Dynamic Manipulation of Flexible Objects with Torque Sequence Using a Deep Neural Network
This addresses robotic manipulation of flexible objects, an incremental advance in automation for tasks like handling fabrics or deformable materials.
The paper tackles dynamic manipulation of flexible objects by proposing a deep neural network to model motion equations and optimize torque commands, enabling control without a physics model. It demonstrated effectiveness on rigid objects, flexible objects with and without contact, and cloth.
For dynamic manipulation of flexible objects, we propose an acquisition method of a flexible object motion equation model using a deep neural network and a control method to realize a target state by calculating an optimized time-series joint torque command. By using the proposed method, any physics model of a target object is not needed, and the object can be controlled as intended. We applied this method to manipulations of a rigid object, a flexible object with and without environmental contact, and a cloth, and verified its effectiveness.