Collective Dynamics (E-Journal)
Not a member yet
185 research outputs found
Sort by
Accurate pedestrian localization in overhead depth images via Height-Augmented HOG
Full text linkWe tackle the challenge of reliably and automatically localizing pedestrians in real-life conditions through overhead depth imaging at unprecedented high-density conditions. Leveraging upon a combination of Histogram of Oriented Gradients-like feature descriptors, neural networks, data augmentation and custom data annotation strategies, this work contributes a robust and scalable machine learning-based localization algorithm, which delivers near-human localization performance in real-time, even with local pedestrian density of about 3 ped/m2, a case in which most stateof- the art algorithms degrade significantly in performance
Determination of pedestrian’s personal space in mass religious gatherings - A case study of Kumbh Mela
Full text linkPersonal space can be defined as the physical distance between two individuals in a social environment. It varies from person to person depending on culture and context and there are significant individual differences too. Studying personal space includes the ability to recognize the various zones of involvement and the activities, relationships, and emotions associated with each zone. This paper tries to formulate and define personal space in high density crowd situations in Kumbh Mela, one of the world’s largest mass religious gatherings. Video data of pilgrims taking part in the Panchkroshi Yatra, a religious walkathon which is a part of KumbhMela, is used for the analysis of factors affecting personal space. To identify the thresholds of personal space, walking speed of individuals, gender, presence of luggage and the number of males and females surrounding an individual have been considered. It is found that the average speed of the individual, the group size,and the gender ratio of group members have a significant effect on the personal space of an individual. Also, it is observed that the personal space follows an asymmetrical pattern rather than a symmetrical pattern
Modeling Environmental Operative Elements in Agent-Based Pedestrian Simulation
Full text linkModels for pedestrian simulation are employed on a day-to-day basis for supporting the design and planning of the built environment in normal and evacuation situations. One of the aspects that are least investigated in the community, probably because it is considered closer to technology transfer than to research, is the modelling of operational elements of the simulated environment. The present paper briefly describes an agent-based approach to the representation of operative elements of the environment with particular attention to the mechanisms of interaction between these active objects and pedestrians
Social group behaviour of triads. Dependence on purpose and gender
Full text linkWe analysed a set of uninstructed pedestrian trajectories automatically tracked in a public area, and we asked a human coder to assess their group relationships. For those pedestrians who belong to the groups, we asked the coder to identify their apparent purpose of visit to the tracking area and apparent gender. We studied the quantitative dependence of the group dynamics on such properties in the case of triads (three people groups) and compared them to the two pedestrian group case (dyads), studied in a previous work. We found that the group velocity strongly depends on relation and gender for both triads and dyads, while the influence of these properties on spatial structure of groups is less clear in the triadic case. We discussed the relevance of these results to the modelling of pedestrian and crowd dynamics, and examined the possibility of the future works on this subject
Estimating social relation from trajectories
Full text linkThis study focuses on social pedestrian groups in public spaces and makes an effort to identify the social relation between the group members. We particularly consider dyads having coalitional or mating relation. We derive several observables from individual and group trajectories, which are suggested to be distinctive for these two sorts of relations and propose a recognition algorithm taking these observables as features and yielding an estimation of social relation in a probabilistic manner at every sampling step. On the average, we detect coalitional relation with 87% and mating relation with 81% accuracy. To the best of our knowledge, this is the first study to infer social relation from joint (loco)motion patterns and we consider the detection rates to be a satisfactory considering the inherent challenge of the problem
Congestion in Computational Evacuation Modelling
Full text linkThe time-based analysis of egress scenarios is a long-standing and well-established method to evaluate occupant safety. It is based on the necessary condition that the required egress time is smaller than the available egress time. The former is derived by the application of evacuation models, the latter by calculation of smoke and heat spread in the case of a fire incident. In the calculation of required egress time the time-dependent development of occupant density and consequently the emergence of congestion often play a crucial role. There is a demand to evaluate the development of local occupant density and jam situations independent of the above time-based criterion. This is for example reflected in national guidelines and standards. It is however difficult to obtain general valid evaluation criteria for congestion due to the multitude of influencing parameter and the highly situation-dependent nature of the accompanying boundary conditions. In addition, prediction of localization and duration of congestion may differ from model to model if applied to equal scenarios. Furthermore, close inspection reveals the difficulty to define proper terms for a quantitative definition of congestion. This issue is further analysed in this paper based on three case studies
Experimental study on mixed traffic flow of bicycles and pedestrians
Full text linkThe mixed flow of bicycles and pedestrians is frequently observed on bicycle-pedestrian-shared roads. Unfortunately, studies on dynamics of this kind of mixed flow are very limited. This paper reports an experimental study of this kind of mixed traffic flow with equal numbers of pedestrians and cyclists asked to walk/ride in a ring-shaped track. In the uni-/bi-directional flow scenarios, pedestrians and bicycles moved in the same/opposite direction. Under both scenarios, bicycles and pedestrians formed their own lanes. Pedestrians walked in the inner lane and cyclists rode in the outer lane. Widths of both the pedestrian lane and the bicycle lane were more uniform in bidirectional flow. The pedestrian flow rate is larger in the unidirectional flow scenario than in the bidirectional flow scenario. In contrast, at low densities, the bicycle flow rate is essentially the same under the two scenarios. When the density is large, the bicycle flow rate becomes larger in the unidirectional flow scenario. Comparing the two modes, pedestrian flow rate is smaller/larger than bicycle flow rate at small/large densities under both scenarios
Anomaly Detection of Pedestrian Flow: A Machine Learning Method for Monitoring-Data of Visitors to a Building
Full text linkMany public facilities such as community halls and gymnasiums are supposed to be evacuation sites when disasters occur. From the viewpoint of managing such facilities, it is necessary to monitor the usage and to respond immediately when an anomaly occurs. In this study, an integrated system of IoT sensors and machine learning for anomaly detection of pedestrian flow was proposed for buildings that are expected to be used as emergency evacuation sites in the event of a disaster. For trial practice of the system, infrared sensors were installed in a research building of a university, and data of visitors to the fourth floor of the building was collected as a time series data of pedestrian flow. As a result, it was shown that anomalies of pedestrian flow at an arbitrary time of a day with an occurrence probability of 5 % or less can be detected properly using the data collected
Evacuation Data from a Hospital Outpatient Drill The Case Study of North Shore Hospital
Full text linkAssessing the fire safety of buildings is fundamental to reduce the impact of this threat on their occupants. Such an assessment can be done by combining existing models and existing knowledge on how occupants behave during fires. Although many studies have been carried out for several types of built environment, only few of those investigate healthcare facilities and hospitals. In this study, we present a new behavioural data-set for hospital evacuations. The data was collected from the North Shore Hospital in Auckland (NZ) during an unannounced drill carried out in May 2017. This drill was recorded using CCTV and those videos are analysed to generate new evacuation model inputs for hospital scenarios. We collected pre-movement times, exit choices and total evacuation times for each evacuee. Moreover, we estimated pre-movement time distributions for both staff members and patients. Finally, we qualitatively investigated the evacuee actions of patients and staff members to study their interaction during the drill. The results show that participants were often independent from staff actions with a majority able to make their own decision