analiza aiqbench
This research introduces Dynamic Execution Horizon Prediction (DEHP), a lightweight reinforcement learning module that dynamically adjusts action chunk sizes for robot policies without retraining the underlying model. By predicting shorter horizons for fine-grained manipulation and longer ones for free-space motion, the method significantly improves success rates in high-precision tasks while maintaining computational efficiency. The approach offers a practical solution for integrating adaptive planning into existing black-box chunk-based architectures like diffusion or flow policies.
Dynamic Execution Horizon Prediction for Chunk-based Robot Policies
arXiv:2606.11408v1 Announce Type: new Abstract: Action chunking has become a standard design in modern robot policies, from diffusion/flow policies to vision-language-action models, where the policy predicts a sequence of actions and executes a fixed number of them instead of acting one step at a time. However, this paradigm relies on a key assumption: a fixed execution horizon. During chunk execution, the policy operates open-loop, which is particularly problematic for fine-grained manipulation tasks that require frequent replanning. In practice, the execution horizon is typically chosen through empirical tuning and is highly task-dependent. To this end, we propose Dynamic Execution Horizon Prediction (DEHP), an effective method that trains a lightweight execution-horizon prediction branch using online reinforcement learning while keeping the pretrained chunk policy completely frozen. This makes the method compatible with black-box chunk policies and isolates the effect of adapting the execution horizon from changes to the underlying action generator. Across our evaluations, DEHP improves the success rate of different high-precision and long-horizon manipulation tasks by a large margin. Our qualitative analysis further shows that DEHP predicts shorter execution horizons during fine-grained stages of the task and longer horizons during free-space motion. In this way, DEHP balances the efficiency of open-loop chunk execution with the reactivity of closed-loop single-step control. Project page: https://dehp-chunking.github.io/
Abstract:Action chunking has become a standard design in modern robot policies, from diffusion/flow policies to vision-language-action models, where the policy predicts a sequence of actions and executes a fixed number of them instead of acting one step at a time. However, this paradigm relies on a key assumption: a fixed execution horizon. During chunk execution, the policy operates open-loop, which is particularly problematic for fine-grained manipulation tasks that require frequent replanning. In practice, the execution horizon is typically chosen through empirical tuning and is highly task-dependent. To this end, we propose Dynamic Execution Horizon Prediction (DEHP), an effective method that trains a lightweight execution-horizon prediction branch using online reinforcement learning while keeping the pretrained chunk policy completely frozen. This makes the method compatible with black-box chunk policies and isolates the effect of adapting the execution horizon from changes to the underlying action generator. Across our evaluations, DEHP improves the success rate of different high-precision and long-horizon manipulation tasks by a large margin. Our qualitative analysis further shows that DEHP predicts shorter execution horizons during fine-grained stages of the task and longer horizons during free-space motion. In this way, DEHP balances the efficiency of open-loop chunk execution with the reactivity of closed-loop single-step control. Project page: this https URL
Robotics (cs.RO)
arXiv:2606.11408 [cs.RO]
(or arXiv:2606.11408v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.11408 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 301 cuvinte
Subiecte conexe
Building a Foundation Stack for General-Purpose Robots
IEEE Spectrum · 13 iul. 2026
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · 4 sept. 2026
Knowing When to Stop: Adaptive Action Chunking via Internal Cross-Attention Dynamics in VLAs
arXiv · 2 sept. 2026
PATCH: Action-Chunk-Conditioned Latent Patch Innovation Monitoring for Robot Manipulation
arXiv · 16 iun. 2026