Advancing architectural design and education through the use of virtual environments technology Lee Anderson School of Architecture University of Minnesota
[email protected]
Victoria Interrante, Ross Treddinick, Brian Ries and Jason Lindquist Dept of Computer Science and Engineering University of Minnesota {interran,riesbr}@cs.umn.edu; {tredi002,lind0234}@umn.edu
ABSTRACT
DISCUSSION
The environment of the computer screen, though widely accepted, has several limitations on its use for design representation.
In many types of design, the design must be represented indirectly. Unlike a potter, for example, an architect must design using only a representation of the final result. Consequently architecture has always been concerned with the means of representation and the effect that different means of representation have on the design process and outcome.
This presentation describes design tools and methods developed for Virtual Environments by the Digital Design Consortium at the University of Minnesota that seek to overcome these screen limitations. The Digital Design Consortium is a composed of faculty from the departments of Architecture and Computer Science. Our purpose is to develop digital new tools and methods for architectural design. Keywords
Immersive virtual environments, architectural design. INTRODUCTION
Designs that we create are often affected in subtle but systematic ways, of which we’re typically not even consciously aware, by the tools that we use to create them. Our primary visualization tool, the computer screen, provides an environment that has several limitations on its use for design representation, among them: 1) It separates our vision from the rest of out body, so we have no physical memory to combine with what we have seen, as we do with normal vision. Thus our sense of “space” is limited, and we tend to objectify our design, and often design from an external viewpoint; 2) View changes are done in a formal manner, without the subtlety and quick spontaneity of our normal visual explorations; 3) The screen can present a perspective, but not one likely to be correct for the actual viewpoint, and one that is tightly framed in a rectangle. Even if stereo is provided (which is rare) the tools for constructing and modifying the model remain rooted in a 2D methodology. Virtual Environments technology has the potential to enhance the both the process and products of design by reducing or eliminating limitations related to the traditional screen environment.
The rapidly increasing use of Building Information Modeling (BIM) and digitally-assisted fabrication in architectural education and practice is placing an increased emphasis on the development of the three-dimensional model as the representation of a building project. While this model is very useful, it tends to “objectify” the building when seen on a computer screen. Students increasingly are creating designs that are evaluated almost entirely within the environment of the screen. The screen provides a compelling image, but, for the reasons stated, little actual sense of what it would mean to occupy the actual built design. This is especially true for students, who have little or no experience in having their designs built. Students designing on a screen can often be observed creating designs that have an aesthetic that respects the aspect ratio of the screen; the image on the screen is not interpreted as a view into a larger space that persists beyond the boundaries of the computer monitor - the boundaries of the screen are the boundaries of the imagination space. Another common situation is “designing from a privileged point of view”. Unlike in a real environment, that easily affords changing one’s viewpoint, in a CAD environment there is a mental overhead associated with changing one’s view of a model; thus, designs are often initially conceived from a single point of view, rather than holistically; with the result that the design is explored from other points of view farther on in the process of conceptualization than affords maximum flexibility in the consideration of design alternatives. The use of immersive together with tools to environment for design, many of the limitations of
Virtual Environments (IVE), take best advantage of this provide a way of overcoming a standard screen environment.
An immersive, head-mounted display (HMD) environment recombines our visual sense with the rest of our body. The subtle, instinctive movements and head nods that are so characteristic of a person examining an object or space are present again when a designer is in an immersive VE. The designer is presented with an environment that gives a complete sense of surroundings, leading to an exploration of space, as opposed to a screen environment that tends to restrict explorations to those of “form” (a design term that has come into vogue concurrent with screen-based designing). In an IVE, the design can be presented at full scale, giving the designer, by his/her movements a strong sense of scale and a gestalt that combines body memory of place with visual memory of place.
implications of their design, yet little experience with which to evaluate the final result.
Naturally, the tools appropriate for designing in a screen environment must be, at least, adapted for use in a virtual environment. The following projects show two of our efforts to create new tools for designing within Immersive Virtual Environments. AN IMMERSIVE VIRTUAL ENVIRONMENT FOR DESIGN (HMD-BASED)
In our first VE design system, GlCreator (Figure 1), our intent is not to create a full modeler, but to provide basic modeling tools for exploring issues of designing in Virtual Environments as experienced from an egocentric perspective and at arbitrarily large scale, up to life-sized or larger.. GLCreator uses a virtual wand in conjunction with a kiosk that serves as a toolbox. The toolbox/kiosk metaphor was chosen to keep all aspects of the application in three dimensions. Designing is accomplished with use of the virtual wand. Capabilities include standard object creation and transformation tools, including Boolean operations. GlCreator uses a unique spinning cylinder from which components, textures, images and animations can be selected to use in the virtual world. All objects dropped into a specific folder appear on the wheel. GlCreator has provisions for working at a design station (Figure 2) with table and chair existing in both virtual and real worlds. It also provides a virtual pin-up space that allows saving 2D images from the design for review in a virtual critiquing space. Clicking on a 2D pinup will restore the design to the place it was when the 2D image was saved, providing a type of “paper trail”. An illustration of the difference between the virtual design environment and the screen environment is that designers (typically students in architectural design studios) who observe their design on a standard screen after the virtual design session is complete, don’t recognize it as the same design. This illustrates the value in design education, where students have a special need for understanding the
Figure 1: An illustration of some of capabilities of the GLCreator system.
added to SketchUp to provide navigation with the virtual environment. The system allows working in a standard, screen environment while being immersed in the design, in any view, while it is being created.
