2022
Mayr, Christina Maria; Köster, Gerta
Guiding crowds when facing limited compliance: Simulating strategies Journal Article
In: PLOS ONE, vol. 17, no. 11, pp. 1-24, 2022.
@article{mayr-2022-cdyn,
title = {Guiding crowds when facing limited compliance: Simulating strategies},
author = {Christina Maria Mayr and Gerta Köster},
doi = {10.1371/journal.pone.0276229},
year = {2022},
date = {2022-11-11},
urldate = {2022-11-11},
journal = {PLOS ONE},
volume = {17},
number = {11},
pages = {1-24},
abstract = {At traffic hubs, it is important to avoid congestion of pedestrian streams to ensure safety and a good level of service. This presents a challenge, since distributing crowds on different routes is much more difficult than opening valves to, for example, regulate fluid flow. Humans may or may not comply with re-directions suggested to them typically with the help of signage, loudspeakers, apps, or by staff. This remains true, even if they perceive and understand the suggestions. Yet, simulation studies so far have neglected the influence of compliance. In view of this, we complement a state-of-the-art model of crowd motion and crowd behavior, so that we can vary the compliance rate. We consider an abstracted scenario that is inspired by a metro station in the city of Munich, where traffic regulators wish to make some passengers abandon the obviously shortest route so that the flow evens out. We investigate the effect of compliance for two very simple guiding strategies. In the first strategy, we alternate routes. In the second strategy, we recommend the path with the lowest crowd density. We observe that, in both cases, it suffices to reroute a small fraction of the crowd to reduce travel times. But we also find that taking densities into account is much more efficient when facing low compliance rates.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
At traffic hubs, it is important to avoid congestion of pedestrian streams to ensure safety and a good level of service. This presents a challenge, since distributing crowds on different routes is much more difficult than opening valves to, for example, regulate fluid flow. Humans may or may not comply with re-directions suggested to them typically with the help of signage, loudspeakers, apps, or by staff. This remains true, even if they perceive and understand the suggestions. Yet, simulation studies so far have neglected the influence of compliance. In view of this, we complement a state-of-the-art model of crowd motion and crowd behavior, so that we can vary the compliance rate. We consider an abstracted scenario that is inspired by a metro station in the city of Munich, where traffic regulators wish to make some passengers abandon the obviously shortest route so that the flow evens out. We investigate the effect of compliance for two very simple guiding strategies. In the first strategy, we alternate routes. In the second strategy, we recommend the path with the lowest crowd density. We observe that, in both cases, it suffices to reroute a small fraction of the crowd to reduce travel times. But we also find that taking densities into account is much more efficient when facing low compliance rates.
Rahn, Simon; Gödel, Marion; Köster, Gerta; Hofinger, Gesine
Modelling airborne transmission of SARS-CoV-2 at a local scale Journal Article
In: PLOS ONE, vol. 17, no. 8, 2022.
@article{rahn-2022,
title = {Modelling airborne transmission of SARS-CoV-2 at a local scale},
author = {Simon Rahn and Marion Gödel and Gerta Köster and Gesine Hofinger},
doi = {10.1371/journal.pone.0273820},
year = {2022},
date = {2022-08-30},
urldate = {2022-08-30},
journal = {PLOS ONE},
volume = {17},
number = {8},
publisher = {Public Library of Science},
abstract = {The coronavirus disease (COVID-19) pandemic has changed our lives and still poses a challenge to science. Numerous studies have contributed to a better understanding of the pandemic. In particular, inhalation of aerosolised pathogens has been identified as essential for transmission. This information is crucial to slow the spread, but the individual likelihood of becoming infected in everyday situations remains uncertain. Mathematical models help estimate such risks. In this study, we propose how to model airborne transmission of SARS-CoV-2 at a local scale. In this regard, we combine microscopic crowd simulation with a new model for disease transmission. Inspired by compartmental models, we describe virtual persons as infectious or susceptible. Infectious persons exhale pathogens bound to persistent aerosols, whereas susceptible ones absorb pathogens when moving through an aerosol cloud left by the infectious person. The transmission depends on the pathogen load of the aerosol cloud, which changes over time. We propose a 'high risk' benchmark scenario to distinguish critical from non-critical situations. A parameter study of a queue shows that the new model is suitable to evaluate the risk of exposure qualitatively and, thus, enables scientists or decision-makers to better assess the spread of COVID-19 and similar diseases.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The coronavirus disease (COVID-19) pandemic has changed our lives and still poses a challenge to science. Numerous studies have contributed to a better understanding of the pandemic. In particular, inhalation of aerosolised pathogens has been identified as essential for transmission. This information is crucial to slow the spread, but the individual likelihood of becoming infected in everyday situations remains uncertain. Mathematical models help estimate such risks. In this study, we propose how to model airborne transmission of SARS-CoV-2 at a local scale. In this regard, we combine microscopic crowd simulation with a new model for disease transmission. Inspired by compartmental models, we describe virtual persons as infectious or susceptible. Infectious persons exhale pathogens bound to persistent aerosols, whereas susceptible ones absorb pathogens when moving through an aerosol cloud left by the infectious person. The transmission depends on the pathogen load of the aerosol cloud, which changes over time. We propose a 'high risk' benchmark scenario to distinguish critical from non-critical situations. A parameter study of a queue shows that the new model is suitable to evaluate the risk of exposure qualitatively and, thus, enables scientists or decision-makers to better assess the spread of COVID-19 and similar diseases.
Lehmberg, Daniel
Operator-informed machine learning: Extracting geometry and dynamics from time series data PhD Thesis
Technical University of Munich and Hochschule München, 2022.
@phdthesis{lehmberg-2022-cdyn,
title = {Operator-informed machine learning: Extracting geometry and dynamics from time series data},
author = {Daniel Lehmberg},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
school = {Technical University of Munich and Hochschule München},
abstract = {This thesis explores an operator-informed approach to extract geometric and dynamic coordinates from time series. The main architecture consists of time delay embedding, the Laplace-Beltrami, and the Koopman operators. I transfer the numerical frameworks to a software solution. By analyzing concrete data scenarios I show that the approach is useful to accurately identify and predict the system. The model's components provide insight into the system, also in real-world and large-scale settings.},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
This thesis explores an operator-informed approach to extract geometric and dynamic coordinates from time series. The main architecture consists of time delay embedding, the Laplace-Beltrami, and the Koopman operators. I transfer the numerical frameworks to a software solution. By analyzing concrete data scenarios I show that the approach is useful to accurately identify and predict the system. The model's components provide insight into the system, also in real-world and large-scale settings.
Gödel, Marion; Lehmberg, Daniel; Brydon, Rebecca; Bosina, Ernst; Köster, Gerta
Towards learning dynamic origin-destination matrices from crowd density heatmaps Journal Article
In: JSTAT, 2022.
@article{goedel-2022b-cdyn,
title = {Towards learning dynamic origin-destination matrices from crowd density heatmaps},
author = {Marion Gödel and Daniel Lehmberg and Rebecca Brydon and Ernst Bosina and Gerta Köster},
doi = {10.1088/1742-5468/ac6255},
year = {2022},
date = {2022-01-01},
journal = {JSTAT},
abstract = {Knowing the origins and destinations of pedestrians' paths is key to the initialization of crowd simulations. Unfortunately, they are difficult to measure in the real world. This is one major challenge for live predictions during events such as festivals, soccer games, protest marches, and many others. Sensor data can be used to feed real-world observations into simulations in real-time. As input data for this study, we use density heatmaps generated from real-world trajectory data obtained from stereo sensors. Density information is compact, of constant size, and in general easier to obtain than e.g., individual trajectories. Therefore, the information limitation improves the applicability to other scenarios. We include the absolute pedestrian trip counts from origins to destinations during a brief time interval in an OD matrix, including unknown destinations due to sensor errors. Our goal is to estimate these OD matrices from a series of density heatmaps for the same interval. For this, we compute the ground truth OD matrices and density heatmaps using real-world trajectory data from a train station. We employ linear regression as a statistical learning method for estimation. We observe that the linear share of the relationship between density and OD matrix is estimated successfully. Nevertheless, a portion of the data remains that cannot be explained. We attempt to overcome this difficulty with random forest as a nonlinear model. The results indicate that both a linear and a nonlinear model can estimate some features of the OD matrices. However, there is no clear winner in terms of the chosen metric, the $R2^$ score. Overall, our findings are a strong indicator that OD matrices can indeed be estimated from density heatmaps extracted automatically from sensors.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Knowing the origins and destinations of pedestrians' paths is key to the initialization of crowd simulations. Unfortunately, they are difficult to measure in the real world. This is one major challenge for live predictions during events such as festivals, soccer games, protest marches, and many others. Sensor data can be used to feed real-world observations into simulations in real-time. As input data for this study, we use density heatmaps generated from real-world trajectory data obtained from stereo sensors. Density information is compact, of constant size, and in general easier to obtain than e.g., individual trajectories. Therefore, the information limitation improves the applicability to other scenarios. We include the absolute pedestrian trip counts from origins to destinations during a brief time interval in an OD matrix, including unknown destinations due to sensor errors. Our goal is to estimate these OD matrices from a series of density heatmaps for the same interval. For this, we compute the ground truth OD matrices and density heatmaps using real-world trajectory data from a train station. We employ linear regression as a statistical learning method for estimation. We observe that the linear share of the relationship between density and OD matrix is estimated successfully. Nevertheless, a portion of the data remains that cannot be explained. We attempt to overcome this difficulty with random forest as a nonlinear model. The results indicate that both a linear and a nonlinear model can estimate some features of the OD matrices. However, there is no clear winner in terms of the chosen metric, the $R2^$ score. Overall, our findings are a strong indicator that OD matrices can indeed be estimated from density heatmaps extracted automatically from sensors.
Gödel, Marion
Systematic parameter analysis to reduce uncertainty in crowd simulations PhD Thesis
Technical University of Munich and Hochschule München, 2022.
