Modeling and analysis of 3D deformable object grasping

Robotic grasping has been extensively studied in the last two decades. Most of the researches in this field have been dedicated to rigid body grasping and only few of them have considered the case of deformable objects. Nevertheless, robotized grasping of deformable objects has many potential applications in various areas including bio-medical processing, food processing industry, service robotics, robotized surgery, etc. In this paper we address the problem of the grasping of 3D deformable object using a multi-fingered robotic hand. The deformable object behavior is modeled by using a non-linear isotropic mass-spring system. In order to deal with the interactions between the manipulated body and the robotic hand fingers, a contact model is proposed. It is based on force calculations between the hemispherical finger tips and each facet of the meshed contact surfaces. The initial fingers' configuration is defined so that a force closure grasping is obtained, which eliminates any relative Euclidean motion of the manipulated object, meanwhile, a grasping criterion is optimized. Several grasping criteria are evaluated along a pick and place operation on the deformable object.