ODE Policy: Robot Learning via Action Dynamics, Not Actions
ODE Policy: Robot Learning via Action Dynamics, Not Actions
This paper introduces ODE Policy, a robot learning framework that models the dynamics of actions using neural ordinary differential equations to ensure continuous motion trajectories. Evaluated on unimanual, bimanual, dexterous, locomotion, and humanoid tasks, it outperforms baselines in success rate, generalization, and generation speed under both imitation learning and reinforcement learning settings.
The authors propose ODE Policy, a method that learns continuous action dynamics via neural ordinary differential equations rather than discrete action values, achieving higher success rates and faster generation speeds in both imitation learning and reinforcement learning benchmarks across unimanual, bimanual, dexterous, locomotion, and humanoid tasks.
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.
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
hly than human demonstrations. On 14 real-world tasks across single-arm, dual-arm, dexterous, and humanoid robots, it achieves the highest success rate. In RL, it further speeds up SuperMimic for humanoid motion tracking and smooths its actions. Efficient in learning, data, parameters, and execution, ODE policy offers a promising path to scaling robot learning.
Open source S5
n head of vision-language-action (VLA) and world action models (WAMs), and as the actor of model-based RL against 14 baselines on unimanual, bimanual, dexterous, locomotion, and loco-manipulation tasks. In IL, against flow matching, it reaches 5.9$\times$ the success, 5.4$\times$ the generalization, and 3.9$\times$ the mode balance after one epoch, and generates actions 10.8$\times$ faster. It mostly moves more smoot
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org