@phdthesis{goedel-2022c-cdyn,
title = {Systematic parameter analysis to reduce uncertainty in crowd simulations},
author = {Marion Gödel},
year = {2022},
date = {2022-01-01},
urldate = {2021-01-01},
school = {Technical University of Munich and Hochschule München},
abstract = {Crowd simulations are an indispensable tool to review evacuation concepts. Since uncertainties limit their reliability, I propose and implement a three-step approach to reduce epistemic uncertainties: I identify influential parameters, calibrate them, and quantify the output uncertainty for a safety-relevant scenario. The result is a specific model that provides predictions with considerably less uncertainty. In this way, this work contributes to increasing the reliability of crowd simulations.},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Crowd simulations are an indispensable tool to review evacuation concepts. Since uncertainties limit their reliability, I propose and implement a three-step approach to reduce epistemic uncertainties: I identify influential parameters, calibrate them, and quantify the output uncertainty for a safety-relevant scenario. The result is a specific model that provides predictions with considerably less uncertainty. In this way, this work contributes to increasing the reliability of crowd simulations.
Gödel, Marion; Bode, Nikolai; Köster, Gerta; Bungartz, Hans-Joachim
Bayesian inference methods to calibrate crowd dynamics models for safety applications Journal Article
In: Safety Science, vol. 147, pp. 105586, 2022.
@article{goedel-2022-cdyn,
title = {Bayesian inference methods to calibrate crowd dynamics models for safety applications},
author = {Marion Gödel and Nikolai Bode and Gerta Köster and Hans-Joachim Bungartz},
doi = {10.1016/j.ssci.2021.105586},
year = {2022},
date = {2022-01-01},
journal = {Safety Science},
volume = {147},
pages = {105586},
abstract = {Crowd simulation is a crucial tool to assess risks and engineer crowd safety at events and in built infrastructure. Simulations can be used for what-if studies, for real-time predictions, as well as to develop regulations for crowd safety. A reliable prediction requires a carefully calibrated model. Model parameters are often calibrated as point estimates, single parameter values for which the model evaluation fits given data best. In contrast, Bayesian inference provides a full posterior distribution for the fitted parameters that includes the residual uncertainty after calibration. In this work, we calibrate a microscopic model and an emulator derived from a microscopic model for crowd dynamics using point estimates and Approximate Bayesian Computation. We calibrate on data measuring the flow through a key scenario of crowd safety: a bottleneck. We vary the bottleneck width and demonstrate via three case studies the advantages and shortcomings of the two calibration techniques. In a case with a unimodal posterior, both methods yield similar results. However, one safety-relevant case study, that mimics the dynamics of evacuating people squeezing through an opening, exhibits a faster-is-slower dynamic where multiple free-flow speeds lead to the same flow. In this case, only Bayesian inference reveals the true bimodal shape of the posterior distribution. For multidimensional calibration, we illustrate that Bayesian inference allows accurate calibration by describing parameter relations. We conclude that, in practice, point estimation often seems sufficient, but Bayesian inference methods are necessary to capture important structural information about the uncertain parameters, and thus the physics of safety.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Crowd simulation is a crucial tool to assess risks and engineer crowd safety at events and in built infrastructure. Simulations can be used for what-if studies, for real-time predictions, as well as to develop regulations for crowd safety. A reliable prediction requires a carefully calibrated model. Model parameters are often calibrated as point estimates, single parameter values for which the model evaluation fits given data best. In contrast, Bayesian inference provides a full posterior distribution for the fitted parameters that includes the residual uncertainty after calibration. In this work, we calibrate a microscopic model and an emulator derived from a microscopic model for crowd dynamics using point estimates and Approximate Bayesian Computation. We calibrate on data measuring the flow through a key scenario of crowd safety: a bottleneck. We vary the bottleneck width and demonstrate via three case studies the advantages and shortcomings of the two calibration techniques. In a case with a unimodal posterior, both methods yield similar results. However, one safety-relevant case study, that mimics the dynamics of evacuating people squeezing through an opening, exhibits a faster-is-slower dynamic where multiple free-flow speeds lead to the same flow. In this case, only Bayesian inference reveals the true bimodal shape of the posterior distribution. For multidimensional calibration, we illustrate that Bayesian inference allows accurate calibration by describing parameter relations. We conclude that, in practice, point estimation often seems sufficient, but Bayesian inference methods are necessary to capture important structural information about the uncertain parameters, and thus the physics of safety.
2021
Lehmberg, Daniel; Dietrich, Felix; Köster, Gerta
Modeling Melburnians—Using the Koopman operator to gain insight into crowd dynamics Journal Article
In: Transportation Research Part C: Emerging Technologies, vol. 133, pp. 103437, 2021.
@article{lehmberg-2021-cdyn,
title = {Modeling Melburnians—Using the Koopman operator to gain insight into crowd dynamics},
author = { Daniel Lehmberg and Felix Dietrich and Gerta Köster},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0968090X21004265},
doi = {10.1016/j.trc.2021.103437},
year = {2021},
date = {2021-12-01},
urldate = {2021-12-01},
journal = {Transportation Research Part C: Emerging Technologies},
volume = {133},
pages = {103437},
abstract = {Describing and forecasting city traffic is challenging, given the array of factors influencing the movement of pedestrians and vehicles. Faced with this complexity, research has focused on machine learning as a way to capture spatio-temporal traffic patterns, based on past sensor data. While the methods can accurately forecast high-dimensional observations, it is often hard to explain the models and their predictions. In many cases, only the predictive performance of an otherwise black box model is evaluated. An accurate but explainable alternative is provided by the Koopman operator, an emerging methodology for data-driven identification of high-dimensional and nonlinear dynamical systems. In this study, we showcase the method by analyzing pedestrian traffic data from Melbourne, Australia. We formulate our computations in the Extended Dynamic Mode Decomposition framework, where we approximate the Koopman operator in a function basis computed with time delay embedding and the Diffusion Map algorithm. The model captures the distinct temporal patterns of 11 traffic sensors simultaneously. Because the dynamics become linear in the operator-theoretic perspective, we can decompose the model into its spectral components. Importantly, these components facilitate interpretation and can increase scientific understanding of the underlying system and the model’s operations. For the Melbourne data, we show that the spectral components connect to the underlying state space geometry and indicate the model’s stability over a prediction horizon. Our study showcases how the Koopman operator framework offers explainable and accurate data-driven predictions in a real-world traffic system. The results can easily be transferred to other traffic systems.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Describing and forecasting city traffic is challenging, given the array of factors influencing the movement of pedestrians and vehicles. Faced with this complexity, research has focused on machine learning as a way to capture spatio-temporal traffic patterns, based on past sensor data. While the methods can accurately forecast high-dimensional observations, it is often hard to explain the models and their predictions. In many cases, only the predictive performance of an otherwise black box model is evaluated. An accurate but explainable alternative is provided by the Koopman operator, an emerging methodology for data-driven identification of high-dimensional and nonlinear dynamical systems. In this study, we showcase the method by analyzing pedestrian traffic data from Melbourne, Australia. We formulate our computations in the Extended Dynamic Mode Decomposition framework, where we approximate the Koopman operator in a function basis computed with time delay embedding and the Diffusion Map algorithm. The model captures the distinct temporal patterns of 11 traffic sensors simultaneously. Because the dynamics become linear in the operator-theoretic perspective, we can decompose the model into its spectral components. Importantly, these components facilitate interpretation and can increase scientific understanding of the underlying system and the model’s operations. For the Melbourne data, we show that the spectral components connect to the underlying state space geometry and indicate the model’s stability over a prediction horizon. Our study showcases how the Koopman operator framework offers explainable and accurate data-driven predictions in a real-world traffic system. The results can easily be transferred to other traffic systems.
Rahn, Simon; Gödel, Marion; Fischer, Rainer; Köster, Gerta
Dynamics of a Simulated Demonstration March: An Efficient Sensitivity Analysis Journal Article
In: Sustainability, vol. 13, no. 6, pp. 3455, 2021.
@article{rahn-2021-cdyn,
title = {Dynamics of a Simulated Demonstration March: An Efficient Sensitivity Analysis},
author = {Simon Rahn and Marion Gödel and Rainer Fischer and Gerta Köster},
doi = {10.3390/su13063455},
year = {2021},
date = {2021-03-21},
urldate = {2021-03-21},
journal = {Sustainability},
volume = {13},
number = {6},
pages = {3455},
publisher = {MDPI},
abstract = {Protest demonstrations are a manifestation of fundamental rights. Authorities are responsible for guiding protesters safely along predefined routes, typically set in an urban built environment. Microscopic crowd simulations support decision-makers in finding sustainable crowd management strategies. Planning routes usually requires knowledge about the length of the demonstration march. This case study quantifies the impact of two uncertain parameters, the number of protesters and the standard deviation of their free-flow speeds, on the length of a protest march through Kaiserslautern, Germany. Over 1000 participants walking through more than 100,000 m2 lead to a computationally demanding model that cannot be analyzed with a standard Monte Carlo ansatz. We select and apply analysis methods that are efficient for large topographies. This combination constitutes the main novelty of this paper: We compute Sobol’ indices with two different methods, based on polynomial chaos expansions, for a down-scaled version of the original set-up and compare them to Monte Carlo computations. We employ the more accurate of the approaches for the full-scale scenario. The global sensitivity analysis reveals a shift in the governing parameter from the number of protesters to the standard deviation of their free-flow speeds over time, stressing the benefits of a time-dependent analysis. We discuss typical actions, for example floats that reduce the variation of the free-flow speed, and their effectiveness in view of the findings.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Protest demonstrations are a manifestation of fundamental rights. Authorities are responsible for guiding protesters safely along predefined routes, typically set in an urban built environment. Microscopic crowd simulations support decision-makers in finding sustainable crowd management strategies. Planning routes usually requires knowledge about the length of the demonstration march. This case study quantifies the impact of two uncertain parameters, the number of protesters and the standard deviation of their free-flow speeds, on the length of a protest march through Kaiserslautern, Germany. Over 1000 participants walking through more than 100,000 m2 lead to a computationally demanding model that cannot be analyzed with a standard Monte Carlo ansatz. We select and apply analysis methods that are efficient for large topographies. This combination constitutes the main novelty of this paper: We compute Sobol’ indices with two different methods, based on polynomial chaos expansions, for a down-scaled version of the original set-up and compare them to Monte Carlo computations. We employ the more accurate of the approaches for the full-scale scenario. The global sensitivity analysis reveals a shift in the governing parameter from the number of protesters to the standard deviation of their free-flow speeds over time, stressing the benefits of a time-dependent analysis. We discuss typical actions, for example floats that reduce the variation of the free-flow speed, and their effectiveness in view of the findings.
