No abstract

A navigation test was carried out in a spatially immersive virtual environment. The test was a game-like experience where task of subjects was to find as many gates as possible while they navigated through a track guided by auditory and/or visual cues of the gates. The results are presented as a function of the number of found gates, searching times, and normalized path lengths. Audio-visual navigation was clearly the most efficient. Visual navigation was the second, and the auditory navigation the least efficient. Further analysis of travel paths indicate that auditory cue was utilized in the beginning to locate the next gate, and visual cue was the most important in the final approach to the gate.

The modern working environment is gradually developing towards a more 'intelligent' one, which means remarkable technological, architectural, psychosocial and medical innovations. This article defines the concept of an intelligent work environment and evaluates its meaning from a user's perspective. The authors briefly discuss the history of the intelligent environment and take a multi-professional approach to foresee the possibilities and challenges awaiting future users of intelligent working environments: they stress the importance of accessibility and ergonomics when designing intelligent work environments of the future. In addition, multimodal and architectural aspects of the intelligent working environment are closely examined.

Historical background and motivation for ICAD2003 paper is presented. In addition, the research results followed up is briefly overviewed and future work is elaborated. HISTORICAL CONTEXTIdeally virtual reality (VR) is multimodal, meaning that applications should provide feedback to all senses, including visual, aural, and haptic feedback, in order to immerse a user into a virtual world. However, audio is seldom applied and 3D auditory displays are rarely constructed in cavelike systems [Cruz-Neira et al. 1992]. An example of this underemphasis of audio is in a recent book on virtual reality [Sherman and Craig 2003] in which visual displays are presented with 48 pages, haptic devices with 23 pages, and only 13 pages are dedicated for spatial auditory displays.Nevertheless, we foresee that spatial audio has a lot of potential in VR applications. 3D sound could be used in many application fields such as training, simulation, interactive design, entertainment, and visualization. Spatial audio could be applied in several common tasks, e.g., localization of an object, way-finding in navigation, data representation, object selection, and object manipulation.We have had a cavelike virtual environment at the Helsinki University of Technology since 1999 and we each have over 10 years experience and interest in the field of auditory displays. Thus, it is quite natural that our research mission has been to apply spatial audio in VR applications. For that purpose, our virtual environment facility has a multichannel 3D sound reproduction system, which provides good positional audio. In addition, our experience in sonification, psychoacoustics, room acoustics modeling, and spatial audio reproduction forms a strong basis for audio related VR.

In a CAVE-like virtual environment spatial audio is typically reproduced with amplitude panning on loudspeakers behind the screens. We arranged a localization experiment where the subjects' task was to point to the perceived location of a sound source. Measured accuracy for a static source was as good as the accuracy in previous headphone experiments using head-related transfer functions. We also measured the localization accuracy of a moving auditory stimulus. The accuracy was decreased by an amount comparable to the minimum audible movement angle.