🤖 AI Summary
This work addresses the challenge of balancing accuracy and efficiency in robotic physical simulation of deformable linear objects (DLOs). We propose a high-fidelity simulation framework that deeply integrates the discrete elastic rods (DER) model with MuJoCo. Methodologically, we are the first to embed the DER model directly into the MuJoCo engine, enabling geometrically and dynamically consistent DLO modeling; we further combine open-loop policy optimization with end-to-end simulation-to-real transfer, achieving dynamic DLO manipulation on real robots using simulation-only training. Our key contribution is a sim-to-real pipeline that requires no real-world fine-tuning: it successfully accomplishes precise wire throwing—achieving a 76.7% success rate on physical hardware—while significantly reducing computational overhead and markedly improving policy generalization from simulation to reality.
📝 Abstract
We show that it is possible to learn an open-loop policy in simulation for the dynamic manipulation of a deformable linear object (DLO) -- e.g., a rope, wire, or cable -- that can be executed by a real robot without additional training. Our method is enabled by integrating an existing state-of-the-art DLO model (Discrete Elastic Rods) with MuJoCo, a robot simulator. We describe how this integration was done, check that validation results produced in simulation match what we expect from analysis of the physics, and apply policy optimization to train an open-loop policy from data collected only in simulation that uses a robot arm to fling a wire precisely between two obstacles. This policy achieves a success rate of 76.7% when executed by a real robot in hardware experiments without additional training on the real task.
