Home im.ve Group im.ve Lab Projects Publications Teaching News, Events, Activities Contact us News

Dynamic, Interactive Virtual Environments (PhD Project Kristopher J. Blom)

The creation of interesting and compelling Virtual Environments is a goal shared by many areas that rely on VEs to create alternative realities for users. While the need for such engaging environments is well known and accepted, little formal work exists in defining what such environments are.  Numerous attempts to support their creation exist, but were created without a fundamental understanding of what kind of environments they are trying to support. Our work focuses on creating a new method of support by developing an understanding of the environments to be created and the support necessary to build them. Based on that understanding of the environments desired, a method of support has been created. 

Dynamic, Interactive Environments?

When  considering what environments fall into the catagories of interesting, engaging and compelling environments many different factors can be thought of. Umong the most important are the creativity of the authors that allows a "story" to be developed with the content. In this work, what we are interested in developing support for that creativity. When considering the factors that we can directly support as computer scientists, two major factors can be identified: dynamics and interaction. Worlds that have components that are changing over time are much more interesting than static environments. This is important, as many of the VR environments one can experience throughout the world's VR research centers are rather static.  Often the only dynamics to environments is the user's interactions, and those are often limited to travelling through the world.  The gaming world manages to create such worlds, but at the cost of time and energy. Games also have the advantage of limiting interaction with the world through the game structure, something that is often difficult to accomplish in VR.  While dynamics are necessary to create interest and evey story content, interaction increases the user's agency. This is what makes the environment engaging and that the users feels part of the environment. Hence, the core of this work is on Dynamic, Interactive Virtual Environments.

 

Dynamics, Interaction, Interactive Dynamics, and Dynamic Interaction

In our work, we performed an analysis of the content of such environments. An initial step was to exploration the design space of the environments in question. This was performed on hand the two components of dynamics and interaction. Exploring these two spaces, we identified spaces of high interest: interactive dynamics and dynamic interactions. Interative Dynamics are those dynamics of the environment that are interative. Understanding the design space, shows all the different potential ways of interacting with these changing components. Similarly the design space of Dynamic Interactions exposes those interactions that occur over time, i.e. they cause a dynamic in the environment. 

Based on a newly won understanding of the design spaces of these different catagories of content, we undertook an analysis of the content. In particular we sought ways in which their development could be supported. One of the most interesting ways of looking at the content was in terms of time (as dynamics are a major portion this makes sense). When looking at time, it is interesting to see that many of these dynamics, interactive dynamics and dynamic interactions are continuous in nature.  When comparing the time representations required with the support methods in existing systems a mismatch is seen, as almost all existing systems are based on discrete events. While discrete changes  also exist, events are generally there as part of changing the dynamic. Arranging our design spaces by time looks like this:

Video of an example world, containing dynamics, interaction, Interactive Dynamics, and Dynamic Interactions


 

Supporting the Creation of Dynamic, Interactive Virtual Environments

 Based on the analysis of DIVEs requirements for a support system were created. Additional System requirements out of prior VR system design experience rounded out the requirements list.
A new system designed to help support the creation of Dynamic, Interactive Virtual Environments in the constraints of Virtual Reality settings was created.. This system, combines the expressive power of the recently developed Functional Reactive Programming (FRP) paradigm with the support of current VR systems. This Functional Reactive Virtual Reality (FRVR  pronounced Fervor) system allows the VE author to describe the world in terms of continuous time behaviors and discrete events that change the behaviors of the world.  The basic components for the author are continuous functions (integral) where time is a "first class"  but invisible part of the system and event occurrence that signal switches between behaviours. The FRVR system builds on the FRP system, extending it to be more useful in the context of VR and convering the design spaces of DIVEs.   This includes advanced functionalities like a simple undo that works across dynamic interactions and interactive dynamics.

In depth information to the FRVR system can be found on its webpage.

 

Future Directions

Numerous directions of interest that have been identified during this work.  A sampling of the future work, including the efforts that have already started are given here.