Kurtc, Valentina; Köster, Gerta; Fischer, Rainer
Sensitivity Analysis for Resilient Safety Design: Application to a Bottleneck Scenario Inproceedings
In: Littlewood, John; Howlett, Robert J; Jain, Lakhmi C (Ed.): Sustainability in Energy and Buildings 2020, pp. 255–264, Springer Science and Business Media Deutschland GmbH, 2021.
@inproceedings{kurtc-2021-cdyn,
title = {Sensitivity Analysis for Resilient Safety Design: Application to a Bottleneck Scenario},
author = {Valentina Kurtc and Gerta Köster and Rainer Fischer},
editor = {John Littlewood and Robert J Howlett and Lakhmi C Jain},
doi = {10.1007/978-981-15-8783-2_21},
year = {2021},
date = {2021-01-01},
booktitle = {Sustainability in Energy and Buildings 2020},
volume = {203},
pages = {255--264},
publisher = {Springer Science and Business Media Deutschland GmbH},
abstract = {Microscopic crowd simulations have become an essential tool
to devise safety strategies within built environments. Evacuating a building
means guiding pedestrians trough a series of corridors and doors, that
is, a series of bottlenecks. Thus, bottlenecks are of particular importance
in simulation studies. Simulation models depend on a number of parameters
whose exact values are often unknown. Prominent examples are the
number of pedestrians or their free-ow speeds. We carry out sensitivity
studies to analyze the system behaviour when crucial model parameters
are varied. We compute Sobol' indices using polynomial chaos expansion
to identify parameters with a strong impact on important evacuation
quantities such as the density in front of the bottleneck. A further di-
culty in this process arises from the fact that many simulation models,
including the one we use, are not strictly deterministic. In this paper,
we propose a pragmatic approach to deal with this stochasticity of the
simulation model.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Microscopic crowd simulations have become an essential tool
to devise safety strategies within built environments. Evacuating a building
means guiding pedestrians trough a series of corridors and doors, that
is, a series of bottlenecks. Thus, bottlenecks are of particular importance
in simulation studies. Simulation models depend on a number of parameters
whose exact values are often unknown. Prominent examples are the
number of pedestrians or their free-ow speeds. We carry out sensitivity
studies to analyze the system behaviour when crucial model parameters
are varied. We compute Sobol' indices using polynomial chaos expansion
to identify parameters with a strong impact on important evacuation
quantities such as the density in front of the bottleneck. A further di-
culty in this process arises from the fact that many simulation models,
including the one we use, are not strictly deterministic. In this paper,
we propose a pragmatic approach to deal with this stochasticity of the
simulation model.
to devise safety strategies within built environments. Evacuating a building
means guiding pedestrians trough a series of corridors and doors, that
is, a series of bottlenecks. Thus, bottlenecks are of particular importance
in simulation studies. Simulation models depend on a number of parameters
whose exact values are often unknown. Prominent examples are the
number of pedestrians or their free-ow speeds. We carry out sensitivity
studies to analyze the system behaviour when crucial model parameters
are varied. We compute Sobol' indices using polynomial chaos expansion
to identify parameters with a strong impact on important evacuation
quantities such as the density in front of the bottleneck. A further di-
culty in this process arises from the fact that many simulation models,
including the one we use, are not strictly deterministic. In this paper,
we propose a pragmatic approach to deal with this stochasticity of the
simulation model.
Mayr, Christina Maria; Köster, Gerta
Social distancing with the Optimal Steps Model Journal Article
In: Collective Dynamics, vol. 6, 2021.
@article{mayr-2021-cdyn,
title = {Social distancing with the Optimal Steps Model},
author = {Christina Maria Mayr and Gerta Köster},
doi = {10.17815/CD.2021.116},
year = {2021},
date = {2021-01-01},
urldate = {2020-01-01},
journal = {Collective Dynamics},
volume = {6},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Kleinmeier, Benedikt
Modeling of Behavioral Changes in Agent-Based Simulations PhD Thesis
Technische Universität München, 2021.
@phdthesis{kleinmeier-2021,
title = {Modeling of Behavioral Changes in Agent-Based Simulations},
author = {Benedikt Kleinmeier},
year = {2021},
date = {2021-01-01},
address = {München},
school = {Technische Universität München},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Zönnchen, Benedikt Sebastian
Efficient parallel algorithms for large-scale pedestrian simulation PhD Thesis
Technische Universität München, 2021.
@phdthesis{zoennchen-2021,
title = {Efficient parallel algorithms for large-scale pedestrian simulation},
author = {Benedikt Sebastian Zönnchen},
year = {2021},
date = {2021-01-01},
address = {München},
school = {Technische Universität München},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Mayr, Christina Maria; Schuhbäck, Stefan; Wischhof, Lars; Köster, Gerta
Analysis of information dissemination through direct communication in a moving crowd Journal Article
In: Safety Science, vol. 142, pp. 105386, 2021, ISSN: 0925-7535.
@article{mayr-2021,
title = {Analysis of information dissemination through direct communication in a moving crowd},
author = {Christina Maria Mayr and Stefan Schuhbäck and Lars Wischhof and Gerta Köster},
url = {https://www.sciencedirect.com/science/article/pii/S0925753521002307},
doi = {https://doi.org/10.1016/j.ssci.2021.105386},
issn = {0925-7535},
year = {2021},
date = {2021-01-01},
journal = {Safety Science},
volume = {142},
pages = {105386},
abstract = {New generation mobile communication protocols, such as the 5G standards, allow direct communication between devices. This allows to disseminate information directly in a moving crowd. In a safety concept, this information could be used to redirect pedestrians away from danger. We couple state-of-the-art computer models of pedestrian motion and mobile device-to-device communication to build a model of this complex socio-technical system. The model captures the interplay between information dissemination and human behavior. We further harness methods of uncertainty quantification to pinpoint the parameters that most influence the systems functionality for a scenario where pedestrians are redirected. We bundle successful analysis methods to suggest a procedure for future studies. We find that, in our scenario, there are rare cases of information dissemination delayed by shadowing and additional network load from apps, where agents cannot be redirected in time. Our simulation tools and methodology can help to detect such problems and serve as a basis to investigate more complex scenarios and rerouting strategies.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
New generation mobile communication protocols, such as the 5G standards, allow direct communication between devices. This allows to disseminate information directly in a moving crowd. In a safety concept, this information could be used to redirect pedestrians away from danger. We couple state-of-the-art computer models of pedestrian motion and mobile device-to-device communication to build a model of this complex socio-technical system. The model captures the interplay between information dissemination and human behavior. We further harness methods of uncertainty quantification to pinpoint the parameters that most influence the systems functionality for a scenario where pedestrians are redirected. We bundle successful analysis methods to suggest a procedure for future studies. We find that, in our scenario, there are rare cases of information dissemination delayed by shadowing and additional network load from apps, where agents cannot be redirected in time. Our simulation tools and methodology can help to detect such problems and serve as a basis to investigate more complex scenarios and rerouting strategies.
2020
Kleinmeier, Benedikt; Köster, Gerta; Drury, John
Agent-based simulation of collective cooperation: from experiment to model Journal Article
In: Journal of The Royal Society Interface, vol. 17, no. 171, pp. 20200396, 2020.
@article{kleinmeier-2020b,
title = {Agent-based simulation of collective cooperation: from experiment to model},
author = {Benedikt Kleinmeier and Gerta Köster and John Drury},
doi = {10.1098/rsif.2020.0396},
year = {2020},
date = {2020-10-07},
journal = {Journal of The Royal Society Interface},
volume = {17},
number = {171},
pages = {20200396},
abstract = {Simulation models of pedestrian dynamics have become an invaluable tool for evacuation planning. Typically, crowds are assumed to stream unidirectionally towards a safe area. Simulated agents avoid collisions through mechanisms that belong to each individual, such as being repelled from each other by imaginary forces. But classic locomotion models fail when collective cooperation is called for, notably when an agent, say a first-aid attendant, needs to forge a path through a densely packed group. We present a controlled experiment to observe what happens when humans pass through a dense static crowd. We formulate and test hypotheses on salient phenomena. We discuss our observations in a psychological framework. We derive a model that incorporates: agents’ perception and cognitive processing of a situation that needs cooperation; selection from a portfolio of behaviours, such as being cooperative; and a suitable action, such as swapping places. Agents’ ability to successfully get through a dense crowd emerges as an effect of the psychological model.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Simulation models of pedestrian dynamics have become an invaluable tool for evacuation planning. Typically, crowds are assumed to stream unidirectionally towards a safe area. Simulated agents avoid collisions through mechanisms that belong to each individual, such as being repelled from each other by imaginary forces. But classic locomotion models fail when collective cooperation is called for, notably when an agent, say a first-aid attendant, needs to forge a path through a densely packed group. We present a controlled experiment to observe what happens when humans pass through a dense static crowd. We formulate and test hypotheses on salient phenomena. We discuss our observations in a psychological framework. We derive a model that incorporates: agents’ perception and cognitive processing of a situation that needs cooperation; selection from a portfolio of behaviours, such as being cooperative; and a suitable action, such as swapping places. Agents’ ability to successfully get through a dense crowd emerges as an effect of the psychological model.
Zönnchen, Benedikt; Köster, Gerta
GPGPU Computing for Microscopic Pedestrian Simulation Inproceedings
In: Foster, Ian; Joubert, Gerhard R; Kučera, Luděk; Nagel, Wolfgang E; Peters, Frans (Ed.): Parallel Computing: Technology Trends, pp. 93-104, 2020.
@inproceedings{zoennchen-2020-cdyn,
title = {GPGPU Computing for Microscopic Pedestrian Simulation},
author = {Benedikt Zönnchen and Gerta Köster},
editor = {Ian Foster and Gerhard R Joubert and Luděk Kučera and Wolfgang E Nagel and Frans Peters},
doi = {10.3233/APC200029},
year = {2020},
date = {2020-01-01},
booktitle = {Parallel Computing: Technology Trends},
volume = {36},
pages = {93-104},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Lehmberg, Daniel; Dietrich, Felix; Köster, Gerta; Bungartz, Hans-Joachim
datafold: data-driven models for point clouds and time series on manifolds Journal Article
In: Journal of Open Source Software, vol. 5, no. 51, pp. 2283, 2020.
