Modular fixturing kits are sets of components used for flexible, rapid construction of fixtures. A modular vise is a parallel-jaw vise, with a regular grid of precisely positioned holes on each jaw. Parts are held by placing pins in the holes so that when the vise is closed, the part is reliably located and completely constrained. The modular vise concept can be adapted easily to the design of modular parallel-jaw grippers for robots. By attaching a grid-plate to each jaw of a parallel-jaw gripper, we gain the ability to easily construct high-quality grasps for a wide variety of parts from a standard set of hardware. Wallack and Canny (1994, 1996) developed an algorithm for planning planar grasp configurations for the modular vise. In this paper, we expand this work to produce a 3D fixture/gripper design tool. We describe several analyses we have added to the planar algorithm, including a 3D grasp quality metric based on force information, 3D geometric loading analysis, and inter-gripper interference analysis. Finally, we describe two applications of our code. One of these is an internal application at Sandia, while the other shows a potential use of our code for designing part of an agile assembly line.

Modular fixturing kits are sets of components used for flexible, rapid construction of fixtures. A modular vise is a parallel-jaw vise, each jaw of which is a modular fixture plate with a regular grid of precisely positioned holes. To fixture a part, one places pins in some of the holes so that when the vise is closed, the part is reliably located and completely constrained. The modular vise concept can be adapted easily to the design of modular parallel-jaw grippers for robots. By attaching a grid-plate to each jaw of a parallel-jaw gripper we gain the ability to easily construct high-quality grasps for a wide variety of parts from a standard set of hardware. Wallack and Canny (1994) developed an algorithm for planning planar grasp configurations for the modular vise. We expand this work to produce a 3D fixture/gripper design tool. We describe several analyses we have added to the planar algorithm, including a 3D grasp quality metric based on force information, 3D geometric loading analysis, and inter-gripper interference analysis. Finally, we describe two applications of our code. One of these is an internal application at Sandia, while the other shows a potential use of our code for designing part of an agile assembly line.

Presents an empirical example of a strongly cooperative manipulation system (the pusher/steerer system), whose capabilities derive from the redistribution of resources typically used in a single robot system. The authors define strongly cooperative strategies to be, intuitively, those which are not trivially serializable. In the authors' system, one robot steers, and the other pushes; the object lies between them. The steerer is the only agent that has information about the path, while the pusher exerts the necessary motive force, and rotates to follow changes in the object's orientation. The system is asynchronous, with no explicit communication between the robots. The authors are interested in studying manipulation systems of this type, because they are strongly cooperative; that is, the two robots must act in concert to achieve the goal. The authors have performed over one hundred pusher/steerer manipulation experiments with two of their mobile robots, TOMMY and LILY. As the authors' analysis predicts, they have found that their system allows a wide variety of paths and manipulable objects for these robots.