.png){kind=icon}
Return
DiffOG: Differentiable Policy Trajectory Optimization With Generalizability
.png){kind=icon}
PRE
AI
DOI：10.1109/TRO.2025.3636302.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Imitation-learning-based visuomotor policies excel at manipulation tasks but often produce suboptimal action trajectories compared to model-based methods. Directly mapping camera data to actions via neural networks can result in jerky motions and difficulties in meeting critical constraints, compromising safety and robustness in real-world deployment. For tasks that require high robustness or strict adherence to constraints, ensuring trajectory quality is crucial. However, the lack of interpretability in neural networks makes it challenging to generate constraint-compliant actions in a controlled manner. This article introduces differentiable policy trajectory optimization with generalizability (DiffOG), a learning-based trajectory optimization framework designed to enhance visuomotor policies. By leveraging the proposed differentiable formulation of trajectory optimization with transformer, DiffOG seamlessly integrates policies with a generalizable optimization layer. DiffOG refines action trajectories to be smoother and more constraint-compliant while maintaining alignment with the original demonstration distribution, thus avoiding degradation in policy performance. We evaluated DiffOG across 11 simulated tasks and two real-world tasks. The results demonstrate that DiffOG significantly enhances the trajectory quality of visuomotor policies while having minimal impact on policy performance, outperforming trajectory processing baselines such as greedy constraint clipping and penalty-based trajectory optimization. Furthermore, DiffOG achieves superior performance compared to existing constrained visuomotor policy.
Keywords:
Differentiable optimization
imitation learning
robot learning
Journal
IF:
10.5
Papers:
3.3K
Citations:
2.8W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSave
View more