sim2real.world · gap index · about · arXiv:2611.07261
An Empirical Study of Reinforcement Learning for Keypoint Positioning of Deformable Linear Objects
- task
- Keypoint positioning of deformable linear objects
- hardware
- not stated
- simulator
- MuJoCo Playground
- transfer
- zero-shot
- sim result
- 96.4% for single-point tracking, 98.0% for disturbance recovery, and 88.8% for two-point tracking
- real result
- 9/10, 10/10, and 8/10 physical trials
The paper does not report a comparable success rate on both sides, so the gap cannot be computed.
What the paper says
Policies trained with observation noise, delay, domain randomization, and disturbances achieve simulation success rates of 96.4% for single-point tracking, 98.0% for disturbance recovery, and 88.8% for two-point tracking.
Using markerless RGB-D keypoint estimation, the policies transfer without real-world fine-tuning, succeeding in 9/10, 10/10, and 8/10 physical trials, respectively.
What it does not report
- real-world completion time
- perception keypoint error
- sim-to-real success-rate gap
Extraction is automated; every quote above was checked as a verbatim substring of the paper text. If a row is wrong, say so and the correction is recorded and credited: [email protected].