@article{lehmberg-2020-cdyn,
title = {datafold: data-driven models for point clouds and time series on manifolds},
author = {Daniel Lehmberg and Felix Dietrich and Gerta Köster and Hans-Joachim Bungartz},
doi = {10.21105/joss.02283},
year = {2020},
date = {2020-01-01},
journal = {Journal of Open Source Software},
volume = {5},
number = {51},
pages = {2283},
publisher = {The Open Journal},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Zönnchen, Benedikt; Kleinmeier, Benedikt; Köster, Gerta
Vadere---A simulation framework to compare locomotion models Inproceedings
In: Zuriguel, Iker; Garcimartín, Ángel; Cruz, Raúl (Ed.): Traffic and Granular Flow 2019, Springer, 2020.
@inproceedings{zoennchen-2019b-cdyn,
title = {Vadere---A simulation framework to compare locomotion models},
author = {Benedikt Zönnchen and Benedikt Kleinmeier and Gerta Köster},
editor = {Iker Zuriguel and Ángel Garcimartín and Raúl Cruz},
doi = {10.1007/978-3-030-55973-1_41},
year = {2020},
date = {2020-01-01},
booktitle = {Traffic and Granular Flow 2019},
publisher = {Springer},
series = {Springer Proceedings in Physics},
abstract = {Unlike many dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, many different approaches compete, of which some are more suitable for a specific scenario, like evacuations, than others. We showcase how to compare microscopic models based on a real-world experiment using the open-source simulation framework Vadere and two models: the Optimal Steps Model and the Behavioral Heuristics Model. Aside from quantitative aspects, we discuss visual results. Both models are able to reproduce the density-speed relation to a reasonable degree. We also identify model characteristics that led to deviations, thus enhancing our understanding of both models and facilitating the decision which model to choose to investigate a particular real-world situation.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Unlike many dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, many different approaches compete, of which some are more suitable for a specific scenario, like evacuations, than others. We showcase how to compare microscopic models based on a real-world experiment using the open-source simulation framework Vadere and two models: the Optimal Steps Model and the Behavioral Heuristics Model. Aside from quantitative aspects, we discuss visual results. Both models are able to reproduce the density-speed relation to a reasonable degree. We also identify model characteristics that led to deviations, thus enhancing our understanding of both models and facilitating the decision which model to choose to investigate a particular real-world situation.
Kleinmeier, Benedikt; Köster, Gerta; Drury, John
Agent-Based Simulation of Collective Cooperation: From Experiment to Model Journal Article
In: Journal of the Royal Society Interface, vol. 17, pp. 20200396, 2020, ISSN: 1742-5662.
@article{kleinmeier-2020-cdyn,
title = {Agent-Based Simulation of Collective Cooperation: From Experiment to Model},
author = {Benedikt Kleinmeier and Gerta Köster and John Drury},
doi = {10.1098/rsif.2020.0396},
issn = {1742-5662},
year = {2020},
date = {2020-01-01},
journal = {Journal of the Royal Society Interface},
volume = {17},
pages = {20200396},
abstract = {Simulation models of pedestrian dynamics have become an invaluable tool for evacuation planning. Typically crowds are assumed to stream unidirectionally towards a safe area. Simulated agents avoid collisions through mechanisms that belong to each individual, such as being repelled from each other by imaginary forces. But classic locomotion models fail when collective cooperation is called for, notably when an agent, say a first-aid attendant, needs to forge a path through a densely packed group. We present a controlled experiment to observe what happens when humans pass through a dense static crowd. We formulate and test hypothesis on salient phenomena. We discuss our observations in a psychological framework. We derive a model that incorporates: agents' perception and cognitive processing of a situation that needs cooperation; selection from a portfolio of behaviours, such as being cooperative; and a suitable action, such as swapping places. Agents' ability to successfully get through a dense crowd emerges as an effect of the psychological model.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Simulation models of pedestrian dynamics have become an invaluable tool for evacuation planning. Typically crowds are assumed to stream unidirectionally towards a safe area. Simulated agents avoid collisions through mechanisms that belong to each individual, such as being repelled from each other by imaginary forces. But classic locomotion models fail when collective cooperation is called for, notably when an agent, say a first-aid attendant, needs to forge a path through a densely packed group. We present a controlled experiment to observe what happens when humans pass through a dense static crowd. We formulate and test hypothesis on salient phenomena. We discuss our observations in a psychological framework. We derive a model that incorporates: agents' perception and cognitive processing of a situation that needs cooperation; selection from a portfolio of behaviours, such as being cooperative; and a suitable action, such as swapping places. Agents' ability to successfully get through a dense crowd emerges as an effect of the psychological model.
Kleinmeier, Benedikt; Köster, Gerta
Experimental Setups to Observe Evasion Maneuvers in Low and High Densities Inproceedings
In: Zuriguel, Iker; Garcimartín, Ángel; Cruz, Raúl (Ed.): Traffic and Granular Flow 2019, Springer, 2020.
@inproceedings{kleinmeier-2019b-cdyn,
title = {Experimental Setups to Observe Evasion Maneuvers in Low and High Densities},
author = {Benedikt Kleinmeier and Gerta Köster},
editor = {Iker Zuriguel and Ángel Garcimartín and Raúl Cruz},
doi = {10.1007/978-3-030-55973-1_15},
year = {2020},
date = {2020-01-01},
booktitle = {Traffic and Granular Flow 2019},
publisher = {Springer},
series = {Springer Proceedings in Physics},
abstract = {Crowd simulations depend on empirical evidence as basis for
model development. However, for many scenarios with high practical impact such evidence is still scarce. There are compelling reasons for this: Experiments involving human participants are expensive, labor intensive, and they carry a high risk for bias. This applies to both sides, experimenter and participants. In this contribution we present two experiment setups to observe pedestrian motion through high and low densities. We focus on the measures we take to avoid observer bias, undue influence on the participants and learning effects. In the first experiment, a waiting crowd of 13 participants is passed by a walking proband. In the second experiment, a waiting dyad is passed by a walking proband. Our experiment designs ensure that we can provide the scientific community with reliable data on a crowd phenomenon where evidence is still missing: single pedestrians maneuvering through a crowd.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Crowd simulations depend on empirical evidence as basis for
model development. However, for many scenarios with high practical impact such evidence is still scarce. There are compelling reasons for this: Experiments involving human participants are expensive, labor intensive, and they carry a high risk for bias. This applies to both sides, experimenter and participants. In this contribution we present two experiment setups to observe pedestrian motion through high and low densities. We focus on the measures we take to avoid observer bias, undue influence on the participants and learning effects. In the first experiment, a waiting crowd of 13 participants is passed by a walking proband. In the second experiment, a waiting dyad is passed by a walking proband. Our experiment designs ensure that we can provide the scientific community with reliable data on a crowd phenomenon where evidence is still missing: single pedestrians maneuvering through a crowd.
model development. However, for many scenarios with high practical impact such evidence is still scarce. There are compelling reasons for this: Experiments involving human participants are expensive, labor intensive, and they carry a high risk for bias. This applies to both sides, experimenter and participants. In this contribution we present two experiment setups to observe pedestrian motion through high and low densities. We focus on the measures we take to avoid observer bias, undue influence on the participants and learning effects. In the first experiment, a waiting crowd of 13 participants is passed by a walking proband. In the second experiment, a waiting dyad is passed by a walking proband. Our experiment designs ensure that we can provide the scientific community with reliable data on a crowd phenomenon where evidence is still missing: single pedestrians maneuvering through a crowd.
Gödel, Marion; Fischer, Rainer; Köster, Gerta
Sensitivity Analysis for Microscopic Crowd Simulation Journal Article
In: Algorithms, vol. 13, 2020.
@article{goedel-2020-cdyn,
title = {Sensitivity Analysis for Microscopic Crowd Simulation},
author = {Marion Gödel and Rainer Fischer and Gerta Köster},
doi = {10.3390/a13070162},
year = {2020},
date = {2020-01-01},
issuetitle = {Methods and Applications of Uncertainty Quantification in Engineering and Science},
journal = {Algorithms},
volume = {13},
abstract = {Microscopic crowd simulation can help to enhance the safety of pedestrians in situations that range from museum visits to music festivals. To obtain a useful prediction, the input parameters must be chosen carefully. In many cases, a lack of knowledge or limited measurement accuracy add uncertainty to the input. In addition, for meaningful parameter studies, we first need to identify the most influential parameters of our parametric computer models. The field of uncertainty quantification offers standardized and fully automatized methods that we believe to be beneficial for pedestrian dynamics. In addition, many methods come at a comparatively low cost, even for computationally expensive problems. This allows for their application to larger scenarios. We aim to identify and adapt fitting methods to microscopic crowd simulation in order to explore their potential in pedestrian dynamics. In this work, we first perform a variance-based sensitivity analysis using Sobol' indices and then crosscheck the results by a derivative-based measure, the activity scores. We apply both methods to a typical scenario in crowd simulation, a bottleneck. Because constrictions can lead to high crowd densities and delays in evacuations, several experiments and simulation studies have been conducted for this setting. We show qualitative agreement between the results of both methods. Additionally, we identify a one-dimensional subspace in the input parameter space and discuss its impact on the simulation. Moreover, we analyze and interpret the sensitivity indices with respect to the bottleneck scenario.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Microscopic crowd simulation can help to enhance the safety of pedestrians in situations that range from museum visits to music festivals. To obtain a useful prediction, the input parameters must be chosen carefully. In many cases, a lack of knowledge or limited measurement accuracy add uncertainty to the input. In addition, for meaningful parameter studies, we first need to identify the most influential parameters of our parametric computer models. The field of uncertainty quantification offers standardized and fully automatized methods that we believe to be beneficial for pedestrian dynamics. In addition, many methods come at a comparatively low cost, even for computationally expensive problems. This allows for their application to larger scenarios. We aim to identify and adapt fitting methods to microscopic crowd simulation in order to explore their potential in pedestrian dynamics. In this work, we first perform a variance-based sensitivity analysis using Sobol' indices and then crosscheck the results by a derivative-based measure, the activity scores. We apply both methods to a typical scenario in crowd simulation, a bottleneck. Because constrictions can lead to high crowd densities and delays in evacuations, several experiments and simulation studies have been conducted for this setting. We show qualitative agreement between the results of both methods. Additionally, we identify a one-dimensional subspace in the input parameter space and discuss its impact on the simulation. Moreover, we analyze and interpret the sensitivity indices with respect to the bottleneck scenario.
