Use this URL to cite or link to this record in EThOS:
Title: Modelling pedestrian systems
Author: Zachariadis, V.
ISNI:       0000 0004 5357 9386
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The thesis is organised in two parts. The objective of the first part is to review existing approaches to the simulation of microscopic pedestrian movement, to identify weaknesses and to propose an alternative model that addresses some of them. A three-layered classification framework is used to sort models based on their treatment of state-space, action-space and information-space. We identify three critical modelling parameters: the discretisation of the state space, the synchronisation of behavioural action (the cause) and its result (the effect), and the treatment of uncertainty and its impact on behaviour. We continue by presenting a pedestrian micronavigation model, which is driven by two types of actions taking place in distinct spatiotemporal contexts; behavioural actions, applied using a discrete time and action-space platform, and physical actions, triggered in continuous time and space. Finally, we present simulation outputs for simple modelling exercises and demonstrate the flexibility of the proposed modelling framework in tackling scenarios that require more complex decision-making processes. The second part focuses on route choice modelling. We start by discussing the effect of spatial morphology and configuration on pedestrian behaviour and routing decisions. Following the three-layered classification framework that was presented in the first part, we review methods of spatial abstraction and discuss their validity for representing state, action and information spaces for pedestrian routing. We propose a network-based abstraction of space derived from the visibility characteristics. In chapter 10, we address one of the most challenging aspects of pedestrian route-choice modelling: route-assignment under transient traffic conditions. The proposed route-choice models are based on structures that propagate feedback of experienced route costs and system-wide self-learning. This approach is appropriate for two reasons: it facilitates the integration of micro-navigation movement simulation with macroscopic route-choice behaviour modelling, and permits the simulation of variable levels of prior experience.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available