The effectiveness of an AR-based context-aware assembly support system in object assembly

Abstract

This study evaluates the effectiveness of an AR-based contextaware assembly support system with AR visualization modes proposed in object assembly. Although many AR-based assembly support systems have been proposed, few keep track of the assembly status in real-time and automatically recognize error and completion states at each step. Naturally, the effectiveness of such contextaware systems remains unexplored. Our test-bed system displays guidance information and error detection information corresponding to the recognized assembly status in the context of building block (LEGO) assembly. A user wearing a head mounted display (HMD) can intuitively build a building block structure on a table by visually confirming correct and incorrect blocks and locating where to attach new blocks. We proposed two AR visualization modes, one of them that displays guidance information directly overlaid on the physical model, and another one in which guidance information is rendered on a virtual model adjacent to the real model. An evaluation was conducted to comparatively evaluate these AR visualization modes as well as determine the effectiveness of context-aware error detection. Our experimental results indicate the visualization mode that shows target status next to real objects of concern outperforms the traditional direct overlay under moderate registration accuracy and marker-based tracking.

Attention guiding techniques using peripheral vision and eye tracking for feedback in augmented-reality-based assistance systems

“…Khuong et al [13] apply two AR visualization methods for supporting assembly tasks using real-time detection of the assembly status. They found that displaying guidance information as an overlay to the physical target leads to longer completion times than displaying information in a side-by-side manner adjacent to it.…”

In-Situ Instructions Exceed Side-by-Side Instructions in Augmented Reality Assisted Assembly

“…In past experiments these performed worse than projectionbased approaches [5], but there was already some evidence that approaches on modern AR glasses may perform similarly well with the advantage of not being bound to a specific workspace [4]. Previous comparisons also indicated that side-by-side instructions outperform in-situ instructions using AR glasses [14]. However, we anticipate that these results do no longer hold in light of modern hardware.…”

“…Stanimirovic et al [25] use a tablet computer for showing in-situ visualizations in form of superimposed 3D animations and additional billboarded textual descriptions. Also in a LEGO DUPLO assembly scenario, Khuong et al [14] proposed a side-by-side visualization for instructing workers. Compared to a wireframe-based in-situ instruction, their results show that spatial separation of the real and the virtual model performed better than the target-aligned in-situ variant.…”

“…paper-based -instructions. Past mobile AR devices thereby often had performance issues leading to slow and instable tracking [14,26] or a low visual fidelity due to low-resolution displays [13]. Only recently mobile AR devices with reasonable processing power, tracking accuracy and latency, as well as display quality were released to the market, e.g.…”

“…We evaluated four different AR instruction techniques for placing a brick in this scenario: An improved 3D in-situ visualization of the brick matching the exact target position, orientation, and color; a 2D in-situ visualization simulating a projected instruction; an animated wireframe representation of the brick, which should avoid occlusions of the target assembly position; and a side-by-side instruction based on the approach proposed by Khuong et al [14], which contains all previously assembled bricks.…”