A Representation of Partial Spatial Knowledge: A Cognitive Map Approach for Evacuation Simulations

Article · January 2018with 28 Reads
Abstract
Usually, routing models in evacuation simulations assume that agents have comprehensive and global knowledge about the building's structure. They neglect the fact that pedestrians might possess no or only parts of information about their position relative to final exits and possible routes leading to them. For the sake of a more realistic description of the routing process, we introduce the systematics of using partial spatial knowledge. Particularly, we present an agent-based approach modeling the inaccurate mental representation of pedestrians' spatial knowledge (the cognitive map). In addition, the model considers further principles and constraints of human wayfinding. Furthermore, we present results of a field study we conducted in an office building. The purpose of this study was to investigate route choices of people in dependency on their familiarity with the building. Our modeling approach is then calibrated using the obtained results. In this context, the distribution of routes which were used by the subjects are compared with results of the model.

Do you want to read the rest of this article?

Request full-text
Request Full-text Paper PDF
  • ... This section reviews previous studies on people's evacuation behaviour with a specific focus on evacuation departure time and destination choices and highlights the need for captur- ing the interdependencies and underlying correlations between them. A substantial body of the evacuation-related literature has focused on evacuees' behaviour in terms of evacu- ation participation (see, for example, Fu and Wilmot 2004;Dash and Gladwin 2007;Hasan et al. 2011;Murray-Tuite et al. 2012) and evacuation route choice behaviour (see, for exam- ple, Carnegie and Deka 2010;Robinson and Khattak 2010;Wu, Lindell, and Prater 2012;Sadri, Ukkusuri, and Murray-Tuite 2014;Andresen, Chraibi, and Seyfried 2018). However, relatively little attention has been paid to evacuees' departure time and destination choice behaviour. ...
    Article
    Full-text available
    Over the past decades, both the frequency and severity of disastrous events have increased all over the world. Thus far, many studies have focused on people’s evacuation behaviour during advance-notice events. However, little is known about their behaviour during no-notice events mainly because of the scarcity of available data. In an effort to address such a gap, this study presents a joint model to estimate evacuee’s destination and departure time choices in the context of no-notice emergency events. These two critical decisions can directly influence spatial and temporal traffic distributions in the network in case of disastrous events. A copula-based joint structure is proposed to explore the interdependencies between these evacuation attributes that stem from the shared factors influencing them and/or the causal effects that they might have on each other. The proposed joint model comprises a multinomial logit model as the discrete component to estimate evacuation destination and an accelerated hazard model as the continuous component to estimate the departure time. The estimation results reveal that socio-economic attributes of evacuees, disaster characteristics, built-environment factors, and issuance of evacuation order by the government are key determinants of the two decisions.
  • Article
    Full-text available
    This study examined the relative influence of environmental variables (corridor width and brightness) and signage (directional and exit signs), when presented in competition, on participants' route-choices in two situational variables (everyday vs. emergency), during indoor wayfinding in virtual environments. A virtual reality-based methodology was used. Thus, participants attempted to find a room (everyday situation) in a virtual hotel, followed by a fire-related emergency egress (emergency situation). Different behaviours were observed. In the everyday situation, for no-signs condition, participants choose mostly the wider and brighter corridors, suggesting a heavy reliance on the environmental affordances. Conversely, for signs condition, participants mostly complied with signage, suggesting a greater reliance on the signs rather than on the environmental cues. During emergency, without signage, reliance on environmental affordances seems to be affected by the intersection type. In the sign condition, the reliance on environmental affordances that started strong decreases along the egress route.
  • Chapter
    Full-text available
    The navigation behaviour of pedestrians in street networks can be forecast by computer simulations based on routing models. These models characterise pedestrians’ route choices regarding a variety of factors. However, the spatial cognition aspects are often omitted in routing models; thus, the diversity of predictable routes is limited. Here, we present a unified routing model that describes route choices of pedestrians by integrating the spatial cognitive aspects of allocentric-based and egocentric-based navigation. We achieved this by combining graph-based routing methods, each formalising a single spatial cognitive aspect. In addition, we present a generic calibration method for our model. For validation, we show that our model is able to correctly predict the routing behaviour of pedestrians in a case study.
  • Article
    Full-text available
    Usually, routing models in pedestrian dynamics assume that agents have fulfilled and global knowledge about the building's structure. However, they neglect the fact that pedestrians possess no or only parts of information about their position relative to final exits and possible routes leading to them. To get a more realistic description we introduce the systematics of gathering and using spatial knowledge. A new wayfinding model for pedestrian dynamics is proposed. The model defines for every pedestrian an individual knowledge representation implying inaccuracies and uncertainties. In addition, knowledge-driven search strategies are introduced. The presented concept is tested on a fictive example scenario.
