Event
Autonomous Data Collection and Offline Reinforcement Learning Method for Local Contact-Rich Manipulation
1 reports · 1 sources
Oliver Hausdörfer et al. propose a modular method: autonomously collect data in local segments of contact-rich tasks, train local policies with offline reinforcement learning, then integrate with traditional motion planning to complete long-horizon assembly tasks. Across four real tasks, using only 2-2.5 hours of autonomously collected data, the average success rate on long-horizon tasks reached 96%, compared with 55% for the planning-only baseline and 37% for the strongest end-to-end baseline; the local policies also reduced applied force by 49%. Performance remained at 90% in scene-level out-of-distribution scenarios, with over 1500 real-world evaluation rollouts completed in total.
Report timeline
- Autonomous Data Collection and Offline Reinforcement Learning for Local Contact-Rich Manipulation Tasks
The paper proposes a modular method that autonomously collects data in local segments of contact-rich tasks and uses offline reinforcement learning to train local policies, which are then integrated with traditional motion planning to complete long-horizon assembly tasks. Across four real-world tasks, using only 2-2.5 hours of autonomous data collection, the method achieved an average success rate of 96% on long-horizon tasks, compared with 55% for a planning-only baseline and 37% for the strongest end-to-end baseline; the local policies also reduced applied force by 49%. Performance remained at 90% in scene-level out-of-distribution scenarios, with more than 1,500 real-world evaluation rollouts completed in total.
arXiv cs.RO ·