful robot reference modelling during training: exposing the policy to a diverse set of robot morphologies and dynamics leads to improved generalization, reducing the velocity tracking error by up to 30% compared to the baseline method. Despite PAL not surpassing the best-performing reference-free controller in all cases, our analysis uncovers critical design choices and informs improvements to the state of the art. |
| title | Reference Free Platform Adaptive Locomotion for Quadrupedal Robots using a Dynamics Conditioned Policy |
| topic | Robotics |
| url | https://arxiv.org/abs/2505.16042 |