Spice: a cognitive agent framework for computational crowd simulations in complex environments

Kielar, Peter M.; Borrmann, André

doi:10.1007/s10458-018-9383-2

Cited by 15 publications

(5 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S5 ). In addition, for comparison, we created a second agent model that was driven by the shortest path algorithm A* 42 which is often used in the field of pedestrian modeling and represents an optimal search behavior that assumes global knowledge of the navigation environment 43 ). We used the observations of human wayfinding from the desktop VR study as our benchmark to which both shortest-path agents’ and visibility-based cognitive agents’ performances can be compared to.…”

Section: Resultsmentioning

confidence: 99%

Visibility matters during wayfinding in the vertical

Gath-Morad

Thrash

Schicker

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Visibility is the degree to which different parts of the environment can be observed from a given vantage point. In the absence of previous familiarity or signage, the visibility of key elements in a multilevel environment (e.g., the entrance, exit, or the destination itself) becomes a primary input to make wayfinding decisions and avoid getting lost. Previous research has focused on memory-based wayfinding and mental representation of 3D space, but few studies have investigated the direct effects of visibility on wayfinding. Moreover, to our knowledge, there are no studies that have explicitly observed the interaction between visibility and wayfinding under uncertainty in a multilevel environment. To bridge this gap, we studied how the visibility of destinations, as well as the continuity of sight-lines along the vertical dimension, affects unaided and goal-directed wayfinding behavior in a multilevel desktop Virtual Reality (VR) study. We obtained results from a total of 69 participants. Each participant performed a total of 24 wayfinding trials in a multilevel environment. Results showcase a significant and nonlinear correlation between the visibility of destinations and wayfinding behavioral characteristics. Specifically, once the destination was in sight, regardless of whether it was highly or barely visible, participants made an instantaneous decision to switch floors and move up towards the destination. In contrast, if the destination was out-of-sight, participants performed ‘visual exploration’, indicated by an increase in vertical head movements and greater time taken to switch floors. To demonstrate the direct applicability of this fundamental wayfinding behavioral pattern, we formalize these results by modeling a visibility-based cognitive agent. Our results show that by modeling the transition between exploration and exploitation as a function of visibility, cognitive agents were able to replicate human wayfinding patterns observed in the desktop VR study. This simple demonstration shows the potential of extending our main findings concerning the nonlinear relationship between visibility and wayfinding to inform the modeling of human cognition.

show abstract

Section: Resultsmentioning

confidence: 99%

Visibility matters during wayfinding in the vertical

Gath-Morad

Thrash

Schicker

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In the following section, we give details of the hypothesis models and data augmentations of our prototype implementation of the framework. directly dependent on the perception, memory, and tactical aspects of this pedestrian behavior architecture [1,2,3,4,14]. Memory, perception, and tactical information cannot be acquired easily in the data phase.…”

Section: Replay Phase: Basismentioning

confidence: 99%

“…The research on pedestrian walking properties identified various influence factors on movement behavior. Researchers apply these findings and develop operational models that describe peoples' walking behavior [1,2,3,4]. In computational crowd simulation, operational models are used as low-level behavior units.…”

Section: Introductionmentioning

confidence: 99%

An Artificial Neural Network Framework for Pedestrian Walking Behavior Modeling and Simulation

Kielar

Borrmann

2020

Coll Dyn

Self Cite

View full text Add to dashboard Cite

Movement behavior models of pedestrian agents form the basis of computational crowd simulations. In contemporary research, a large number of models exist. However, there is still no walking behavior model that can address the various influence factors of movement behavior holistically. Thus, we endorse the use of artificial neural networks to develop walking behavior models because machine learning methods can integrate behavioral factors efficiently, automatically, and data-driven. In this paper, we support this approach by providing a framework that describes how to include artificial neural networks into a pedestrian research context. The framework comprises 5 phases: data, replay, training, simulation, and validation. Furthermore, we describe and discuss a prototype of the framework.

show abstract

“…Understanding specific behavioral response under extreme situations can significantly affect an evacuation [2]. Agent-based models are a good candidate for crowd behavior forecasts because pedestrians can be simulated by individual agents [3]. Crowd behavior visualization is a complex multiagent modeling problem [4] and various techniques for crowd simulation have been proposed in the past 20 years [5].…”

Section: Introductionmentioning

confidence: 99%

Predictive Agent-Based Crowd Model Design Using Decentralized Control Systems

Berceanu

Banu

Husebø

et al. 2023

IEEE Systems Journal

View full text Add to dashboard Cite

As a complex system, crowd dynamics emerge bottom-up from the local interactions between pedestrians as component subsystems. This article proposes a predictive agent-based crowd simulation model to analyze the outcomes of emergency evacuation scenarios taking into account collisions between pedestrians, smoke, fire sprinklers, and exit indicators. The crowd model is based on a decentralized control system structure, where each pedestrian agent is governed through a deliberative-reactive control architecture. The simulation model for evacuation includes a routing-based control system for dynamic-guided evacuation. A design case illustrates the modeling process. Results show that the crowd simulation model based on agent autonomy and local interactions is able to generate higher level crowd dynamics through emergence.

show abstract

Spice: a cognitive agent framework for computational crowd simulations in complex environments

Cited by 15 publications

References 79 publications

Visibility matters during wayfinding in the vertical

Visibility matters during wayfinding in the vertical

An Artificial Neural Network Framework for Pedestrian Walking Behavior Modeling and Simulation

Predictive Agent-Based Crowd Model Design Using Decentralized Control Systems

Contact Info

Product

Resources

About