Identification Accuracy and Efficiency of Haptic Virtual Objects Using Force-feedback

In the field of constructive solid geometry, modelers com-bine (or decombine) geometry primitives like cones, cylinders, prisms, pyramids and spheres to create digital models of phys-ical items. Haptically recognizing the shape of such com-plex models in virtual environments represents a challenging task if just minimal information is conveyed while using force-feedback with a single point of contact. However conducting object identification experiments is very important for deter-mining chances and limits offered by such a haptic feedback system. Therefore this paper describes two experiments using planes and surfaces based on geometry primitives, geometry primitives themselves, frustums and combined geometry prim-itives for recognition task. In the first experiment the models are postured upright (not rotated). In the second experiment the same models are randomly rotated to investigate influences of posture. Test persons ’ averaged identification accuracy and exploration time are determined. Furthermore important state-ments and sources of identification errors are listed that have to be considered to allow intuitive, reliable and fast exploration. Index Terms: haptic, force-feedback, virtual objects, identifi-cation, geometry, shap