Event
StabBot Pipe Alignment System Paper Published
1 reports · 1 sources
Mohamed Ashraf Mostafa and other authors submitted a paper proposing the StabBot robotic pipe alignment system: it uses a pendulum Kalman filter to track the pipe joint in stereo images, approaches at a matched speed, and confirms the final few millimeters via touch. The authors claim that in 200 randomized trials, covering nine sizes from 2-3/8-inch tubing to 20-inch casing, StabBot completed alignment in all cases, and the pipe contacted only the rollers, versus 74% for a conventional fixed guide. The authors report that the median contact speed dropped from 18.4 cm/s to 2.3 cm/s, and the median peak head-pipe force dropped from 11.3 kN to 1.4 kN. All the above results come from a 1:4 Froude-scaled MuJoCo model and are reported scaled to full size.
Report timeline
- StabBot: Robotic Pipe Stabbing with Learned Stereo Vision and Residual Reinforcement Learning
The StabBot robotic stabbing system uses a pendulum Kalman filter to track pipe joints in stereo images, approaches at matched speed, and confirms the final few millimeters by touch. In 200 randomized jobs, StabBot completed every stab, versus 74% for a conventional fixed guide; median contact speed fell from 18.4 cm/s to 2.3 cm/s, and median peak head-pipe force fell from 11.3 kN to 1.4 kN.
arXiv cs.RO ·