DreamControl-v2: Simpler and Scalable Autonomous Humanoid Skills via Trainable Guided Diffusion Priors
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912993519861760 |
|---|---|
| author | Harithas, Sudarshan Kwak, Sangkyung Katara, Pushkal Deolasee, Srujan Kalaria, Dvij Sridhar, Srinath Vemprala, Sai Kapoor, Ashish Huang, Jonathan Chung-Kuan |
| author_facet | Harithas, Sudarshan Kwak, Sangkyung Katara, Pushkal Deolasee, Srujan Kalaria, Dvij Sridhar, Srinath Vemprala, Sai Kapoor, Ashish Huang, Jonathan Chung-Kuan |
| contents | Developing robust autonomous loco-manipulation skills for humanoids remains an open problem in robotics. While RL has been applied successfully to legged locomotion, applying it to complex, interaction-rich manipulation tasks is harder given long-horizon planning challenges for manipulation. A recent approach along these lines is DreamControl, which addresses these issues by leveraging off-the-shelf human motion diffusion models as a generative prior to guide RL policies during training. In this paper, we investigate the impact of DreamControl's motion prior and propose an improved framework that trains a guided diffusion model directly in the humanoid robot's motion space, aggregating diverse human and robot datasets into a unified embodiment space. We demonstrate that our approach captures a wider range of skills due to the larger training data mixture and establishes a more automated pipeline by removing the need for manual filtering interventions. Furthermore, we show that scaling the generation of reference trajectories is important for achieving robust downstream RL policies. We validate our approach through extensive experiments in simulation and on a real Unitree-G1. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_00202 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DreamControl-v2: Simpler and Scalable Autonomous Humanoid Skills via Trainable Guided Diffusion Priors Harithas, Sudarshan Kwak, Sangkyung Katara, Pushkal Deolasee, Srujan Kalaria, Dvij Sridhar, Srinath Vemprala, Sai Kapoor, Ashish Huang, Jonathan Chung-Kuan Robotics Developing robust autonomous loco-manipulation skills for humanoids remains an open problem in robotics. While RL has been applied successfully to legged locomotion, applying it to complex, interaction-rich manipulation tasks is harder given long-horizon planning challenges for manipulation. A recent approach along these lines is DreamControl, which addresses these issues by leveraging off-the-shelf human motion diffusion models as a generative prior to guide RL policies during training. In this paper, we investigate the impact of DreamControl's motion prior and propose an improved framework that trains a guided diffusion model directly in the humanoid robot's motion space, aggregating diverse human and robot datasets into a unified embodiment space. We demonstrate that our approach captures a wider range of skills due to the larger training data mixture and establishes a more automated pipeline by removing the need for manual filtering interventions. Furthermore, we show that scaling the generation of reference trajectories is important for achieving robust downstream RL policies. We validate our approach through extensive experiments in simulation and on a real Unitree-G1. |
| title | DreamControl-v2: Simpler and Scalable Autonomous Humanoid Skills via Trainable Guided Diffusion Priors |
| topic | Robotics |
| url | https://arxiv.org/abs/2604.00202 |