Task-oriented approach to simulate a grasping action through underactuated haptic devices

Force rendering is important in underactuated haptic systems. Underactuation means that some force directions at the contacts cannot be rendered because of the lack of actuation. In this paper we propose to exploit the knowledge of the task to mitigate the effect of the underactuation. The simulation of a grasp is considered and two alternative algorithms are proposed to improve the sensitivity in the underactuated system. The basic idea is to exploit the actuated force direction, optimizing the force feedback according to the type of forces involved in the specific grasping task. These forces can be squeezing forces or forces able to move the grasped object. Experiments show that the proposed task–oriented force rendering considerably increases the ability of perceiving the properties of the grasped virtual object.