In this Thesis we research the impact of Augmented Reality (AR), to aid Industrial Maintenance operations, developed in response to challenges and tasks abstracted from an operational Nuclear Fusion Plant. We argue that AR interfaces reliably improve task performance when completing realistic industrial tasks under representative conditions. We consider the potential of AR to improve performance through the reduction of mental load, unification of instruction and adaptive assistance. Our motivation is to understand the possible role of inter- active AR guidance for hazardous, repetitive tasks. We acknowledge that AR has already been shown to deliver improvements in task performance, however, there is limited research exploring the integration of AR realistic industrial routines, with consideration of the wider context, which presents a barrier to adoption. Our contribution is therefore to evaluate AR guidance, introducing tasks and processes taken from a live plant, and bring insight to the challenges and opportunities for Industry presented by immersive tools. Adopting a prototype led approach we undertake detailed research with operators of the Joint European Torus (JET) Fusion Plant. We explore both AR for tele-operated and manual maintenance operations, consider human factors, conduct thematic analysis and present a detailed overview of AR system requirements. In response we detail the development of a prototype AR system for Manual workers, utilising ‘hands free’ operation, hybrid tracking, and adaptive guidance to integrate with Plant processes. In trials we observe an improvement of 21% efficiency, 50% accuracy and 19% reduced task load. For comparison, we consider the value of AR cues for Tele-operators and present an evaluation of rotational AR cues for Nuclear Glovebox operations. With the addition of basic AR rotational cues we find a 34% reduction in error rate and 17% reduction in time to reach a correct outcome. We contribute findings which demonstrate the impact of AR within workflows extracted from Industrial Maintenance operations, a Taxonomy of cues drawn from existing literature, and a simulation based method for testing AR cues for remote operators. As a result of this work we propose guidelines for deploying AR within an Industrial context and recommend the most appropriate tasks and challenges. We present our analysis of various AR cue styles, human factors and consider further work to investigate the potential of ‘hands free’ operation within a Plant setting.
Augmented Reality guidance for fusion plant maintenance
Bale, T. A. K. (Author). 3 Oct 2023
Student thesis: Doctoral Thesis › Doctor of Philosophy (PhD)