Point It, Strike It: Direction-Conditioned Dynamic Manipulation of Deformable Linear Objects

Key points

• We therefore study single-swing rope striking with goals that specify the tip's 3D position and arrival direction, across the workspace and on different ropes.1
• To address these challenges, we extend the state-of-the-art DLO simulator DeformX with GPU acceleration, a stable Cosserat rod solver, and a cross-flow aerodynamic model, yielding DeformX2.0, which is more than $20{,}000\times$ faster.1
• We then propose TRACE (Trace-rooted Adaptive Cross-Entropy), which generates striking data by warm-starting each new target from the stored swing whose tip path passes closest to it.1
• Finally, we propose RECAP (Residual Calibration Policy), which fits the simulator's rope and rig parameters to a few calibration swings and adapts actions with a correction policy trained in simulation.1

Sources (1)

• [1]Point It, Strike It: Direction-Conditioned Dynamic Manipulation of Deformable Linear Objects
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 07:15 AM

  Goal-conditioned dynamic manipulation of deformable linear objects has mainly specified goals as positions for a rope tip to reach.

  We therefore study single-swing rope striking with goals that specify the tip's 3D position and arrival direction, across the workspace and on different ropes.

Extractive summary: sentences quoted from the sources.