Gödel, Marion; Köster, Gerta; Lehmberg, Daniel; Gruber, Manfred; Kneidl, Angelika; Sesser, Florian
Can we learn where people go? Journal Article
In: Collective Dynamics, 2020.
@article{goedel-2018-cdyn,
title = {Can we learn where people go?},
author = {Marion Gödel and Gerta Köster and Daniel Lehmberg and Manfred Gruber and Angelika Kneidl and Florian Sesser},
doi = {10.17815/CD.2020.43},
year = {2020},
date = {2020-01-01},
booktitle = {Proceedings of Pedestrian and Evacuation Dynamics 2018},
journal = {Collective Dynamics},
abstract = {In most agent-based simulators, pedestrians navigate from origins to destinations. Consequently, destinations are essential input parameters to the simulation. While many other relevant parameters as positions, speeds and densities can be obtained from sensors, like cameras, destinations cannot be observed directly. Our research question is: Can we obtain this information from video data using machine learning methods? We usedensity heatmaps, which indicate the pedestrian density within a given camera cutout, as input to predict the destination distributions. For our proof of concept, we train a Random Forest predictor on an exemplary data set generated with the Vadere microscopic simulator. The scenario is a crossroad where pedestrians can head left, straight or right. In addition, we gain first insights on suitable placement of the camera. The results motivate an in-depth analysis of the methodology.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In most agent-based simulators, pedestrians navigate from origins to destinations. Consequently, destinations are essential input parameters to the simulation. While many other relevant parameters as positions, speeds and densities can be obtained from sensors, like cameras, destinations cannot be observed directly. Our research question is: Can we obtain this information from video data using machine learning methods? We usedensity heatmaps, which indicate the pedestrian density within a given camera cutout, as input to predict the destination distributions. For our proof of concept, we train a Random Forest predictor on an exemplary data set generated with the Vadere microscopic simulator. The scenario is a crossroad where pedestrians can head left, straight or right. In addition, we gain first insights on suitable placement of the camera. The results motivate an in-depth analysis of the methodology.
2019
Schöttl, Jakob; Seitz, Michael J; Köster, Gerta
Investigating Passengers' Seating Behavior in Suburban Trains Inproceedings
In: Hamdar, Samer H (Ed.): Traffic and Granular Flow '17, pp. 405–413, Springer International Publishing, Cham, 2019, ISBN: 978-3-030-11440-4.
@inproceedings{schoettl-2019b-cdyn,
title = {Investigating Passengers' Seating Behavior in Suburban Trains},
author = {Jakob Schöttl and Michael J Seitz and Gerta Köster},
editor = {Samer H Hamdar},
isbn = {978-3-030-11440-4},
year = {2019},
date = {2019-01-01},
booktitle = {Traffic and Granular Flow '17},
pages = {405--413},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {In pedestrian dynamics, individual-based models serve to simulate the behavior of crowds so that evacuation times and crowd densities can be estimated or the efficiency of public transportation optimized. Often train systems are investigated where seat choice may have a great impact on capacity utilization. Thus it is necessary to reproduce passengers' behavior inside trains. Yet there is surprisingly little research on the subject. In this contribution, we collect data on seating behavior in Munich's suburban trains, analyze it, and subsequently introduce a model that matches what we observe. For example, within a compartment, passengers tend to choose the seat group with the smallest number of other passengers. Within a seat group, passengers prefer window seats and forward-facing seats. When there is already another person, passengers tend to choose the seat diagonally across from that person. These and other aspects are incorporated in our model. We demonstrate the applicability of our model and present a qualitative validation with a simulation example. The model's implementation is part of the free and open-source VADERE simulation framework for pedestrian dynamics and thus available for cross-validation. The model can be used as one component in larger systems for the simulation of public transport.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In pedestrian dynamics, individual-based models serve to simulate the behavior of crowds so that evacuation times and crowd densities can be estimated or the efficiency of public transportation optimized. Often train systems are investigated where seat choice may have a great impact on capacity utilization. Thus it is necessary to reproduce passengers' behavior inside trains. Yet there is surprisingly little research on the subject. In this contribution, we collect data on seating behavior in Munich's suburban trains, analyze it, and subsequently introduce a model that matches what we observe. For example, within a compartment, passengers tend to choose the seat group with the smallest number of other passengers. Within a seat group, passengers prefer window seats and forward-facing seats. When there is already another person, passengers tend to choose the seat diagonally across from that person. These and other aspects are incorporated in our model. We demonstrate the applicability of our model and present a qualitative validation with a simulation example. The model's implementation is part of the free and open-source VADERE simulation framework for pedestrian dynamics and thus available for cross-validation. The model can be used as one component in larger systems for the simulation of public transport.
Kleinmeier, Benedikt; Zönnchen, Benedikt; Gödel, Marion; Köster, Gerta
Vadere: An Open-Source Simulation Framework to Promote Interdisciplinary Understanding Journal Article
In: Collective Dynamics, vol. 4, 2019.
@article{kleinmeier-2019-cdyn,
title = {Vadere: An Open-Source Simulation Framework to Promote Interdisciplinary Understanding},
author = {Benedikt Kleinmeier and Benedikt Zönnchen and Marion Gödel and Gerta Köster},
doi = {10.17815/CD.2019.21},
year = {2019},
date = {2019-01-01},
journal = {Collective Dynamics},
volume = {4},
abstract = {Pedestrian dynamics is an interdisciplinary field of research. Psychologists, sociologists, traffic engineers, physicists, mathematicians and computer scientists all strive to understand the dynamics of a moving crowd.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Pedestrian dynamics is an interdisciplinary field of research. Psychologists, sociologists, traffic engineers, physicists, mathematicians and computer scientists all strive to understand the dynamics of a moving crowd.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.
Zönnchen, Benedikt; Laubinger, Matthias; Köster, Gerta
Towards faster navigation algorithms on floor fields Inproceedings
In: Hamdar, Samer H (Ed.): In Traffic and Granular Flow '17, pp. 307–315, Springer International Publishing, Cham, 2019, ISBN: 978-3-030-11440-4.
@inproceedings{zoennchen-2019c,
title = {Towards faster navigation algorithms on floor fields},
author = {Benedikt Zönnchen and Matthias Laubinger and Gerta Köster},
editor = {Samer H Hamdar},
doi = {10.1007/978-3-030-11440-4_34},
isbn = {978-3-030-11440-4},
year = {2019},
date = {2019-01-01},
booktitle = {In Traffic and Granular Flow '17},
pages = {307--315},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {Many microscopic models for crowd dynamics use floor fields to navigate agents through geometries. Recently, dynamic floor fields were introduced which adapt to changes in geometry and the density of crowds. They significantly increase the realism of floor field-based simulations. However, the computation of floor fields is time consuming. In case of multiple or dynamic floor fields, which require frequent recomputations, the total simulation run time is dominated by their computation. We present an algorithm to construct floor fields for continuous space models that uses unstructured meshes. Due to the geometrical flexibility of unstructured meshes, our method reduces the computational complexity by using fewer but well-positioned mesh points.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Many microscopic models for crowd dynamics use floor fields to navigate agents through geometries. Recently, dynamic floor fields were introduced which adapt to changes in geometry and the density of crowds. They significantly increase the realism of floor field-based simulations. However, the computation of floor fields is time consuming. In case of multiple or dynamic floor fields, which require frequent recomputations, the total simulation run time is dominated by their computation. We present an algorithm to construct floor fields for continuous space models that uses unstructured meshes. Due to the geometrical flexibility of unstructured meshes, our method reduces the computational complexity by using fewer but well-positioned mesh points.
Köster, Gerta; Lehmberg, Daniel
Walking on stairs: field experiment Journal Article
In: 2019.
@article{koester-2019b-cdyn,
title = {Walking on stairs: field experiment},
author = {Gerta Köster and Daniel Lehmberg},
editor = {Forschungszentrum Jülich. Institute for Advanced Simulation 7: Civil Safety Research},
url = {https://ped.fz-juelich.de/da/doku.php?id=walking_stairs},
year = {2019},
date = {2019-01-01},
urldate = {2019-11-06},
abstract = {The basis of the data are two videos of pedestrians walking on stairs from field experiments at the University of Applied Sciences Munich (Hochschule München).},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The basis of the data are two videos of pedestrians walking on stairs from field experiments at the University of Applied Sciences Munich (Hochschule München).
Schöttl, Jakob; Seitz, Michael J; Köster, Gerta
Investigating the Randomness of Passengers' Seating Behavior in Suburban Trains Journal Article
In: Entropy, vol. 21, no. 6, 2019, ISSN: 1099-4300.
@article{schoettl-2019-cdyn,
title = {Investigating the Randomness of Passengers' Seating Behavior in Suburban Trains},
author = {Jakob Schöttl and Michael J Seitz and Gerta Köster},
doi = {10.3390/e21060600},
issn = {1099-4300},
year = {2019},
date = {2019-01-01},
journal = {Entropy},
volume = {21},
number = {6},
abstract = {In pedestrian dynamics, individual-based models serve to simulate the behavior of crowds so that evacuation times and crowd densities can be estimated or the efficiency of public transportation optimized. Often, train systems are investigated where seat choice may have a great impact on capacity utilization, especially when passengers get in each other's way. Therefore, it is useful to reproduce passengers' behavior inside trains. However, there is surprisingly little research on the subject. Do passengers distribute evenly as it is most often assumed in simulation models and as one would expect from a system that obeys the laws of thermodynamics? Conversely, is there a higher degree of order? To answer these questions, we collect data on seating behavior in Munich's suburban trains and analyze it. Clear preferences are revealed that contradict the former assumption of a uniform distribution. We subsequently introduce a model that matches the probability distributions we observed. We demonstrate the applicability of our model and present a qualitative validation with a simulation example. The model's implementation is part of the free and open-source Vadere simulation framework for pedestrian dynamics and thus available for further studies. The model can be used as one component in larger systems for the simulation of public transport.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In pedestrian dynamics, individual-based models serve to simulate the behavior of crowds so that evacuation times and crowd densities can be estimated or the efficiency of public transportation optimized. Often, train systems are investigated where seat choice may have a great impact on capacity utilization, especially when passengers get in each other's way. Therefore, it is useful to reproduce passengers' behavior inside trains. However, there is surprisingly little research on the subject. Do passengers distribute evenly as it is most often assumed in simulation models and as one would expect from a system that obeys the laws of thermodynamics? Conversely, is there a higher degree of order? To answer these questions, we collect data on seating behavior in Munich's suburban trains and analyze it. Clear preferences are revealed that contradict the former assumption of a uniform distribution. We subsequently introduce a model that matches the probability distributions we observed. We demonstrate the applicability of our model and present a qualitative validation with a simulation example. The model's implementation is part of the free and open-source Vadere simulation framework for pedestrian dynamics and thus available for further studies. The model can be used as one component in larger systems for the simulation of public transport.
