Architectural evaluation of simulated pedestrian spatial behaviour

Abstract

The purpose of the software system described in this article is to support architects and urban designers to better assess the impact of pedestrian behaviour on planned urban spaces and streetscapes. A new method for modelling and simulating pedestrians in urban environments is proposed. It analyses pedestrian behaviour with a combined focus on movement trajectories, walking speed and the angle between the movement vector and the gaze vector of individuals in large groups of simulated pedestrians. The system learns a statistical model characterizing normal behaviour, based on sample observations of regular pedestrian movements without the impact of significant visual attractions in the environment. Sudden changes of the pedestrians' behavioural characteristics, caused by the visual detection of ‘attractive’ objects, are considered as abnormal behaviour. The simulated environment can be automatically generated using scanned line drawings of two-dimensional street maps or public spaces. In the simulation model, a variety of scenarios can be defined and modified by altering different parameters. Using the example of Wheeler Place in Newcastle (Australia), our pilot experiments demonstrate how pedestrian behaviour characteristics can depend on selected abstract features in urban space. Among 10 different example scenarios, the analyser was able to identify the most frequently visited positions associated with attractive objects.

Keywords:

• Design evaluation
• human behaviour analysis
• pedestrian behaviour modelling

ACKNOWLEDGEMENT

This research was supported by ARC DP1092679 ‘Modelling and predicting patterns of pedestrian movement: using robotics and machine learning to improve the design of urban space’.