Human Factors Design and Evaluation of Augmented Reality Visualization Techniques for Avoidance of Out-of-View Objects

Abstract

Recent advancements in augmented reality (AR) hardware, software and application capabilities have introduced exciting benefits and advantages, especially in industrial and occupational fields. The use of head-worn AR in real work settings affords the benefit of a hands-free tool, allowing workers to perform usual tasks and move around freely in their environment with easier access to critical information. The integration of novel technologies, like AR, into the real-world requires an understanding of how to utilize the technology as a catalyst for improved work-place efficiency, rather than further hinder or slow down performance. The ongoing, dynamic nature of industrial and occupational settings can benefit from AR assistive interfaces, such as visualizations that can cue users to task- or goal-related objects in their environment. Although the benefits of such cueing visualization techniques (VTs) are acknowledged, the consideration of FOV limitations and perceptual and cognitive challenges need to be further examined for effective visualization technique design. Perceptual issues, such as occlusion of real-world information, along with cognitive issues, such as heightened mental workload, can also occur with the use of AR, sometimes resulting in a negative impact on performance. Researchers and designers must consider the strengths and weaknesses of the human visual system, the allocation of limited attentional resources, the interface information density, and information encoding tactics used to cue the users to objects. Many AR visualization techniques already exist to aid users in locating and identifying objects outside of their FOV, but no technique universally solves the perceptual and cognitive issues associated with these interfaces. Previous work related to visualization techniques for attention guidance explored cues to guide users' attention towards an object of interest, for navigation, object collection or user task support. Our work, however, looked at how AR visualization techniques, specifically Whiskers, Wedge, and Compass, can be used to guide users away from objects, for safety or hazard avoidance. We explore the use of these cues in a fast-paced, outdoor and indoor environment, and apply our work to the domain of Search and Rescue, which is a time-critical and often dangerous mission. This work aimed to (1) explore existing attention guidance visualization techniques and the design elements and preattentive properties used to encode out-of-view object information, (2) examine cue conveyance effectiveness of visualization techniques for out-of-view object avoidance and related cognitive impacts on user safety and performance, and (3) investigate the integration of visualization techniques, and the properties used to encode crucial information, into Search and Rescue through expertise-driven examination with subject matter experts. Through quantitative and qualitative analysis, we derived visualization technique design guidelines for cueing to out-of-view objects, which may be applicable to many other industrial and occupational settings. Rather than identifying the best visualization technique, we identify the information encoded, preattentive properties, and design elements ideal for cue conveyance, user safety, and dual task support.