Lehmberg, Daniel; Dietrich, Felix; Kevrekidis, Ioannis G; Bungartz, Hans-Joachim; Köster, Gerta
Exploring Koopman Operator Based Surrogate Models to Accelerate Analysis of Critical Pedestrian Densities Conference
Traffic and Granular Flow '19, 2019.
@conference{lehmberg-2019-cdyn,
title = {Exploring Koopman Operator Based Surrogate Models to Accelerate Analysis of Critical Pedestrian Densities},
author = {Daniel Lehmberg and Felix Dietrich and Ioannis G Kevrekidis and Hans-Joachim Bungartz and Gerta Köster},
year = {2019},
date = {2019-01-01},
booktitle = {Traffic and Granular Flow '19},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Zönnchen, Benedikt; Laubinger, Matthias; Köster, Gerta
Towards faster navigation algorithms on floor fields Inproceedings
In: Hamdar, Samer H (Ed.): In Traffic and Granular Flow '17, pp. 307–315, Springer International Publishing, Cham, 2019, ISBN: 978-3-030-11440-4.
@inproceedings{zoennchen-2019c-cdyn,
title = {Towards faster navigation algorithms on floor fields},
author = {Benedikt Zönnchen and Matthias Laubinger and Gerta Köster},
editor = {Samer H Hamdar},
doi = {10.1007/978-3-030-11440-4_34},
isbn = {978-3-030-11440-4},
year = {2019},
date = {2019-01-01},
booktitle = {In Traffic and Granular Flow '17},
pages = {307--315},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {Many microscopic models for crowd dynamics use floor fields to navigate agents through geometries. Recently, dynamic floor fields were introduced which adapt to changes in geometry and the density of crowds. They significantly increase the realism of floor field-based simulations. However, the computation of floor fields is time consuming. In case of multiple or dynamic floor fields, which require frequent recomputations, the total simulation run time is dominated by their computation. We present an algorithm to construct floor fields for continuous space models that uses unstructured meshes. Due to the geometrical flexibility of unstructured meshes, our method reduces the computational complexity by using fewer but well-positioned mesh points.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Many microscopic models for crowd dynamics use floor fields to navigate agents through geometries. Recently, dynamic floor fields were introduced which adapt to changes in geometry and the density of crowds. They significantly increase the realism of floor field-based simulations. However, the computation of floor fields is time consuming. In case of multiple or dynamic floor fields, which require frequent recomputations, the total simulation run time is dominated by their computation. We present an algorithm to construct floor fields for continuous space models that uses unstructured meshes. Due to the geometrical flexibility of unstructured meshes, our method reduces the computational complexity by using fewer but well-positioned mesh points.
Zönnchen, Benedikt; Köster, Gerta
GPGPU Computing for Microscopic Pedestrian Simulation Conference
Parallel Computing Conference, Prague, Czech Republic, 2019.
@conference{zoennchen-2019d-cdyn,
title = {GPGPU Computing for Microscopic Pedestrian Simulation},
author = {Benedikt Zönnchen and Gerta Köster},
year = {2019},
date = {2019-01-01},
booktitle = {Parallel Computing Conference},
address = {Prague, Czech Republic},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Köster, Gerta; Lehmberg, Daniel; Kneidl, Angelika
Walking on stairs: Experiment and model Journal Article
In: Phys. Rev. E, vol. 100, pp. 022310, 2019.
@article{koester-2019-cdyn,
title = {Walking on stairs: Experiment and model},
author = {Gerta Köster and Daniel Lehmberg and Angelika Kneidl},
doi = {10.1103/PhysRevE.100.022310},
year = {2019},
date = {2019-01-01},
journal = {Phys. Rev. E},
volume = {100},
pages = {022310},
publisher = {American Physical Society},
abstract = {An increasing global population forces urban planners to construct buildings and infrastructure that is extremely deep and high. Elevators and escalators serve skyscrapers and tunnels, but in an emergency people still have to walk on stairs. Computer simulations can mitigate risks of escape situations. For these situations,pedestrian locomotion models need to match reality well. Motion on stairs, however, is not nearly as well understood as motion in the plane. Publications are scarce and some arecontradictory. As a result, movement on stairs is usually modeled by slowing down pedestrians by a fixed factor. But is this justified? And what happens at intermediate landings? This contribution aims to clarify inconclusive results of previous research and provide new information to directly incorporate empirical results into a parsimoniouscomputer model. The algorithms are freely available through an open-source framework. After outlining the shortcomings of existing approaches, we present three experiments, from which we derive requirements for the computer model. Reenacting computer experiments shows the extent to which our model meets our observations. We conclude with an applied example, simulating an evacuation of Germany's famous Neuschwanstein Castle.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An increasing global population forces urban planners to construct buildings and infrastructure that is extremely deep and high. Elevators and escalators serve skyscrapers and tunnels, but in an emergency people still have to walk on stairs. Computer simulations can mitigate risks of escape situations. For these situations,pedestrian locomotion models need to match reality well. Motion on stairs, however, is not nearly as well understood as motion in the plane. Publications are scarce and some arecontradictory. As a result, movement on stairs is usually modeled by slowing down pedestrians by a fixed factor. But is this justified? And what happens at intermediate landings? This contribution aims to clarify inconclusive results of previous research and provide new information to directly incorporate empirical results into a parsimoniouscomputer model. The algorithms are freely available through an open-source framework. After outlining the shortcomings of existing approaches, we present three experiments, from which we derive requirements for the computer model. Reenacting computer experiments shows the extent to which our model meets our observations. We conclude with an applied example, simulating an evacuation of Germany's famous Neuschwanstein Castle.
Gödel, Marion; Fischer, Rainer; Köster, Gerta
Towards Inferring Input Parameters from Measurements: Bayesian Inversion for a Bottleneck Scenario Conference
Traffic and Granular Flow '19, 2019.
@conference{goedel-2019b-cdyn,
title = {Towards Inferring Input Parameters from Measurements: Bayesian Inversion for a Bottleneck Scenario},
author = {Marion Gödel and Rainer Fischer and Gerta Köster},
year = {2019},
date = {2019-01-01},
booktitle = {Traffic and Granular Flow '19},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Gödel, Marion; Fischer, Rainer; Köster, Gerta
Applying Bayesian inversion with Markov Chain Monte Carlo to Pedestrian Đynamics Conference
UNCECOMP 2019, 3rd ECCOMAS Thematic Conference on Uncertainty Quantification in Computational Sciences and Engineering, 2019.
@conference{goedel-2019-cdyn,
title = {Applying Bayesian inversion with Markov Chain Monte Carlo to Pedestrian Đynamics},
author = {Marion Gödel and Rainer Fischer and Gerta Köster},
doi = {10.7712/120219.6322.18561},
year = {2019},
date = {2019-01-01},
booktitle = {UNCECOMP 2019, 3rd ECCOMAS Thematic Conference on Uncertainty Quantification in Computational Sciences and Engineering},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Kleinmeier, Benedikt; Zönnchen, Benedikt; Gödel, Marion; Köster, Gerta
Vadere: An Open-Source Simulation Framework to Promote Interdisciplinary Understanding Journal Article
In: Collective Dynamics, vol. 4, 2019.
@article{kleinmeier-2019,
title = {Vadere: An Open-Source Simulation Framework to Promote Interdisciplinary Understanding},
author = {Benedikt Kleinmeier and Benedikt Zönnchen and Marion Gödel and Gerta Köster},
doi = {10.17815/CD.2019.21},
year = {2019},
date = {2019-01-01},
journal = {Collective Dynamics},
volume = {4},
abstract = {Pedestrian dynamics is an interdisciplinary field of research. Psychologists, sociologists, traffic engineers, physicists, mathematicians and computer scientists all strive to understand the dynamics of a moving crowd.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Pedestrian dynamics is an interdisciplinary field of research. Psychologists, sociologists, traffic engineers, physicists, mathematicians and computer scientists all strive to understand the dynamics of a moving crowd.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.
In principle, computer simulations offer means to further this understanding. Yet, unlike for many classic dynamical systems in physics, there is no universally accepted locomotion model for crowd dynamics. On the contrary, a multitude
of approaches, with very different characteristics, compete. Often only the experts in one special model type are able to assess the consequences these characteristics have on a simulation study. Therefore, scientists from all disciplines who
wish to use simulations to analyze pedestrian dynamics need a tool to compare competing approaches. Developers, too, would profit from an easy way to get insight into an alternative modeling ansatz. Vadere meets this interdisciplinary demand
by offering an open-source simulation framework that is lightweight in its approach and in its user interface while offering pre-implemented versions of the most widely spread models.
2018
Zönnchen, Benedikt; Köster, Gerta
A parallel generator for sparse unstructured meshes to solve the eikonal equation Journal Article
In: Journal of Computational Science, vol. 32, pp. 141–147, 2018, ISSN: 1877-7503.
@article{zoennchen-2018-cdyn,
title = {A parallel generator for sparse unstructured meshes to solve the eikonal equation},
author = {Benedikt Zönnchen and Gerta Köster},
doi = {10.1016/j.jocs.2018.09.009},
issn = {1877-7503},
year = {2018},
date = {2018-01-01},
journal = {Journal of Computational Science},
volume = {32},
pages = {141--147},
abstract = {Mesh generation is the first step in a wide range of applications including navigation for robots or virtual agents in pedestrian simulations. To find the shortest travel time to a target, a common technique is to solve the eikonal equation on a mesh. We propose EikMesh, an extension of the DistMesh algorithm. EikMesh is a fast parallel mesh generator that reduces the number of mesh points, and thus the computation time, while maintaining precision of numerical solvers on the mesh. It automatically refines where desired, in our case, where the eikonal equation undergoes changes, e.g. near obstacles. The first crucial step is the generation of a sophisticated initial mesh which reduces the number of smoothing steps. In addition, EikMesh avoids expensive Delaunay-re-triangulations. Space filling curves manage storage space in a cache-friendly manner. EikMesh scales better than the parallelized traditional DistMesh and significantly outperforms it for a number of test cases.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mesh generation is the first step in a wide range of applications including navigation for robots or virtual agents in pedestrian simulations. To find the shortest travel time to a target, a common technique is to solve the eikonal equation on a mesh. We propose EikMesh, an extension of the DistMesh algorithm. EikMesh is a fast parallel mesh generator that reduces the number of mesh points, and thus the computation time, while maintaining precision of numerical solvers on the mesh. It automatically refines where desired, in our case, where the eikonal equation undergoes changes, e.g. near obstacles. The first crucial step is the generation of a sophisticated initial mesh which reduces the number of smoothing steps. In addition, EikMesh avoids expensive Delaunay-re-triangulations. Space filling curves manage storage space in a cache-friendly manner. EikMesh scales better than the parallelized traditional DistMesh and significantly outperforms it for a number of test cases.
