Evaluation of Haptic Human-Machine Interfaces for Virtual Reality Applications

Haptic human-machine interfaces (HMI) can be found in a variety of different applications reaching from the gaming industry to medical engineering applications. Nevertheless, to date no method for the standardized evaluation of haptic HMIs is available which allows to rate the performance of different systems relative to each other. Therefore, this work pre-sents an approach which enables a standardized comparison of different haptic HMIs. Sev-eral tasks which are conducted within a virtual reality (VR) environment were identified. Tasks were divided into two subsets of tasks. While the goal of subset one is to analyze basic system characteristics, subset two focusses on the evaluation of the performance of human-machine systems during complex object manipulations. This two-fold approach includes the underlying factors of the performance of human-machine systems on the one hand, but en-sures also that results can be transferred to complex applications on the other hand. In or-der to evaluate the proposed tasks, a user study was conducted including 32 participants, who completed the tasks with a selection of five different haptic HMIs. Devices were chosen to cover a broad bandwidth of size of the workspace and displayed forces. Results of the study showed that system characteristics identified by the tasks in subset one (e.g., the abil-ity to react on haptic feedback, or the ability to perform gross movements as well as fine manipulative adjustments) could be transferred to the more complex object manipulations in subset two. Furthermore, results from subset two provide a detailed understanding about the suitability of different human-machine systems to perform object manipulations in three dimensional spaces. In sum, the user study revealed that the identified tasks in VR are able to evaluate haptic HMIs out of completely different fields of application, which additionally use different control concepts. The results observed in this study can serve as benchmark for future evaluations of further haptic HMIs