Learning Where it Matters: Autonomous Data Collection and Offline Reinforcement Learning for Local Contact-Rich Manipulation
Learning Where it Matters: Autonomous Data Collection and Offline Reinforcement Learning for Local Contact-Rich Manipulation
This research proposes a modular approach that combines autonomous data collection with offline reinforcement learning to improve performance in contact-rich manipulation tasks. By collecting dense data locally and integrating it with traditional motion planning, the method achieves a 96% success rate in long-horizon tasks, outperforming planning-only and end-to-end baselines. The approach also reduces applied forces by 49% and maintains 90% performance in out-of-distribution scenarios.
This paper addresses the challenge of high-precision contact-rich manipulation by combining autonomous data collection with offline reinforcement learning. The approach improves success rates and reduces applied forces in real-world tasks, demonstrating robustness to out-of-distribution scenarios.
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.
- Control
- Reported autonomousOpen source S3
Reported numbers
Reported success rate
96%
View original evidence
e translated into policies with high success rates. We then integrate our local policies with traditional motion planning to solve long-horizon assembly tasks. Across four real-world tasks, using only 2-2.5 h of autonomous data collection, we achieve an average success rate of 96% for long-horizon tasks, compared to 55% for a planning-only baseline and 37% for the strongest end-to-end baseline. Our local policies als
Open source S4real-world evaluation rollouts
>1,500
View original evidence
o reduce applied forces by 49%. Notably, performance remains at 90% for scene-level out-of-distribution scenarios, while end-to-end approaches struggle. In total, we run over 1500 real-world evaluation rollouts. Our results pave the way for targeted data collection in contact-rich tasks and for high success rates in precision applications. Videos and qualitative results are available at https://learnsyslab.github.io/
Open source S5
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ight-tolerance tasks. To reliably solve challenging contact-rich tasks, we propose to leverage robot data and learning where it matters in a modular approach. First, we introduce an autonomous data collection scheme for the real world that collects dense data locally in the contact-rich segment of the task, eliminating reliance on a teleoperator's skill. Second, we show that using offline RL, the collected data can b
Open source S3
e translated into policies with high success rates. We then integrate our local policies with traditional motion planning to solve long-horizon assembly tasks. Across four real-world tasks, using only 2-2.5 h of autonomous data collection, we achieve an average success rate of 96% for long-horizon tasks, compared to 55% for a planning-only baseline and 37% for the strongest end-to-end baseline. Our local policies als
Open source S4
o reduce applied forces by 49%. Notably, performance remains at 90% for scene-level out-of-distribution scenarios, while end-to-end approaches struggle. In total, we run over 1500 real-world evaluation rollouts. Our results pave the way for targeted data collection in contact-rich tasks and for high success rates in precision applications. Videos and qualitative results are available at https://learnsyslab.github.io/
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org