 

Affect of Content on Users

A question of interest that has come to the forefront through this work is: how does having an dynamic, interactive  virtual environment actually effect user?  In our initial considerations, we have been theorizing how they might effect presence, a measure of how well users accept the world as a temporary reality.  There is an implicit acceptance that dynamics and interaction in worlds help form presence, but the topic is almost completely ignored to this point.  We are interested in forming a theory as to the foundations of this relationship.  A first proposal for a new theory of the relationship was presented at the Peach Summer School (part of the Peach EU project) in the form of a poster. This theory is based on the idea that the level of user agency afforded by the environment is how the relationship is built.  Developing research in the group is further striving to define this relationship and make first steps in providing empirical data to back it up.

Building on these ideas, we have started looking at how simple content is percieved by users. An initial study has been performed comparing the affordances of simple objects in virtual and real environments.

Interaction with Dynamics:

Once we finally have dynamics in a virtual environment, the impulse of users is to start to interact with the dynamics. Ongoing research, starting with the Selection of Dynamic Objects  project, is investigating how one can interact with dynamics.  The beginning work is investigating methods of selection (the basic interaction task in VR) for moving objects. Building off this work and the work of the dissertation above on dynamics, dynamic interaction, and interactive dynamics, we are starting to consider and develop methods for interaction with dynamics.

Simplifying the Authors experience:

 The FRVR system is designed to make it easier to create dynamic interactive virtual environments.  It equips the user with tools that support eh true natrue of the environments to be created, and it adds new capabilities that are signficant in the production of interactive dynamics. Unfortunately, a drawback of the current system is that the coding remains more difficult that one could wish.  While the system better supports the DIVE nature, it still requires programming experience and in particular a special extension of a less known language (Haskell).  For authors with mathematical and engineering backgrounds these hurdles should be readily overcome (Haskell has a mathematical-like syntax). However, to make FRVR  more usable we have also been exploring ways to make the system easier to use, particularly to lighten the programming load. To this end, we have been  exploring visual programming of Arrows and Signal Functions, the basis of the current FRP implementation.  Our work on a Visual Programming Environment of Arrowized Haskell can be found here.



Papers


Dynamic, Interactive Virtual Environments

Kristopher J. Blom
Book. Sierke Verlag. 2009. ISBN-13: 9783868441178
Website(s): Book at publisher's website


Dynamic Interactive Virtual Environments

Kristopher J. Blom
Dissertation
University of Hamburg, Department of Informatics,
Advisor: Steffi Beckhaus
Defended Jan. 2009
Website(s): Published version as bound book at publisher's website


On the Creation of Dynamic, Interactive Virtual Environments

Blom, K., Beckhaus, S.
In Proceedings of the IEEE VR 2008 workshop "SEARIS - Software Engineering and Architectures for Interactive Systems". Shaker Verlag. 2008. pp. 57-60.
File(s): Paper (PDF)


On Affordances and Agency as Explanatory Factors of Presence

Blom K.
Extended Abstract in Proceeding of Peach Summer School 2007
File(s): Paper (PDF)


Supporting the Creation of Dynamic, Interactive Virtual Environments

Blom K., Beckhaus S.
Proceedings of the 2007 ACM symposium on Virtual reality software and technology (VRST 2007). ACM. North Beach, California. pages 51-54.
File(s): Paper (PDF)
Website(s): DOI link


Functional Reactive Virtual Reality

Blom K., Beckhaus S.
in Short Paper Proceedings of IPT-EGVE Symposium (2007), B. Fröhlich, R. Blach, and R. van Liere (Editors), Weimar, pg. 295-302
File(s): Paper (PDF)


Integrating Functional Reactive Programming in a High-Level VR Framework

Blom K., Beckhaus S.
Virtuelle und Erweiterte Realität, 4. Workshop der GI-Fachgruppe VR/AR", M. E. Latoschik, B. Fröhlich (Editors), Weimar, pg. 189-196, ISBN 978-3-8322-6367-6
File(s): Paper (PDF)


Presence in Interactive Experiences

Blom K.
1st Peach Summer School 'Presence: towards human machine confluence-Presence technologies and Foundations.'
Poster, 2007.
File(s): Poster (PDF)


People involved:

Blom