Revised 25 th October 2003 1 TRB Annual Meeting 2004 Monitoring the Movement of Pedestrians Using Low-cost Infrared Detectors: Initial Findings Jon Kerridge, Alistair Armitage, David Binnie and Lucy Lei {initial.surname@napier.ac.uk} Napier University, Edinburgh, Scotland, EH10 5DT Neil Sumpter {neil.sumpter@irisys.co.uk} IRISYS Ltd, Towcester, Northants, UK, NN12 6AD ABSTRACT Monitoring the movement of pedestrians in everyday environments is difficult, especially if accurate data is required. Typically data pertaining to the origins and destinations of pedestrians as they move around a space can only be obtained either by locating many observers in the area under surveillance or by analyzing CCTV footage. The former is error prone and perturbs the space being analyzed; the latter is expensive in terms of the technology and time required. Technical advances in low-cost infrared detectors provide an opportunity to unobtrusively observe pedestrian spaces and determine individual pedestrian trajectories automatically. We describe the underpinning detector technology and show how arrays of such detectors can be used to monitor larger spaces. An outline of the algorithm used to create complete trajectories as pedestrians move between detectors is presented. A series of experiments is described where pedestrians were asked to move in a set of defined patterns in a controlled environment. Initial results from these experiments are discussed where we found that at Fruin Levels of Service A to C 93% of pedestrian trajectories could be tracked, which dropped to 79% at Level of Service D. INTRODUCTION Many governments and transport authorities are firmly committed to the development of an integrated transport policy. The success of this policy will depend on ensuring that the different modes of transport work effectively and efficiently together and this, in turn, will require the optimization of flows between modes. In most cases flow between modes is in the form of a pedestrian sector in a journey. Recent attempts to improve pedestrian flow have relied on monitoring methods and systems that are expensive to implement and limited in functionality. The utility and application of these attempts to date has thus been constrained by technological and financial requirements. There are many situations where knowledge of the number and movement patterns of pedestrians would be beneficial including: • people in bus and taxi queues to ensure adequate vehicle deployment • pedestrian density within rail station concourses or on platforms to prevent overcrowding • pedestrian flow rates e.g. out of stations to enable the road vehicle management systems to optimise pedestrian and vehicular traffic movements • passenger flow rates within airports to optimize the number of passport control staff available • town planning authorities require pedestrian flow measures to monitor the viability and vitality of parts of a town. It is the last of these that provides the motivation for the research and development that has been undertaken. The viability of an area is a measure of the number of retail outlets of different types that can be sustained by the local population. The vitality is a measure of how many people use those outlets. Obtaining this last measure is a specific objective of the research programme. Figure 1. A Row of Shops to be Monitored shops Bus stop TRB 2004 Annual Meeting CD-ROM Paper revised from original submittal.