This paper describes a visual tool for teleoperative experimentation involving remote manipulation and contact tasks. Using modest hardware, it recovers in real time the pose of moving polyhedral objects, and presents a synthetic view of the scene to the operator of a teleoperated robot using any chosen viewpoint and viewing direction. To recover pose, the method of line tracking first introduced by Harris (1992) is extended to multiple calibrated cameras, and its dynamic performance improved using robust methods and iterative filtering. Experiments are reported which determine the static and dynamic performance of the vision system, and its use in teleoperation is illustrated in two experiments, a peg-in-hole manipulation task and an impact control task.

Proposes a methodology called visual servoing for dexterous manipulation using dexterous hands. Most of the literature on dexterous manipulation follows the classical "planning then control" scheme. A successful outcome is conditional on the perfectly planned fingers' trajectories. Although tactile sensors have been improved, they are not accurate enough for the object's pose estimation. Visual servoing can be considered as an attractive solution to the problems described above. Nevertheless, from the stability point of view, the introduction of a camera inside the control loop could lead to an unstable response. Using a DAE (differential algebraic equation) formulation we introduce all the system components. We analyzed some situations that can be efficiently carried out. Finally we conclude by providing a numerical example of a 2D dexterous manipulation guided by the camera information.

This paper addresses the problem of dexterous manipulation of objects using an articulated hand. We describe some geometric reasoning techniques that can be applied to solve active compliance tasks. In this context, fine motion strategies are derived from the positions and forces between the grasped object and the fingers of the hand. Our approach is based on the minimum power principle of the grasping system, using the force/position information to determine a sequence of well coordinated manipulation actions (fixed-point, rolling and sliding). Different manipulation situations are illustrated using a semi-spherical model of the fingertips. To test the techniques described in this paper, we present an example showing an articulated hand changing the orientation of a polyhedral object while respecting the stability constraints of the grasp.