Dexterous Robotic Manipulation of Deformable Objects with Multi-Sensory Feedback - a Review

Book excerpt: In an attempt to support the ongoing effort of development for robotic solutions to the manipulation of deformable objects with multi-sensory feedback, this chapter reviewed the major trends adopted over the last decades in autonomous robotic interaction, which remains mainly guided by vision and force/tactile sensing. This extensive survey aimed at providing and classifying a critical list of relevant references that broadly cover the field. Starting from an overview of classical modeling and control techniques with application to the robotic manipulation of rigid objects, the review investigated how these approaches are being extended to the case of deformable objects manipulation. The main issues related with the significant differences between rigid and non-rigid objects were highlighted and consideration was given to a wide range of solutions that have been proposed, often in direct correspondence with a specific application. It is noticeable that most of the control methods available in the literature are applied to manipulate 1D and 2D deformable objects. The study of how to control a robot arm to handle a 3D deformable object still remains an open subject. Only a few early attempts to produce a generalized approach for handling 3D deformable objects were reported. Also, most of the proposed solutions currently available address the modeling problem of 3D deformable objects without attempting to solve the control problem simultaneously. Furthermore, the manipulation process does not involve any dexterity considerations. The study of these aspects is essential for the current effort of the robotic research community to establish a novel framework for the purpose of dexterous handling of 3D deformable objects. It involves the development of sophisticated multi-sensory systems to work in coordination with a robot arm and hand, taking into account their mechanical structure and control scheme, that influence the accuracy, and the dexterity. The integration of such complementary techniques will ensure that more elaborate manipulation can be achieved in the near future.