  • Conference Paper
    Full-text available
    There is a wide range of simulation software used to investigate pedestrian dynamics. Most of these software are commercial or do not give an insight into the functionalities of their models. While this might be of little importance for many users, it is paramount for researchers to know exactly what the models are performing as this provides a better interpretation of the results. In this paper we introduce JuPedSim, the Jlich Pedestrian Simulator. JuPedSim is an extensible framework for simulating and analyzing pedestrians' motion at a microscopic level. It consists at the moment of three modules which are loosely coupled and can be used independently. The first module simulates the movement of pedestrians given a geometry and an initial configuration. Three-dimensional geometries are also supported. The start configuration includes the desired destinations, speeds, route choices and other demographic parameters. The second module visualizes the geometry and the trajectories. High resolution videos can also be recorded directly from the module interface. The third module analyses the results from the simulation or any other source, for instance experiments, and generates different type of plots such as densities, velocities and flows. This paper also introduces a new force-based model, which is designed for computing pedestrian trajectories in continuous space based on repulsive and attractive forces. In addition we describe the verification and validation process of the framework.
  • Conference Paper
    Full-text available
    The life safety assessment in case of fire has already been investigated in numerous publications and many evacuation models are capable to consider fire-relevant field quantities. Commonly, the modelling of human-fire-interaction is based on a visibility-dependent reduction of movement speeds and dose-effect-relationships for toxic or irritant fire effluents. With regards to the tactical level of evacuation models, another challenge is the modelling of individual behavioural patterns like dynamic routing strategies during fire. Experimental as well as post-incident studies imply the consideration of these aspects in coupled fire and evacuation analyses. In our work, we aim to set up a modelling framework to address these patterns based on individual knowledge about a structure as well as the perception of fire-related effects. For this purpose, we present a coupled fire and evacuation analysis based on a simplified test scenario. The fire-relevant field quantities are calculated with NIST's Fire Dynamics Simulator Version 6.1.2. For the evacuation simulation, we apply the research framework JuPedSim extended by a versatile routing submodel that utilises the cognitive map concept and perception modelling. Finally, the presented modelling approach is tested with a functional verification and a parameter sensitivity analysis.
  • Article
    Route selection is a complex issue in simulating the pedestrian’s walking behaviour during the evacuation. In many current studies, the agent-based model has been widely used to simulate the individualized behaviour, and growing concern is given to the cognitive aspects in route selection. However, the influence of incomplete spatial knowledge and the preferences for different routing strategies has not been fully considered. To represent more reliable route choice by different pedestrians, this study presents an individual cognitive evacuation behaviour model which is mainly focusing on the individual differences in routing strategy and spatial knowledge. The model consists of two major components, namely the multiple routing strategies and personalized spatial cognitive road network. Based on a review of the previous studies, the potential factors that may affect the evacuation behaviour were discussed. The quantitative relationships between these potential factors and the pedestrian’s routing preference, as well as the possibilities of recognizing different roads, were explored using the regression method. As a case study, the proposed model was implemented using the data collected from a field survey of a real outdoor event. When applied in an agent-based simulation, the implemented model will support more reliable simulation of route choices corresponding to the pedestrians’ behaviour in the real event.
  • Article
    Full-text available
    Spatial memory refers to the part of the memory system that encodes, stores, recognizes and recalls spatial information about the environment and the agent’s orientation within it. Such information is required to be able to navigate to goal locations, and is vitally important for any embodied agent, or model thereof, for reaching goals in a spatially extended environment. In this paper, a number of computationally implemented cognitive models of spatial memory are reviewed and compared. Three categories of models are considered: symbolic models, neural network models, and models that are part of a systems-level cognitive architecture. Representative models from each category are described and compared in a number of dimensions along which simulation models can differ (level of modeling, types of representation, structural accuracy, generality and abstraction, environment complexity), including their possible mapping to the underlying neural substrate. Neural mappings are rarely explicated in the context of behaviourally validated models, but they could be useful to cognitive modeling research by providing a new approach for investigating a model’s plausibility. Finally, suggested experimental neuroscience methods are described for verifying the biological plausibility of computational cognitive models of spatial memory, and open questions for the field of spatial memory modeling are outlined.