Künzner, Florian; Köster, Gerta; Dietrich, Felix
Efficient quantification of uncertainties when de-boarding a train Conference
Pedestrian and Evacuation Dynamics 2018, 2018.
@conference{kuenzner-2018-cdyn,
title = {Efficient quantification of uncertainties when de-boarding a train},
author = {Florian Künzner and Gerta Köster and Felix Dietrich},
year = {2018},
date = {2018-01-01},
booktitle = {Pedestrian and Evacuation Dynamics 2018},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Dietrich, Felix; Künzner, Florian; Neckel, Tobias; Köster, Gerta; Bungartz, Hans-Joachim
Fast and flexible uncertainty quantification through a data-driven surrogate model Journal Article
In: International Journal for Uncertainty Quantification, vol. 8, pp. 175–192, 2018.
@article{dietrich-2018-cdyn,
title = {Fast and flexible uncertainty quantification through a data-driven surrogate model},
author = {Felix Dietrich and Florian Künzner and Tobias Neckel and Gerta Köster and Hans-Joachim Bungartz},
doi = {10.1615/Int.J.UncertaintyQuantification.2018021975},
year = {2018},
date = {2018-01-01},
journal = {International Journal for Uncertainty Quantification},
volume = {8},
pages = {175--192},
abstract = {To assess a computer model's descriptive and predictive power, the model's response to uncertainties in the input must be quantified. However, simulations of complex systems typically need a lot of computational resources, and thus prohibit exhaustive sweeps of high-dimensional spaces. Moreover, the time available to compute a result for decision systems is often very limited. In this paper, we construct a data-driven surrogate model from time delays of observations of a complex, microscopic model. We employ diffusion maps to reduce the dimensionality of the delay space. The surrogate model allows faster generation of the quantity of interest over time than the original, microscopic model. It is a non-intrusive method, and hence does not need access to the model formulation. In contrast to most other surrogate approaches, the construction allows quantities of interest that are not closed dynamically, because a closed state space is constructed through Takens delay embedding. Also, the surrogate can be stored to and loaded from storage with very little effort. The surrogate model is decoupled from the original model, and the fast execution speed allows to quickly evaluate many different parameter distributions. We demonstrate the capability of the approach in combination with forward UQ on a parametrized Burgers? equation, and the microscopic simulation of a train station. The surrogate model can accurately capture the dynamical features in both examples, with relative errors always smaller than ten percent. The simulation time in the real-world example can be reduced by an order of magnitude.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
To assess a computer model's descriptive and predictive power, the model's response to uncertainties in the input must be quantified. However, simulations of complex systems typically need a lot of computational resources, and thus prohibit exhaustive sweeps of high-dimensional spaces. Moreover, the time available to compute a result for decision systems is often very limited. In this paper, we construct a data-driven surrogate model from time delays of observations of a complex, microscopic model. We employ diffusion maps to reduce the dimensionality of the delay space. The surrogate model allows faster generation of the quantity of interest over time than the original, microscopic model. It is a non-intrusive method, and hence does not need access to the model formulation. In contrast to most other surrogate approaches, the construction allows quantities of interest that are not closed dynamically, because a closed state space is constructed through Takens delay embedding. Also, the surrogate can be stored to and loaded from storage with very little effort. The surrogate model is decoupled from the original model, and the fast execution speed allows to quickly evaluate many different parameter distributions. We demonstrate the capability of the approach in combination with forward UQ on a parametrized Burgers? equation, and the microscopic simulation of a train station. The surrogate model can accurately capture the dynamical features in both examples, with relative errors always smaller than ten percent. The simulation time in the real-world example can be reduced by an order of magnitude.
2017
Köster, Gerta; Schöttl, Jakob; Seitz., Michael J
Investigating passengers' seating behavior in suburban trains Conference
In Traffic and Granular Flow'17, 2017.
@conference{koster-2017-cdyn,
title = {Investigating passengers' seating behavior in suburban trains},
author = {Gerta Köster and Jakob Schöttl and Michael J Seitz.},
year = {2017},
date = {2017-01-01},
booktitle = {In Traffic and Granular Flow'17},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Zönnchen, Benedikt; Laubinger, Matthias; Köster, Gerta
Towards faster navigation algorithms on floor fields Conference
In Traffic and Granular Flow '17, 2017.
@conference{zoennchen-2017-cdyn,
title = {Towards faster navigation algorithms on floor fields},
author = {Benedikt Zönnchen and Matthias Laubinger and Gerta Köster},
year = {2017},
date = {2017-01-01},
booktitle = {In Traffic and Granular Flow '17},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Dietrich, Felix
Data-Driven Surrogate Models for Dynamical Systems PhD Thesis
Technische Universität München, 2017.
@phdthesis{dietrich-2017,
title = {Data-Driven Surrogate Models for Dynamical Systems},
author = {Felix Dietrich},
year = {2017},
date = {2017-01-01},
address = {München},
school = {Technische Universität München},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
2016
Modellierung sozialpsychologischer Faktoren in Personenstromsimulationen PhD Thesis
Technische Universität München, 2016.
@phdthesis{,
title = {Modellierung sozialpsychologischer Faktoren in Personenstromsimulationen},
year = {2016},
date = {2016-01-01},
address = {München},
school = {Technische Universität München},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Seitz, Michael J
Simulating pedestrian dynamics PhD Thesis
Technische Universität München, 2016.
@phdthesis{seitz-2016b,
title = {Simulating pedestrian dynamics},
author = {Michael J Seitz},
year = {2016},
date = {2016-01-01},
address = {München},
school = {Technische Universität München},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Seitz, Michael J; Dietrich, Felix; Köster, Gerta; Bungartz, Hans-Joachim
The superposition principle: A conceptual perspective on pedestrian stream simulations Journal Article
In: Collective Dynamics, vol. 1, pp. A2, 2016.
@article{seitz-2016b-cdyn,
title = {The superposition principle: A conceptual perspective on pedestrian stream simulations},
author = {Michael J Seitz and Felix Dietrich and Gerta Köster and Hans-Joachim Bungartz},
doi = {10.17815/CD.2016.2},
year = {2016},
date = {2016-01-01},
journal = {Collective Dynamics},
volume = {1},
pages = {A2},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
von Sivers, Isabella; Künzner, Florian; Köster, Gerta
Pedestrian Evacuation Simulation with Separated Families Conference
Proceedings of the 8th International Conference on Pedestrian and Evacuation Dynamics (PED2016), Hefei, China, 2016.
@conference{sivers-2016c-cdyn,
title = {Pedestrian Evacuation Simulation with Separated Families},
author = {Isabella von Sivers and Florian Künzner and Gerta Köster},
year = {2016},
date = {2016-01-01},
booktitle = {Proceedings of the 8th International Conference on Pedestrian and Evacuation Dynamics (PED2016)},
address = {Hefei, China},
abstract = {Crowds are composed of both single persons and small to large groups of people. A family can
be considered a special group within a crowd because of its unique behaviour. While families may become
separated at the beginning of an emergency situation, they tend to evacuate the situation together. That is,
according to ndings in psychology, family members search for each other and evacuate once they are reunited.
However, it is not exceptional that families are separated at the beginning of an emergency. According to
psychological ndings, family members search for each other and evacuate after they are reunited. The model
presented in this paper transfers these ndings into pedestrian evacuation simulation. We describe how we
model the behaviour of separated families and qualitatively validate the model. With the help of uncertainty
quantication and Sobol indices, we analyse the impact of three uncertain parameters of the model on the
evacuation times: the percentage of family members in the crowd, the speed at which parents search for their
children, and the speed of the children evacuating with their parents. As a result, we can show that it is vital
to consider families in evacuation simulation to better estimate of the evacuation times.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Crowds are composed of both single persons and small to large groups of people. A family can
be considered a special group within a crowd because of its unique behaviour. While families may become
separated at the beginning of an emergency situation, they tend to evacuate the situation together. That is,
according to ndings in psychology, family members search for each other and evacuate once they are reunited.
However, it is not exceptional that families are separated at the beginning of an emergency. According to
psychological ndings, family members search for each other and evacuate after they are reunited. The model
presented in this paper transfers these ndings into pedestrian evacuation simulation. We describe how we
model the behaviour of separated families and qualitatively validate the model. With the help of uncertainty
quantication and Sobol indices, we analyse the impact of three uncertain parameters of the model on the
evacuation times: the percentage of family members in the crowd, the speed at which parents search for their
children, and the speed of the children evacuating with their parents. As a result, we can show that it is vital
to consider families in evacuation simulation to better estimate of the evacuation times.
be considered a special group within a crowd because of its unique behaviour. While families may become
separated at the beginning of an emergency situation, they tend to evacuate the situation together. That is,
according to ndings in psychology, family members search for each other and evacuate once they are reunited.
However, it is not exceptional that families are separated at the beginning of an emergency. According to
psychological ndings, family members search for each other and evacuate after they are reunited. The model
presented in this paper transfers these ndings into pedestrian evacuation simulation. We describe how we
model the behaviour of separated families and qualitatively validate the model. With the help of uncertainty
quantication and Sobol indices, we analyse the impact of three uncertain parameters of the model on the
evacuation times: the percentage of family members in the crowd, the speed at which parents search for their
children, and the speed of the children evacuating with their parents. As a result, we can show that it is vital
to consider families in evacuation simulation to better estimate of the evacuation times.
von Sivers, Isabella; Seitz, Michael Jakob; Köster, Gerta
How do people search: a modelling perspective Inproceedings
In: Parallel Processing and Applied Mathematics, 11th International Conference, PPAM 2015, Krakow, Poland, September 6-9, 2015. Revised Selected Papers, Part II, pp. 487–496, Springer, 2016.
