Multimodal virtual reality: Input‐output devices, system integration, and human factorsBurdea, Grigore; Richard, Paul; Coiffet, Philippe
doi: 10.1080/10447319609526138pmid: N/A
Virtual reality (VR) involves multimodal interactions with computer‐simulated worlds through visual, auditory, and haptic feedback. This article reviews the state of the art in special‐purpose input‐output devices, such as trackers, sensing gloves, 3‐D audio cards, stereo displays, and haptic feedback masters. The integration of these devices in local and network‐distributed VR simulation systems is subsequently discussed. Finally, we present human‐factor studies that quantify the benefits of several feedback modalities on simulation realism and sensorial immersion. Specifically, we consider tracking and dextrous manipulation task performance in terms of error rates and learning times when graphics, audio, and haptic feedback are provided.
Postural instability induced by virtual reality exposure: Development of a certification protocolKennedy, Robert S.; Stanney, Kay M.
doi: 10.1080/10447319609526139pmid: 11540107
Exposure to virtual environments often causes users to experience symptoms of motion sickness. An accessory manifestation of motion sickness symptoms is postural disequilibrium. If the postural disequilibrium that occurs persists beyond the time an individual is within the confines of the laboratory or system site, user safety could be compromised and products liability issues could be forthcoming. In this study, a portable, automated postural assessment system is developed that can be employed before and after exposure to a virtual reality (VR) system in order to certify that a user's balance on exiting the system is at least demonstrably as good as it was on entering. It is argued that if the “coming out” balance performance is sufficiently poorer than the “going in” balance, then the user should be retained until the pretest balance performance is regained. The results from a set of normative and validation experiments on postural equilibrium identified several reliable measures of stance that could serve as a basis for certification. Furthermore, a new automated video‐based measure using only head movement showed that performance over sessions is stable and reliable. The head movement changes that occur with stimuli, such as alcohol and simulator exposure, are well behaved, predictable, and significant even with small samples. The implication is that the proposed objective measure of postural stability, in conjunction with procedures for obtaining self‐reports of symptoms, can afford some measure of certification that exposure to a given VR system is without harm.
Navigating large virtual spacesDarken, Rudolph P.; Sibert, John L.
doi: 10.1080/10447319609526140pmid: N/A
As important as navigation is to human performance in virtual worlds, it is an often overlooked problem in the design process. This article reports an experiment intended to show that real‐world wayfinding and environmental design principles are effective in designing virtual worlds that support skilled wayfinding behavior. The study measures participant performance on a complex searching task in a number of virtual worlds with differing environmental cues. The worlds are augmented with either a radial grid, a map, or both a grid and a map. The control condition provided no additional navigational cues. The results showed that navigational performance was superior under both map treatments as compared to the control and grid conditions. The grid was, however, shown to provide superior directional information as compared to the other conditions. The control condition provided the worst performance, with participants often becoming disoriented and experiencing extreme difficulty completing the tasks.
Task‐level interaction with virtual environments and virtual actorsZeltzer, David; Gaffron, Swetlana
doi: 10.1080/10447319609526141pmid: N/A
In many virtual environment (VE) applications, the VE system must be able to display accurate models of human figures that can perform routine behaviors and adapt to events in the virtual world. In order to achieve such adaptive, task‐level interaction with virtual actors, it is necessary to model elementary human motor skills. SkillBuilder is a software system for constructing a set of motor behaviors for a virtual actor by designing motor programs for arbitrarily complicated skills. Motor programs are modeled using finite state machines, and a set of state machine transition and ending conditions for modeling motor skills has been developed. Using inverse kinematics and automatic collision avoidance, SkillBuilder was used to construct a suite of behaviors for simulating visually guided reaching, grasping, and head‐eye tracking motions for a kinematically simulated actor consisting of articulated, rigid body parts. All of these actions have been successfully demonstrated in real time by permitting the user to interact with the virtual environment using a whole‐hand input device.
The virtual academy: A simulated environment for constructionist learningMoshell, J. Michael; Hughes, Charles E.
doi: 10.1080/10447319609526142pmid: N/A
The Virtual Academy is an educational model based on multiage teams of students and adults working through the Internet to build and use virtual worlds for educational purposes. These collaborations are mediated by a range of tools ranging from electronic mail to hypermedia and video links, and result in the creation of simulation‐based role‐playing adventure games within the ExploreNet software environment ExploreNet is an Internet‐based multimedia, multiuser domain constructed specifically for educational experimentation. This article describes the Virtual Academy Model, the ExploreNet software system, and an experiment conducted in the spring of 1995. The article describes the evolution of features of ExploreNet's user interface and their relevance to collaborative work by children.