QuickTime™ and a decompressor are needed to see this picture.
This system has the advantage of being able to work in a view best suited for modeling, while simultaneously observing in a view best suited for understanding the design. It also allows more that one designer to participate in viewing/critiquing the design as it progresses.
Figure 2: An example of an immersive design environment. Top: real world; Bottom: within the GLCreator system. AN IMMERSIVE VIRTUAL ENVIRONMENT FOR DESIGN (IPT-BASED)
The general goal in this second project is to leverage the capabilities of a popular architectural modeling system (SketchUp) and augment those capabilities with possibilities of immersive viewing. The immersive display in this case is a large, curved, screen, rear-projected with three, high-resolution stereo projectors. The user is tracked, and the screen image generated according to the position of the user so that the screen appears to be a window into a virtual world. The user designs on a tablet PC using an augmented version of SketchUp, while simultaneously being immersed in a view of the same design. All changes made to the SketchUp model are detected and used to update the model displayed in the immersive virtual environment. Tools are
Figure 3: top: a view of a model in SketchUp; bottom: a view of our system in use. FUTURE WORK
Many open questions that need to be address in order to fully harness the potential of IVE to facilitate the process and products of architectural design and education: -
Architectural design is critically dependent on an accurate, intuitive sense of space and spatial relationships. In order to harness the full potential of immersive virtual environments for facilitating the process of conceptual design in architecture, we need to ensure that people will be able to achieve as accurate an egocentric, spatial understanding of an environment from experiencing it virtually as they could from experiencing it in real life. The problem is that in numerous previous studies [e.g. Henry and
Furness, etc.], investigators have found that when people initially experience an immersive virtual environment (IVE) presented via a head-mounted display system, they report, and act, as if they perceive distant objects, or locations in space, as being closer to them than they really are – in some cases by up to 50% or more. We have begun to make important progress in elucidating the factors that underlie this phenomenon [Interrante et al], but much work remains to be done in order to determine how best to enable people to attain an accurate spatial understanding in 3D IVEs. -
Large projection display can be more practical than HMDs for use in many situations, such as when many people need to be able to see a design at the same time, as in a design review. However there remain many unanswered questions whose resolution can inform the more effective utilization of IPT-based VE systems: o
o
-
-
To what extent, and under what conditions, do people interpret what they see on a large projection display as being equivalent to an image that represents a view of a scene from their own eyes, e.g. as if they were looking out a window defined by the extent of the screen, versus as equivalent to a view from a camera at an unknown location? To what extent and under what conditions can people accurately infer sizes and spatial relationships in environments presented via large projection display systems?
Architectural designs are usually much too large to permit adequate observation from the limited movement provided within the tracked area of an immersive design environment. This is especially true given that most architectural designs are embedded within a larger context, which is also modeled. Our largest model, for example, covers several square miles of downtown Saint Paul, and is used to evaluate design changes that are proposed in that setting. Even somewhat smaller models, such as the one of our campus, require a considerable amount of virtual “traveling” in order to experience all aspects of a proposed design change. In immersive virtual environments presented via headmounted display systems, it has been found [Usoh] that people report a greater sense of presence when they are enabled to traverse the environment by directly walking than when they are required to rely on
an indirect metaphor for locomotion, such as walkingin-place, or an abstract metaphor, such as flying. However, when the immersive virtual environment represents a space that is larger than the available space within which a user can travel by directly walking, it becomes necessary to consider alternative methods for moving through the space. One possibility that we have been investigating is augmented walking with seven league boots. Initial findings are that accelerated walking is superior to “flying” as a means of navigation. ACKNOWLEDGMENTS This research was supported the National Science Foundation (IIS-0313226 and CNS-0323471), by the Digital Technology Center at the University of Minnesota, and by the Linda and Ted Johnson Digital Design Consortium Endowment and Lab Setup Funds. REFERENCES
1. Anderson, L., Esser J., and Interrante, V. A Virtual Environment for Conceptual Design in Architecture. 9th Eurographics Workshop on Virtual Environments/7th International Workshop on Immersive Projection Technology, (Zurich, Switzerland, May 2003), 57–63. 2. Henry, D. and Furness, T. Spatial Perception in Virtual Environments: Evaluating an Architectural Application, IEEE Virtual Reality Annual International Symposium, (Seattle, WA, September 1993), 33–40. 3. Interrante, V., Ries, B., Lindquist, J., and Anderson, L. Elucidating the Factors that can Facilitate Veridical Spatial Perception in Immersive Virtual Environments, IEEE Virtual Reality, (Charlotte, NC, March 2007), to appear. 4. Treddinick, R., Anderson L., and Interrante, V. A Tablet Based Immersive Architectural Design Tool, Synthetic Landscapes [Proceedings of the 25th Annual Conference of the Association for Computer-Aided Design in Architecture], (Louisville, KY, October 2006), 328–340. 5. Usoh, M., Arthur. K., Whitton, M.C., Bastos, B., Steed, A., Slater, M. and Brooks, F.P., Jr. Walking > Walkingin-Place > Flying, in Virtual Environments, ACM SIGGRAPH, (Los Angeles, CA, August 1999), 359– 364.
