PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization
PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization
This paper presents PhysMoDPO, a Direct Preference Optimization framework for generating physically-plausible humanoid motion. By integrating a Whole-Body Controller into the training pipeline, the method ensures generated motions are compliant with both physics and text instructions. Experiments show improvements in physical realism and task performance on simulated and real robots.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S4
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
tput of WBC becomes compliant both with physics and original text instructions. To train PhysMoDPO we deploy physics-based and task-specific rewards and use them to assign preference to synthesized trajectories. Our extensive experiments on text-to-motion and spatial control tasks demonstrate consistent improvements of PhysMoDPO in both physical realism and task-related metrics on simulated robots. Moreover, we demon
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org