Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : May 29, 2024 - June 01, 2024
This paper proposes a vision-based method for estimating the deformation of a soft robotic hand before grasping an object. Due to the softness of a soft robotic hand, it can adapt and fit to the shape of objects, which enables stable grasping of various objects. However, because of the grasping position and object shape, the hand may not grasp an object in a stable manner. Therefore, it is important to calculate whether the hand can adapt to the object shape prior to grasp, which requires estimating the deformation of the hand. In our method, the deformation model of the hand is introduced. Then, calibration is performed using the image to obtain specific parameter of the hand. Finally, based on the deformation model and predicted contact points, the deformation of the entire hand is estimated. An experiment was conducted to evaluate the accuracy of the proposed method.
