[2603.01953] Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy

Computer Science > Robotics arXiv:2603.01953 (cs) [Submitted on 2 Mar 2026] Title:Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy Authors:Pengyuan Wu, Pingrui Zhang, Zhigang Wang, Dong Wang, Bin Zhao, Xuelong Li View a PDF of the paper titled Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy, by Pengyuan Wu and 4 other authors View PDF HTML (experimental) Abstract:Diffusion-based policies have achieved remarkable results in robotic manipulation but often struggle to adapt rapidly in dynamic scenarios, leading to delayed responses or task failures. We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction. DCDP integrates a self-supervised dynamic feature encoder, cross-attention fusion, and an asymmetric action encoder-decoder to inject environmental dynamics before action execution, achieving real-time closed-loop action correction and enhancing the system's adaptability in dynamic scenarios. In dynamic PushT simulations, DCDP improves adaptability by 19\% without retraining while requiring only 5\% additional computation. Its modular design enables plug-and-play integration, achieving both temporal coherence and real-time responsiveness in dynamic robotic scenarios, including real-world manipulation tasks. The project page is at: this https URL Comments: Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI); Compu...