Trends in Grasping

Research and development in robotics and industrial automation has created a need for good grasp planning algorithms for a variety of object shapes and hand types. This in turn has stimulated research on the inherent computational geometry of grasping. The purpose of this paper is to survey some of the recent grasping results, algorithms, and ideas. We will place particular emphasis on research involving the synthesis of optimal grasps. We include a brief discussion of a few of the many problems that arise in the context of grasping. The main focus of this paper, however, is on the following problem and its variants: Given a description of an object (e.g. Is the object concave or convex? Is it a polygonal object or is it curved?) and a gripper (e.g. How many ngers/contacts does the gripper have? What type of contacts are they? Point contacts? Is there friction? What positions can the ngers obtain?), nd an \optimal" grasp of the object with the gripper or determine that there is no feasible grasp of the object. As we will see, even the notion of optimal will not always be measured by the same metrics, making a general solution di cult if not impossible to nd.