以迭代式事後重標記學習可控執行品質的機械人策略
Learning Quality-Controllable Robot Policies with Iterative Hindsight Relabeling
Quality-Controllable Policy(QCP)讓機械人策略在推理時控制執行速度、動作流暢度等品質屬性,並可按任務需要調整這些屬性。研究提出迭代式事後重標記流程,利用線上執行軌跡更新數據集並微調策略;在 Robomimic、RoboEval 及真實操作任務中,QCP 逐輪提升執行品質和成功率。
Published on Oct 27
Authors:
,
,
,
,
,
,
,
Abstract
Quality-Controllable Policy enables real-time adjustment of robot execution quality and improves performance through iterative hindsight relabeling of online rollouts.
Robot policy evaluation often relies on a binary success signal for task completion. In real-world deployment, however, desired execution behaviors are also defined by quality attributes such as execution speed and motion smoothness, which need to be dynamically controlled according to task requirements. To address this, we introduce Quality-Controllable Policy (QCP), a robot policy that enables inference-time control over both individual quality attributes and their combinations. To extend the controllable quality range beyond initial demonstrations, we further propose an iterative hindsight relabeling loop: QCP collects online rollouts under target quality labels, adds successful trajectories to the dataset, relabels the accumulated trajectories in hindsight according to their measured quality, and fine-tunes the policy. Experiments on Robomimic and RoboEval demonstrate that QCP consistently improves both task execution quality and success rates across iterations, producing successful rollouts that exceed the quality of the initial demonstrations. Finally, we validate QCP on real-world manipulation tasks, where controlling speed increases throughput in conveyor operations and controlling transport stability improves success when carrying wine glasses, demonstrating that quality-controllable policies yield practical gains in physical deployment.
View arXiv page View PDF Project page Add to collection
Get this paper in your agent:
hf papers read 2610.37393
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.37393 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.37393 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.37393 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co