First-Person Perspective Operation Guidance and Quality Inspection with Augmented Reality in Complex Product Assembly

Manual assembly of complex products, such as aerospace engines, demands high precision and efficiency, yet conventional 2D guidance methods lack contextual clarity and fail to support real-time quality assurance. To address these limitations, this study explores the use of augmented reality head-mounted display (AR-HMD) to enhance first-person perspective (FPP) assembly guidance and inspection. Existing AR-based approaches typically focus on either instruction delivery or post-process inspection, with limited integration between guidance and quality control. This study proposes a unified framework that combines FPP guidance and FPP inspection (FPP-GI). The FPP-GI framework leverages hierarchically structured spatial information to provide intuitive guidance for operators and employs an enhanced visual recognition model along with a 3D reconstruction approach to enable quality monitoring during and after the assembly process. Experimental validation was conducted using a 3D-printed turbine engine model, with 24 participants (14 with mechanical assembly experience and 10 without). The results showed that AR-guided assembly reduced average task completion time by 37.0% for experienced participants and by 41.1 % for the inexperienced. User feedback indicated enhanced intuitiveness and usability, though varying perceptions were observed between participants. These findings demonstrate the potential of FPP-GI framework to streamline complex assembly workflows and strengthen quality assurance in industrial settings.