Designing virtual reality laboratories for the evaluation of vibrotactile warnings
Virtual Reality (VR) provides researchers with a tool that allows them to recreate a wide variety of situations and scenarios virtually and use these virtual scenes for research purposes. Thus, VR offers the possibility to simulate immersive hazard scenarios and test them with people without putting the participants in real danger. By making it possible to virtually represent and immersively experience various domains and contexts, such VR laboratories (VR Labs) bridge the gap between laboratory and field testing.
This thesis explores the research question to what extent VR Labs are suitable for testing (directional) vibrotactile warning devices, focusing on the design requirements for VR Labs from a socio-technical perspective. As an application domain, construction site safety is used.
The fundamental challenge in the construction domain that this thesis addresses is the high number of accidents and fatalities in the industry. These are a consequence of many factors, such as time pressure, stress, workload, or the ever-changing work environment itself. The industry is trying to reduce the number of accidents and increase safety on construction sites, among other things, with safety training, safety inspections, and personal protective equipment. This thesis takes up the aspect of personal protective equipment and explores the idea of on-body directional vibration warnings of hazards, as is being researched for Proximity Warning Systems (PWS). Systems, such as PWS, can be viewed as augmenting human capabilities by alerting workers to hazards they may not be aware of. However, evaluations of such systems are often limited to technical components or laboratory testing since field testing could put participants at risk and would require extensive resources. Here, VR Labs allow testing such systems in an immersive, simulated work environment.
This thesis explores design aspects of VR Labs for evaluating directional vibrotactile warnings and construction site safety. First, the theoretical background is provided by discussing and referencing the contributions of cognitive psychology, the basics of tactile feedback, and relevant aspects of VR. Then, the design space is presented, and requirements are derived from a socio-technical analysis of the application domain, a conceptual design of vibrotactile wearables, and the translation of domain knowledge into the modeling of Virtual Environments (VE).
Then, several aspects of VR laboratories are reviewed in the subsequent studies. First, the extent to which the perception of vibrotactile cues in VR differs from that in reality and the impact this has on the choice of locomotion in VR is addressed. For this purpose, a proof-of-concept is presented to compare results made in VR with those from testing in real environments. Then, controller-based locomotion is compared to free movement in VR, and design recommendations are made based on the results and observations.
Based on the experiences made in the two studies, a study is presented on evaluating vibrotactile PWS in VR for hazard recognition and behavior adaptation. Here, directional vibrotactile warnings are applied via a wearable attached to the waist of participants. This study highlights the need to include human factors in the development of PWS and discusses design aspects regarding the design of tasks, vibrotactile signals and patterns, and graphical fidelity of VEs.
Then, a study on VR safety training for experiential learning is presented. In this study, a simulation was used to teach the use of angle grinders and the associated safety aspects. The system was evaluated with trainees of the construction industry on learning effects, user experience, and usability. This study is followed by an analysis of social media data on the topic of VR-related accidents to provide insights into which safety aspects to consider when testing with VR. For this purpose, social media data from the platform ‘Reddit’ was downloaded and analyzed.
Finally, based on the experiences in the studies, a chapter on the design aspects of VR Labs discusses design recommendations and clarifies the socio-technical requirements of VR Labs. This chapter also provides multiple application scenarios for VR Labs, highlighting their flexible applications for research.
In conclusion, this thesis provides a broad understanding of design challenges and recommendations when designing VR Labs for the evaluation of directional vibrotactile warnings and construction site safety.