Computer Science Honours DRAFT Project Proposal


TITLE:

The design of a legacy independent Virtual Reality Computer Interface and subsequent implementation of this system on the Rhodes University CoRgi platform.


AUTHOR: Matthew Mundell

SUPERVISOR: Shaun Bangay

DATE: 15 March 1999



THE PROBLEM TO BE ADDRESSED AND ITS RELATION TO THE STATE OF THE ART:

Virtual and augmented reality could become the next area of computer revolution. Currently the focus of research and development in the virtual reality domain is on applications and improving the methods that are used to create, host and interact with these applications. Among the applications being undertaken, it seems that special attention is being given to data visualization (for simplifying the extraction of information from large data sets) and multi-user environments (as an extension of the ever-popular communications element in computing). The attention to multi-user environments in research is paralleled on the Internet where VRML (Virtual Reality Modelling Language) worlds, virtual worlds displayed in a web browser or stand alone program, allow interaction between users in the world.

Soon these applications will reach a affordable consumer usable point, which will enable complete 3d use of the computer and a move away from the unnaturalness of the 2 dimensionality of current computer displays. This project aims to address the imminent necessity for a 3d interface into the computer via the design of a virtual "meta-environment", which will be implemented on the Rhodes University CoRgi virtual reality platform.

The meta-environment will be a 3d world which contains objects having form and function, very much like a standard virtual reality application. The difference being that the meta-world will provide for

  • interactively building the world: designing and placing objects and associating functions with these objects
  • initiating applications that run as separate worlds
  • initiating applications that run within the meta-world (eg a 3d word processor)
  • switching between running applications
  • configuration of the system wide properties - what form will error messages take, system sounds, preferences for interaction devices etc.
  • possibly allowing object persistence throughout all running world - to allow, for example, the implementation of a clock that would always be visible
  • an extension of the multi-user interface (vxdm?)


Hopefully the project output will demonstrate the current possible level of implementation of a 3d operating system on relatively light hardware (there are no Onyx2 systems at Rhodes) and provide a basis for future development. Testing of the system on various forms of end users at all stages of the development is hoped to provide a stronger understanding of virtual interaction at a meta-environment level and a beter designed system.



LITERATURE SURVEY AND STATE OF THE ART:

There seems to be much literature, especially in the form of internet sites and published papers which are relevant to many specific facets of virtual reality but as the design and implementation of a 3D form of computer interaction meta-environment is new, little information in this specific area has surfaced at present.

An initial result of surveying available literature, showing the likely form of resources to be used, can be found in appendix A.

Many academic institutions, such as the University of Illinois in Chicago, Georgia Tech and the Center for Advanced Studies, Research and Development in Sardinia have virtual reality research focusses. The most promising of these seems to be the Advanced Interfaces Group (AIG) at the University of Manchester. Some papers which have come out of the AIG which may have relevance:

  • Snowdon. D. et al Towards the next generation of Human Computer Interface
  • West. A.J. et al Aviary - A Generic Virtual Reality Interface for Real Applications
  • Hubbold. R. et al Design Issues for Virtual Reality Systems

    Any interactive system requires understanding of human-computer interaction (HCI). Many current HCI texts have sections on 3D HCI. A selection of these can be found in appendix A.

    Implementation of the system on CoRgi is going to require at least an end-user knowledge of the CoRgi system, and probably also a general understanding of the underlying CoRgi structure and inner workings. Documentation for the CoRgi platform is automatically generated and can be found at http://www.cs.ru.ac.za/vrsig/corgidocs/index.html.

    Many projects from previous years have sections explaining different parts of the system.

    The blue-sky nature of the project may require some creative input from other facets of life. Some hypothesized virtual reality and virtual interfacing from the art world:

    Movies: Lawnmower Man, Lawnmower Man II, Johnny Mnemonic, Jurassic Park, Nemesis

    Novels: William Gibson's Neuromancer



    DESIGN CONSIDERATIONS AND IMPLEMENTATION DECISIONS:

    It is hoped that the design will result in a well thought-out operating environment that can evolve into a usable, growable system, providing intuitive, open interaction for the novice and advanced user. It is important that the design remains independent of platform and hardware constraints, and is not undermined by the legacy of 2D interfaces or current computer metaphors, but rather builds upon their positive aspects.

    Implementation on the CoRgi virtual reality platform has been chosen to take advantage of the current development of this system, research that has been already undertaken on the system, and the knowledge and experience that members of the department have with the platform. This will require a C based application which will conceptually run on top of the CoRgi libraries.

    Design and implementation will be segregated to avoid important design decisions being based on underlying systems and hardware.



    THE EXPECTED OUTPUT:

    Research, design and subsequent implementation should result in most of the following output:

    1. A design for an operating system and virtual reality platform independent framework for the creation of a virtual meta-environment to act as the user interface to the virtual reality computer. Focus here is on enabling the user to create 3D objects and associate actions and properties with them.

    The framework must be able to initiate applications and switch between running applications.

    2. Implementation of the designs on the CoRgi platform.

    3. One or more example created meta-environments for the system, including objects linked with various task usually associated with the operating system and operating/desktop environment eg. configuration of display settings.



    POSSIBLE EXTENSIONS:

    4. Extension of the design to provide for an interface into the file system.

    5. Ports of the design onto other systems underlying virtual reality systems.

    6. Further extension of the design to allow the system to run 2D apps. This will probably just be a program in itself, but perhaps the meta-environment will need to specially cater for these type of apps in such ways as displaying the underlying desktop environment components.




    Matthew Mundell
    Email with suggestions or comments welcome.

    Last updated 22 March 1999