The fingertips are one of the most important and sensitive parts of our body. They are the first stimulated areas of the hand when we interact with our environment. Providing haptic feedback to the fingertips in virtual reality could, thus, drastically improve perception and interaction with virtual environments. In this paper, we present a modular approach called HapTip to display such haptic sensations at the level of the fingertips. This approach relies on a wearable and compact haptic device able to simulate 2 Degree of Freedom (DoF) shear forces on the fingertip with a displacement range of ±2 mm. Several modules can be added and used jointly in order to address multi-finger and/or bimanual scenarios in virtual environments. For that purpose, we introduce several haptic rendering techniques to cover different cases of 3D interaction, such as touching a rough virtual surface, or feeling the inertia or weight of a virtual object. In order to illustrate the possibilities offered by HapTip, we provide four use cases focused on touching or grasping virtual objects. To validate the efficiency of our approach, we also conducted experiments to assess the tactile perception obtained with HapTip. Our results show that participants can successfully discriminate the directions of the 2 DoF stimulation of our haptic device. We found also that participants could well perceive different weights of virtual objects simulated using two HapTip devices. We believe that HapTip could be used in numerous applications in virtual reality for which 3D manipulation and tactile sensations are often crucial, such as in virtual prototyping or virtual training.

Abstract. In this paper, we study the perception of tactile directional cues by one or two fingers, using either the index, middle, or ring finger, or any of their combination. Therefore, we use tactile devices able to stretch the skin of the fingertips in 2 DOF along four directions: horizontal, vertical, and the two diagonals. We measure the recognition rate in each direction, as well as the subjective preference, depending on the (couple of) finger(s) stimulated. Our results show first that using the index and/or middle finger performs significantly better than using the ring finger on both qualitative and quantitative measures. The results when comparing one versus two-finger configurations are more contrasted. The recognition rate of the diagonals is higher when using one finger than two, whereas two fingers enable a better perception of the horizontal direction. These results pave the way to other studies on one versus two-finger perception, and raise methodological considerations for the design of multi-finger tactile devices.

No abstract