Evaluating the use of augmented reality to facilitate assembly

Abstract

Assembly is the process in which two or more objects are joined together through particular sequences and operations. Current practice utilises two-dimensional (2D) drawings as the main visualisation means to guide assembly. Other visualisation means such as three-dimensional (3D) manual and Virtual Reality (VR) technology have also been applied to assist in assembly. As an emerging technology, Augmented Reality (AR) integrates 3D images of virtual objects into a real-world workspace. The insertion of digitalised information into the real-world workspace using AR can provide workers with the means to implement correct assembly procedures with improved accuracy and reduced errors. Despite the substantial application of AR in assembly; related research has rarely been explored from a human cognitive perspective. The limited available cognitive research concerning the applications of AR visualisation means in assembly highlights the need for a structured methodology of addressing cognitive and useability issues for the application potentials of AR technology to be fully realised.This dissertation reviews the issues and discrepancies in using four types of visualisation means (2D drawings, 3D manual prints, VR, and AR) for guiding assembly, and investigates potential cognitive theories to underpin the benefits of animated AR in assembly. A theoretical framework is then put forward, which summarises existing mechanisms concerning visual-spatial information processing and THE Working Memory (WM) processing in the context of spatial cognition theory, active vision theory and THE WM theory, and raises the to-be-validated aspects of the above theories when transferring from the psychological arena to practical instances. Moreover, the dissertation formulates the methodology of configuring a prototype-animated AR system, and devising particular assembly tasks that are normally guided by reference to documentation and a test-bed with a series of experiments.Two experiments were conducted with three testing scenarios: experiment I concerns the evaluation in the first and second scenarios, while experiment II concerns the third scenario. In scenario 1, a small scale LEGO model was used as the assembly and experimental tester task to compare 3D manual prints and AR. This scenario measured the task performance and cognitive workload of using the system for assembly. The second scenario applied the knowledge gained from scenario 1 to the real construction piping assembly. Comparisons were then made as to productivity improvements, cost reduction and the reduction of rework between 2D isometric drawings and AR. Common findings from both scenarios revealed that the AR visualisation yielded shorter task completion time, less assembly errors and lower total task load. Evaluation from the real construction scenario also indicated that the animated AR visualisation significantly shortened the completion time (original time and rework time), payment to assemblers and cost on correcting erroneous assembly.Questionnaire feedback (including NASA task load index) (Hart 2006, 908) revealed that the animated AR visualisation better aided assembly comprehension, and better facilitated information retrieval and collaboration between human and guidance medium. Using the same LEGO tester task, the third scenario measured the training effects of using 3D manual prints and AR among novice assemblers. The results revealed that the learning curve of novice assemblers was reduced (faster learning) and task performance relevant to working memory was increased when implementing AR training. Useability evaluation was conducted based on classical useability methods, to assess the user interface regarding system improvements.