Multi-Embodiment Locomotion at Scale with extreme Embodiment Randomization

Nico Bohlinger, Jan Peters
9/2/2025
cs.ROcs.LG

No confident code implementation found for this paper.

Abstract

We present a single, general locomotion policy trained on a diverse collection of 50 legged robots. By combining an improved embodiment-aware architecture (URMAv2) with a performance-based curriculum for extreme Embodiment Randomization, our policy learns to control millions of morphological variations. Our policy achieves zero-shot transfer to unseen real-world humanoid and quadruped robots.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Towards Embodiment Scaling Laws in Robot Locomotion

89% match2025

One Policy to Run Them All: an End-to-end Learning Approach to Multi-Embodiment Locomotion

89% match1 repo2222024

ManyQuadrupeds: Learning a Single Locomotion Policy for Diverse Quadruped Robots

84% match2023

The One RING: a Robotic Indoor Navigation Generalist

83% match2024

The Role of Domain Randomization in Training Diffusion Policies for Whole-Body Humanoid Control

82% match2024

Cross-Embodiment Offline Reinforcement Learning for Heterogeneous Robot Datasets

82% match1 repo42026

Cite this paper

@article{bohlinger2025multiembodiment,
  title  = {Multi-Embodiment Locomotion at Scale with extreme Embodiment Randomization},
  author = {Nico Bohlinger and Jan Peters},
  year   = {2025},
  eprint = {2509.02815},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.02815v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.02815.svg)](https://codeofpaper.com/paper/2509.02815)

Discussion