# Models Developed

### From Oispacewiki

This page lists the models that have been developed in the project.

## Research questions

- What are the similarities and differences between the ontologies, geometries, scales, and dimensions of O-space and I-space?
- How to construct a unified conceptual model of O-space and I-space? What are the key concepts in a unified conceptual model of O- and I-space?
- What theories and technologies employed in O-space can be applied for navigation in I-space? What new theories and technologies need to be developed for implementing the seamless navigation within the two spaces?

To build a unified model of O-space and I-space for navigation, investigating these two spaces separately is a good starting point. Exploring the similarities and differences between O-space and I-space from the perspective of navigation can lay the ontological and formal foundations for a unified model of the two spaces (For more information please see here), in which with applications to seamless navigation.
Many geometric and topological constructs can be used to represent the structure of
both O-space and I-space. For example, networks can often usefully represent both spaces. In O-space, roads are represented by edges and road junctions by nodes,
and in I-space, nodes represent rooms which are connected by edges representing corridors.
I-spaces generally have more “regular" geometries, for example, the boundary
of a room compared with a coastline. In O-space, distances, angles and coordinates play
essential roles (e.g., latitude and longitude), so Euclidean concepts are important. In I-space,
topology plays a more prominent role, for example, connectivity becomes more
important than direction. The scale of O-space is generally larger than I-space. Structurally,
O-space is usually represented in two dimensions, whereas I-space uses three
dimensions (e.g., levels of buildings).
In past decades, GIS is traditionally focused on O-space. We have investigated what theories and technologies applied in O-space can be used or extended for I-space
by comparing the two spaces and how to use them in I-space. For navigation, network
and topology play essential roles. So in this research we will investigate and extend the existing network and topology theories that have already been applied in O-space in order to fit the practical needs of the navigation applications in I-space.

## Ontologies developed

Building several microworlds (proposed by Sowa) is more beneficial than a large comprehensive ontology, because microworlds make it easier to share and reuse knowledge. We have built different levels of microworlds, based on given upper ontologies:

- Domain ontologies (Structure) related to the fabric of the I, O, and OI spaces. For example rooms, corridors, walls for indoor space, and roads, parks, bridges for outdoor space.
- Task ontologies (Navigation) related to navigation in I,O, and OI spaces.
- Application ontologies, applying the navigation tasks in different types of spaces (indoor navigation for pedestrian, outdoor navigation for vehicle, outdoor navigation for pedestrian).

For more information please see here (after Guarino, 1998). We are seeking and constructing the intra-relationships between the concepts within each ontology and inter-relationships between the concepts in different ontologies.

## Data models developed

Data models of O-space, I-space, and OI- space were built using UML. UML helps us to keep our data models consistent with the ontologies developed already.

Following the development of the ontologies and data models, we are in the process of building the geospatial database of Boardman Hall, the office building where our department is located. By this means, we can test our ontologies and data models.