@inproceedings{sivers-2015b-cdyn,
title = {How do people search: a modelling perspective},
author = {Isabella von Sivers and Michael Jakob Seitz and Gerta Köster},
doi = {10.1007/978-3-319-32152-3_45},
year = {2016},
date = {2016-01-01},
booktitle = {Parallel Processing and Applied Mathematics, 11th International Conference, PPAM 2015, Krakow, Poland, September 6-9, 2015. Revised Selected Papers, Part II},
volume = {9574},
pages = {487--496},
publisher = {Springer},
series = {Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Seitz, Michael J; Dietrich, Felix; Köster, Gerta; Bungartz, Hans-Joachim
The superposition principle: A conceptual perspective on pedestrian stream simulations Journal Article
In: Collective Dynamics, vol. 1, pp. A2, 2016.
@article{seitz-2016b,
title = {The superposition principle: A conceptual perspective on pedestrian stream simulations},
author = {Michael J Seitz and Felix Dietrich and Gerta Köster and Hans-Joachim Bungartz},
doi = {10.17815/CD.2016.2},
year = {2016},
date = {2016-01-01},
journal = {Collective Dynamics},
volume = {1},
pages = {A2},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
von Sivers, Isabella; Templeton, Anne; Künzner, Florian; Köster, Gerta; Drury, John; Philippides, Andrew; Neckel, Tobias; Bungartz, Hans-Joachim
Modelling social identification and helping in evacuation simulation Journal Article
In: Safety Science, vol. 89, pp. 288–300, 2016, ISSN: 0925-7535.
@article{sivers-2016d,
title = {Modelling social identification and helping in evacuation simulation},
author = {Isabella von Sivers and Anne Templeton and Florian Künzner and Gerta Köster and John Drury and Andrew Philippides and Tobias Neckel and Hans-Joachim Bungartz},
doi = {10.1016/j.ssci.2016.07.001},
issn = {0925-7535},
year = {2016},
date = {2016-01-01},
journal = {Safety Science},
volume = {89},
pages = {288--300},
abstract = {Social scientists have criticised computer models of pedestrian streams for their treatment of psychological
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Social scientists have criticised computer models of pedestrian streams for their treatment of psychological
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.
Seitz, Michael J; Bode, Nikolai W F; Köster, Gerta
How cognitive heuristics can explain social interactions in spatial movement Journal Article
In: Journal of the Royal Society Interface, vol. 13, no. 121, pp. 20160439, 2016.
@article{seitz-2016c,
title = {How cognitive heuristics can explain social interactions in spatial movement},
author = {Michael J Seitz and Nikolai W F Bode and Gerta Köster},
doi = {10.1098/rsif.2016.0439},
year = {2016},
date = {2016-01-01},
journal = {Journal of the Royal Society Interface},
volume = {13},
number = {121},
pages = {20160439},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Köster, Gerta; Lehmberg, Daniel; Dietrich, Felix
Is slowing down enough to model movement on stairs? Inproceedings
In: Knoop, Victor L; Daamen, Winnie (Ed.): Traffic and Granular Flow '15, pp. 35–42, Springer International Publishing, Nootdorp, the Netherlands, 2016, (27--30 October 2015).
@inproceedings{koster-2015-cdyn,
title = {Is slowing down enough to model movement on stairs?},
author = {Gerta Köster and Daniel Lehmberg and Felix Dietrich},
editor = {Victor L Knoop and Winnie Daamen},
year = {2016},
date = {2016-01-01},
booktitle = {Traffic and Granular Flow '15},
pages = {35--42},
publisher = {Springer International Publishing},
address = {Nootdorp, the Netherlands},
abstract = {There are many well validated models of pedestrian movement on a flat
surface. This is not the case for movement on stairs. Experiments show that pedestrians
slow down when climbing or descending stairs. Hence, it is tempting to model
movement on stairs by simply slowing down by a factor. But this would imply that,
other than being slower, motion on stairs mirrors motion in the plane. Is that assumption
justified? We conduct field observations that reveal similarities but also
significant differences. Thus we argue that modeling movement on stairs by slowing
down free-flow velocities may be an acceptable first shot. True microscopic
behavior, however, like treading from step to step and keeping to a straight line instead
of trying to overtake can only be captured by a dedicated model. We present
an extension to the Optimal Steps Model that achieves this.},
note = {27--30 October 2015},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
There are many well validated models of pedestrian movement on a flat
surface. This is not the case for movement on stairs. Experiments show that pedestrians
slow down when climbing or descending stairs. Hence, it is tempting to model
movement on stairs by simply slowing down by a factor. But this would imply that,
other than being slower, motion on stairs mirrors motion in the plane. Is that assumption
justified? We conduct field observations that reveal similarities but also
significant differences. Thus we argue that modeling movement on stairs by slowing
down free-flow velocities may be an acceptable first shot. True microscopic
behavior, however, like treading from step to step and keeping to a straight line instead
of trying to overtake can only be captured by a dedicated model. We present
an extension to the Optimal Steps Model that achieves this.
surface. This is not the case for movement on stairs. Experiments show that pedestrians
slow down when climbing or descending stairs. Hence, it is tempting to model
movement on stairs by simply slowing down by a factor. But this would imply that,
other than being slower, motion on stairs mirrors motion in the plane. Is that assumption
justified? We conduct field observations that reveal similarities but also
significant differences. Thus we argue that modeling movement on stairs by slowing
down free-flow velocities may be an acceptable first shot. True microscopic
behavior, however, like treading from step to step and keeping to a straight line instead
of trying to overtake can only be captured by a dedicated model. We present
an extension to the Optimal Steps Model that achieves this.
Köster, Gerta; Zönnchen, Benedikt
A Queuing Model Based On Social Attitudes Inproceedings
In: Knoop, Victor L; Daamen, Winnie (Ed.): Traffic and Granular Flow '15, pp. 193–200, Springer International Publishing, Nootdorp, the Netherlands, 2016, (27--30 October 2015).
@inproceedings{koster-2015b-cdyn,
title = {A Queuing Model Based On Social Attitudes},
author = {Gerta Köster and Benedikt Zönnchen},
editor = {Victor L Knoop and Winnie Daamen},
doi = {10.1007/978-3-319-33482-0},
year = {2016},
date = {2016-01-01},
booktitle = {Traffic and Granular Flow '15},
pages = {193--200},
publisher = {Springer International Publishing},
address = {Nootdorp, the Netherlands},
abstract = {Modern pedestrian simulation models have to deal with queuing to obtain
realistic results. Queues control the number of pedestrians entering or leaving an
area and, through this, the number of pedestrians inside that area. Furthermore they
impede passing pedestrians. But how do humans decide on a queuing strategy? And
how does this effect the form of the emerging queue? Based on dynamic floor fields
for navigation and a simple heuristic decision mechanism we present a computer
model that is able to capture different queuing patterns that we observe in every day
life. For this we assume that there are two basic attitudes, aggressive competition
and cooperative getting in line. Pedestrians can switch between these strategies.},
note = {27--30 October 2015},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Modern pedestrian simulation models have to deal with queuing to obtain
realistic results. Queues control the number of pedestrians entering or leaving an
area and, through this, the number of pedestrians inside that area. Furthermore they
impede passing pedestrians. But how do humans decide on a queuing strategy? And
how does this effect the form of the emerging queue? Based on dynamic floor fields
for navigation and a simple heuristic decision mechanism we present a computer
model that is able to capture different queuing patterns that we observe in every day
life. For this we assume that there are two basic attitudes, aggressive competition
and cooperative getting in line. Pedestrians can switch between these strategies.
realistic results. Queues control the number of pedestrians entering or leaving an
area and, through this, the number of pedestrians inside that area. Furthermore they
impede passing pedestrians. But how do humans decide on a queuing strategy? And
how does this effect the form of the emerging queue? Based on dynamic floor fields
for navigation and a simple heuristic decision mechanism we present a computer
model that is able to capture different queuing patterns that we observe in every day
life. For this we assume that there are two basic attitudes, aggressive competition
and cooperative getting in line. Pedestrians can switch between these strategies.
von Sivers, Isabella Katharina Maximiliana; Templeton, Anne; Künzner, Florian; Köster, Gerta; Drury, John; andNeckel, Tobias Philippides Andrew; Bungartz, Hans-Joachim
Modelling social identification and helping in evacuation simulation Journal Article
In: Safety Science, vol. 89, pp. 288–300, 2016, ISSN: 0925-7535.
@article{sivers-2016d-cdyn,
title = {Modelling social identification and helping in evacuation simulation},
author = {Isabella Katharina Maximiliana von Sivers and Anne Templeton and Florian Künzner and Gerta Köster and John Drury and Tobias Philippides Andrew andNeckel and Hans-Joachim Bungartz},
doi = {10.1016/j.ssci.2016.07.001},
issn = {0925-7535},
year = {2016},
date = {2016-01-01},
journal = {Safety Science},
volume = {89},
pages = {288--300},
abstract = {Social scientists have criticised computer models of pedestrian streams for their treatment of psychological
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Social scientists have criticised computer models of pedestrian streams for their treatment of psychological
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.
crowds as mere aggregations of individuals. Indeed most models for evacuation dynamics use analogies
from physics where pedestrians are considered as particles. Although this ensures that the results of
the simulation match important physical phenomena, such as the deceleration of the crowd with
increasing density, social phenomena such as group processes are ignored. In particular, people in a
crowd have social identities and share those social identities with the others in the crowd. The process
of self categorisation determines norms within the crowd and influences how people will behave in
evacuation situations. We formulate the application of social identity in pedestrian simulation
algorithmically. The goal is to examine whether it is possible to carry over the psychological model to
computer models of pedestrian motion so that simulation results correspond to observations from crowd
psychology. That is, we quantify and formalise empirical research on and verbal descriptions of the effect
of group identity on behaviour. We use uncertainty quantification to analyse the model's behaviour when
we vary crucial model parameters. In this first approach we restrict ourselves to a specific scenario that
was thoroughly investigated by crowd psychologists and where some quantitative data is available: the
bombing and subsequent evacuation of a London underground tube carriage on July 7th 2005.