The relativity of wayfinding

Volume 11 Number 1 January 12 - February 8 2015

 

New research into wayfinding has made some surprising revelations about how people take directions and use landmarks, with potential applications in emergency services calls, among other things. By Annie Rahilly.

People talk about places every day. Talking about where you are, where we should meet, or how to find the way to this meeting place is essential for orientation and wayfinding. 

Talking about place, or reading sketches in this regard, may appear so simple from the outside, but actually it is fascinating what people reveal about their spatial knowledge and achieve with this communication.

And of course, it would be nice if our go-everywhere, do-anything, smart phones could make sense of such talk as well. 

Consider the operators in emergency call centres. They have to interpret verbal place descriptions in order to dispatch an emergency team. Often in the emotion of the moment, accuracy is forgotten and words fail.

But what if a system could support call centre operators by identifying location references and accurate details about their local configuration?

Not such an improbable notion if we think of the near future when we will only talk to our otherwise automatically driving cars in order to agree on destinations or routes.

Spatial information science and geomatics expert Stephan Winter from the Department of Infrastructure Engineering at the University of Melbourne specialises in human spatial cognition and communication.

His passion is in developing intelligent spatial machines and he is pioneering new ways of modelling environments, including mobility solutions and crowd-sourced data and knowledge acquisition. The everywhere phone can be used to accept and exchange information.

Professor Winter has made significant contributions towards the formulation of intelligent spatial systems, computational transportation science, and probabilistic time geography. 

He has also contributed to a range of topics such as accessibility and network structure and trial and error methods of collaborative transportation such as ad hoc ride sharing, and bus on demand. Professor Winter is looking for better intelligent human-computer interaction.

While maps tell us much, they do not replace first-hand knowledge of a location.

When in a familiar environment, we use the “map in the mind”.

If we take the University of Melbourne grounds as an example, the mind map is formed by the experience of the campus: sandstone buildings from the 19th century, a clock tower, tall yellow-brick buildings from the 70s, individual experiences, a few lecture theatres and the library where long evenings have been spent over books. Can any of this information be found on a map?

“The world is still by and large un-mapped when it comes to relating to how places are experienced, what they look like, feel like – or how they are spoken about. Collecting this knowledge is still a challenge for the future, one that we might call the search for deep maps,” Professor Winter says.

“And one key source for this knowledge is, of course, human place descriptions. In a research project at the University, we set ourselves the goal to tap into this human knowledge and enrich our spatial databases, working towards these deep maps. We envision databases that store human knowledge about the environment, which reflects the way people have experienced and memorised them, rather than imposing an administrative view,” he says.

At the end of this research project Professor Winter can extract some of this human knowledge from place descriptions. 

“We detect references to locations and spatial relationships between these locations which we fill into databases. Cussonia Court (on the University’s Parkville campus) for example, is next to the ‘Old Quad’. We can compare and guess whether two place descriptions refer to the same place, and if so, whether they are compatible or contradict each other.”

People learn which places are often referred to, and have more prominence than others. 

We learn names of things that are not stored on a map. We even detect when events are used to describe locations. 

People may say: “I am at the Grand Final” instead of “I am in the Melbourne Cricket Ground”. Once detected, these event names can be stored as synonyms for particular places in the place database.

Such descriptions can be shared for collaborative and safe evacuations by our emergency services. It is a process of gaining knowledge by intelligent guesswork.

“Maps are dynamic documents, and as our technologies expand and challenge us, we respond in imaginative ways,” Professor Winter says.

 

Watch a video about this research at: http://youtu.be/8BnAN_53dcQ