Failure Forecasting Boosts Robustness of Sim2Real Rhythmic Insertion Policies
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912472889294848 |
|---|---|
| author | Liu, Yuhan Zhang, Xinyu Chang, Haonan Boularias, Abdeslam |
| author_facet | Liu, Yuhan Zhang, Xinyu Chang, Haonan Boularias, Abdeslam |
| contents | This paper addresses the challenges of Rhythmic Insertion Tasks (RIT), where a robot must repeatedly perform high-precision insertions, such as screwing a nut into a bolt with a wrench. The inherent difficulty of RIT lies in achieving millimeter-level accuracy and maintaining consistent performance over multiple repetitions, particularly when factors like nut rotation and friction introduce additional complexity. We propose a sim-to-real framework that integrates a reinforcement learning-based insertion policy with a failure forecasting module. By representing the wrench's pose in the nut's coordinate frame rather than the robot's frame, our approach significantly enhances sim-to-real transferability. The insertion policy, trained in simulation, leverages real-time 6D pose tracking to execute precise alignment, insertion, and rotation maneuvers. Simultaneously, a neural network predicts potential execution failures, triggering a simple recovery mechanism that lifts the wrench and retries the insertion. Extensive experiments in both simulated and real-world environments demonstrate that our method not only achieves a high one-time success rate but also robustly maintains performance over long-horizon repetitive tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_06519 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Failure Forecasting Boosts Robustness of Sim2Real Rhythmic Insertion Policies Liu, Yuhan Zhang, Xinyu Chang, Haonan Boularias, Abdeslam Robotics Artificial Intelligence This paper addresses the challenges of Rhythmic Insertion Tasks (RIT), where a robot must repeatedly perform high-precision insertions, such as screwing a nut into a bolt with a wrench. The inherent difficulty of RIT lies in achieving millimeter-level accuracy and maintaining consistent performance over multiple repetitions, particularly when factors like nut rotation and friction introduce additional complexity. We propose a sim-to-real framework that integrates a reinforcement learning-based insertion policy with a failure forecasting module. By representing the wrench's pose in the nut's coordinate frame rather than the robot's frame, our approach significantly enhances sim-to-real transferability. The insertion policy, trained in simulation, leverages real-time 6D pose tracking to execute precise alignment, insertion, and rotation maneuvers. Simultaneously, a neural network predicts potential execution failures, triggering a simple recovery mechanism that lifts the wrench and retries the insertion. Extensive experiments in both simulated and real-world environments demonstrate that our method not only achieves a high one-time success rate but also robustly maintains performance over long-horizon repetitive tasks. |
| title | Failure Forecasting Boosts Robustness of Sim2Real Rhythmic Insertion Policies |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2507.06519 |