  • Article
    Full-text available
    We travel through the environment to reach places that satisfy our needs and wants. Successful travel requires that we know where to go and how to get there; it also requires that we can move along the intended route in the intended direction without having accidents or getting unnecessarily delayed. Taken together, these are requirements of navigation: coordinated and goal-directed movement through the environment. Navigation occurs over a wide spectrum of spatio-temporal scales. We navigate to the other side of the room, to the post office, to visit our relatives in another town, or to vacation half way around the world. In order to navigate effectively, we apply our psychological skills of perception, cognition, and motor behavior, within the contexts of physical and social environments, and with the assistance of technologies of information and transportation. There are consequently a host of human factors issues relevant to navigation. Attention to these issues can result in improvements to the design of technologies, environments, and training procedures that increase the ease, accuracy, efficiency, comfort, and safety of navigation.
  • Article
    Understanding the influence of external information at a lower level of awareness during the processes of route selection could be a key factor to predict user's movements within complex buildings, avoiding wayfinding problems and improving egress in emergency situations. This study aims to verify whether corridor intersection configuration attributes, such as width and brightness, act as factors of attraction to improve the affordance of indoor hallways during an emergency egress situation, using a VR-based methodology. The main hypotheses are that users tend to move along either, wider or brighter corridors. Thirty volunteers participated in this study, moving along 57 different corridors, according to the experimental conditions of the study. The results suggest that people prefer to follow brighter pathways in “T-type” and “F-type” intersections, and wider corridors in “T-type” intersections. In situations where these variables are in conflict, there is a preference for brighter paths in both intersection configurations.
  • Article
    We investigated the efficiency with which men and women find their way to novel destinations using directions containing landmarks or cardinal descriptors and how such wayfinding performance is related to differences in spatial anxiety and wayfinding strategies. In two experiments, participants navigated through a model town using landmark or cardinal directions. Men and women were faster and more accurate when navigating based on cardinal directions than when navigating based on landmark directions. In addition, participants who reported greater spatial anxiety made significantly more navigation errors. As reliance on orientation strategies increased, navigation efficiency increased, suggesting that wayfinding strategies are related to navigation performance. These findings are discussed in relation to broader theoretical ideas concerning the dynamics of wayfinding processes.
  • Article
    Full-text available
    People often get lost in buildings, including but not limited to libraries, hospitals, conference centers, and shopping malls. There are at least three contributing factors: the spatial structure of the building, the cognitive maps that users construct as they navigate, and the strategies and spatial abilities of the building users. The goal of this article is to discuss recent research on each of these factors and to argue for an integrative framework that encompasses these factors and their intersections, focusing on the correspondence between the building and the cognitive map, the completeness of the cognitive map as a function of the strategies and individual abilities of the users, the compatibility between the building and the strategies and individual abilities of the users, and complexity that emerges from the intersection of all three factors. We end with an illustrative analysis in which we apply this integrative framework to difficulty in way-finding.
  • Conference Paper
    Full-text available
    Human spatial representations are known to be remarkably robust and efficient, and to be structured hierarchically. In this paper, we describe a biologically inspired computational model of spatial working memory attempting to account for these properties, based on the LIDA cognitive architecture. We also present preliminary results regarding a virtual reality experiment, which the model is able to account for, and the quantitative properties of the representation.
  • Book
    In today's complex maze of urban structures, wayfinding is no longer simply a matter of putting up directional signs, it is multi-faceted problem that requires sharp design skills. This text spells out the principles of wayfinding and applies them to architectural design. beginning with spatial, orientation, and perception factors, the book shows how well-designed routes and destinations must integrate a wide range of stimuli - graphic, verbal, auditory, and tactile. Plus, convenient checklists show how to prevent potential wayfinding problems in the design stage and to troubleshoot them in existing structures. Supported by numerous real-life examples and hundreds of lively illustrations, this book is a complete guide to spatial direction and logical orientation in buildings and public spaces - jam-packed with proven techniques for helping people find their way through today's confusing world.
  • Article
    A person's cognitive map, or knowledge of large-scale space, is built up from observations gathered as he travels through the environment. It acts as a problem-solver to find routes and relative positions, as well as describing the current location. The TOUR model describes the multiple representations that make up the cognitive map, the problem-solving strategies it uses, and the mechanisms for assimilating new information. The representations have rich collections of states of partial knowledge, which support many of the attractive qualities of common-sense knowledge.
  • Article
    This article uses a limited example to demonstrate that studies of wayfinding and spatial learning can benefit from a more rigorous analytic description of building layout and exploration paths that exhibit their own pattern. In at least one case, search patterns are strongly shaped according to the degree of integration of each space and each choice node of the circulation system within the overall layout. Given this overriding trend, we formulate the idea of a search structure whereby the intelligible properties of layouts interact with navigation rules to produce characteristic patterns of exploration.