DLO Dynamic Manipulation and Vibration Suppression through Reinforcement Learning

The dynamic robotic manipulation of deformable linear objects (DLOs) such as cables is currently a challenge. In this work, we propose a reinforcement learning framework called DynDLO, aimed at learning policies to dynamically manipulate the DLO and suppress the vibration caused by high-speed movements. The trained policies are then deployed on a robot to successfully execute the task in the real world.