A Novel Measurement Method for Performance Assessment of Hands-Free, XR-Based Human–Machine Interfaces

This article presents an innovative measurement method for assessing the information transfer performance of hands-free human-machine interfaces (HMIs) based on extended reality (XR) technology. The proposed method primarily involves the design and implementation of a dedicated XR environment, which serves as a testbed for data acquisition. Following this, an experimental campaign is conducted, involving multiple acquisition cycles for different individuals. Finally, the proposed method enables extraction of two primary metrics, namely, selection accuracy and information transfer rate (ITR), indicative of the potential of the considered HMIs to transfer information. These metrics account for both intraindividual and interindividual variabilities within the HMIs, thus providing a metrologically sound assessment of performance. The proposed method is validated through a practical case study. Three NHMIs are considered: eye-tracking, head-tracking, and brain-computer interfaces (BCIs) based on steady-state visually evoked potentials (SSVEPs), as they allow hands-free interactions solely through visual observation. Without loss of generality, Microsoft HoloLens 2 and Unicorn Hybrid Black were used as XR and BCI platforms, respectively. The experimental findings obtained from eight healthy individuals allowed a comparative analysis of the performance of the three distinct HMIs, facilitating a better understanding of which interface might be more robust for a given application scenario. Overall, the proposed method represents a reliable performance assessment of innovative HMIs. This becomes increasingly significant considering the evolution of wearable HMIs and the current lack of comprehensive strategies for their characterization.