The Effectiveness of Artificial Landmarks for Orientation Tasks within a Large-Scale Virtual Environment

Trial Setup: Portable stereo rig running Bionatics Blueberry software with MultiGen Vega Prim.

Virtual environments are used widely in the academic world, both for research and for training. Within many of these implementations, orientation and navigation are key activities that need to be conducted accurately and efficiently for the purpose of the environment to be achieved. There has been extensive research into how people construct the knowledge required to perform these tasks, and aids that are used. Much of this research however has focused on smaller scale environments that are densely populated with features which either limit movement or act as landmarks to aid navigation.

Dynamically coloured artificial landmarks.

In his MSc GIS dissertation Adam Roussell (2006/7) investigated how artificial landmarks are used to aid in orientation technique within at a much winder geographical virtual environment that is only sparsely populated with man-made objects. It was evident that being able to see a number of artificial landmarks greatly improves accuracy in orientation tasks. However, Adam found that if a number of landmarks are visible, then a topological approach is taken to determine the user’s location rather than directly using distances from the landmarks. He also concluded that Urban Image Theory appears not only to hold for mental representation of urban areas which are densely populated with features, but also for much larger scale environments where there are little in the way of visible features. In these cases, the items used for the representations appear to be on a larger scale, such as general terrain cover and coastlines rather than urban zones.

Adam’s work used Bionatics Blueberry software working with Multigen Vega Prime on SPLINT’s flat screen portable stereo projection rig.

Staff involvement: Claire Jarvis, Jing Li, Bill Hickin