Abstract
The advancements in the field of XR devices and systems are interesting from an industrial point of view, as they present new opportunities for improving productivity and operations through-smart tooling, digitally enhanced assembly and maintenance, inspection, remote collaborations, etc. Typically, the XR headsets claim to provide a full 6-DoF tracking, while this may be good enough for consumer or entertainment applications; for an industrial application, we need to determine the exact errors and tolerances of the tracking for practical applications. In this paper, we present our methods and critical measurements from evaluating HTC Vive XR Elite and Magic Leap 2 for full 6-DoF tracking, depth perception accuracy, and drift accumulation over time. Through these tests, we measured a significant difference between individual XR devices' tracking accuracy, depth perception, and drifts, which could range from moderate to severe impact for the on-job deployment of these devices. By systematically analyzing error margins and tracking fidelity, this study aims to provide new valuable insights into the strengths and limitations of tracking capabilities of these XR devices, and the methodology which can be adopted to evaluate others. Further, this study could also help design AR symbology and user experience for an industrial application.