Research

Pedestrian Models: Exosomatic Visual Architecture

Spatial agents

Space syntax deals with the configurational properties of environments. Spatial agents use vision to assess the configuration, and move towards open space by a stochastic process: choosing a destination at random from the available space, and walking towards it. In this way, they are configurational explorers. The rules are: walk 3 steps, look around and choose a new destination, walk 3 steps, and so on. If their field of view is set to 170º (approximating human vision) the agents start to move, on aggregate, in a human like manner. The images show agent trails: as the agents walk over 1m grid squares they lay trails behind them. Black areas represent few agent trails and white areas many agent trails.

The scatterplots show agent gate counts against people gate counts for an area of the City around the Baltic House site, and the Tate Britain Gallery on Millbank (agent trails for the two layouts are shown on the left). The agents demonstrate good correspondence with aggregate people movement patterns. In the City, there is a correlation coefficient o f R² = 0.67 with actual people gate counts and in the Tate gallery R² = 0.76. For comparison, axial line analysis of the City gives R² = 0.66 and convaxial analysis of the Tate gallery gives R² = 0.68. The correlations are similar, so do spatial agents represent the dynamic form of space syntax? The answer is not simple. The agents correlate with space syntax no better than with reality: in the City R² = 0.67 for agents against axial analysis. Is there something else the agents are capturing that space syntax is not? Again, the answer is not simple. If we perform a multiple regression of axial analysis and agents against people movement in the City, there is a relationship of R² = 0.73. There is something independent that the two analyses are capturing, but little of it is to do with people movement. The fundamental question is still unanswered: what is it about the way people approach the configuration that space syntax is finding?

Further details published in:
Turner, A. and Penn, A., 2002, Encoding natural movement as an agent-based system: an investigation into human pedestrian behaviour in the built environment. Environment and Planning B: Planning and Design 29:473–490
Turner, A., 2003, Analysing the visual dynamics of spatial morphology. Environment and Planning B: Planning and Design 30:657–676

Download EVAS software

City and Tate Gallery movement data was collected by the Space Syntax Laboratory. The help of Space Syntax Limited is gratefully acknowledged.