With the increasing number of studies on crowd behavior analysis, there has been a widespread interest in treating subgroups as an important topic. A previous experimental study has investigated the decision-making and motion behavior of subgroups when facing a static obstacle during movement. However, it is hard to quantify social attributes (e.g., interpersonal relationships and sense of identity) and little is known about how they affect the movement process of subgroups. Here, we propose a vision-driven model to solve this problem, in which two key model parameters are defined to control the spatial cohesion and attraction intensity, respectively. Numerical simulations demonstrate that the optimal regions of model parameters vary depending on different conditions of the three control variables (obstacle width, time pressure, and subgroup size). The spatial cohesion and attraction intensity barely change the movement process of subgroups in the maintaining state but significantly affect it in the splitting-merging state. This model can reproduce the herding effect of subgroup members in the merging process, which is affected to varying degrees by the modulation of model parameters. Overall, this work contributes to the simulation of subgroup behaviors from a sociopsychological perspective.

Due to the universal existence of human subgroups in reality, an increasing number of studies have incorporated them into the modeling process of crowd evacuation. However, such models seldom explain subgroup behaviors from the aspect of what individuals see and how they respond to visual input. Here, we propose a vision-driven model based on cognitive heuristics, in which the mechanisms of avoidance with the environment and attraction to other members within the field of view are explicitly clarified. Numerical simulations demonstrate that various spatial characteristics of subgroup members can be effectively represented by this model, and both the intensity and heterogeneity of spatial cohesion have significant impacts on subgroup evacuation. By comparing with an empirical study, the reproducibility of our model has been validated in terms of the temporal and spatial dimensions. This model produces more natural and realistic subgroup behaviors in multiple interaction contexts than existing models, and also quantitatively exhibits the superiority in reproduction effects. Overall, this work provides an interpretable mathematical framework for modeling subgroups from the perspective of visual perception.

In today’s high-density urban environments, understanding pedestrian behavior in emergency evacuations is increasingly crucial. This study develops a sophisticated model integrating small-world network dynamics with emotional contagion to dissect pedestrian behaviors in such situations. Utilizing agent-based and complex network analyses, it delves into phase transitions in pedestrian order under various risk scenarios. The findings demonstrate significant behavioral variations in medium-risk environments, which are especially susceptible to transitions from orderly to disorderly states. This model emphasizes the important roles of emotional contagion in shaping crowd dynamics and the psychological factors alongside physical ones in evacuation strategies. Notably, this research illuminates the impacts of crowd size and speed on disorder evolution, which offers valuable insights for urban planning, architectural design, intelligent transportation systems, and emergency management. Overall, our study enriches the understanding of pedestrian dynamics in emergencies and provides a foundation for developing more effective public safety measures.

The study of identifying human subgroups from videos is a significant topic, which has received a lot of attention in multiple disciplines. So far, however, there has been little consideration about combining it with relevant conceptions in network science. Therefore, this article proposes a novel method for the automatic identification of human subgroups in dynamic pedestrian flows. The spatial proximity and temporal continuity are combined to calculate the interaction intensity between pedestrians, by which a time-dependent pedestrian flow network is constructed. Based on the objective function of weighted partition density, the optimal threshold is used to determine community structures that correspond to human subgroups in frame images. Numerical experiments demonstrate that our method achieves high identification accuracy under various evaluation datasets, and exhibits better performance than existing methods in terms of different crowd densities, various numbers of subgroup members, and certain levels of trajectory noise. Furthermore, this work provides valuable implications for the understanding of subgroup behaviors and the modeling of subgroup movements.

The study of crowd evacuation has received considerable attention as the frequent occurrence of crowd disasters in public places. Notably, the increasing proportion of disabled pedestrians makes vulnerable crowds an indispensable part of the evacuation process. However, most previous research neglects to introduce the motion characteristics of disabled pedestrians into the modeling of crowd evacuation. Therefore, we develop an extended model to simulate the evacuation process involving nondisabled, visual-disabled, acoustic-disabled, and physical-disabled pedestrians. Numerical simulations indicate that this model achieves a more realistic mixed crowd evacuation in the library scene and reproduces the escape movement of multitype disabled pedestrians. Moreover, several management strategies are provided to guide the evacuation of disabled pedestrians, and the appropriate strategy can be determined by comprehensively considering multiple factors such as efficiency, safety, and cost.

Human crowds exhibit rich self-organizing behaviors through local interactions. The understanding of interaction mechanisms has important implications for the management of large-scale crowds. Although most vision-based heuristic models are successful, some features such as distracted pedestrians, and empirical phenomena like sudden turns are difficult to be explained. Here, a heuristic interaction model is proposed, which incorporates the extracted laws of pedestrian heterogeneity and mutual anticipation. We argue that pedestrians are heterogeneous in terms of speed and attention (e.g., distracted by cell phones), and have anticipations for other pedestrians’ velocities during the interaction. Numerical simulations indicate that our model realistically simulates the self-organizing phenomenon in the “distraction experiment,” along with related experimental findings. The “freezing-by-heating” phenomenon, as well as interesting phenomena such as “sidewalk shuffling” are successfully predicted, as exhibited in empirical observations. Taken together, our model may serve various potential fields involving the control of traffic flows and navigation of autonomous swarm robots.

With the increasing scale of crowds in public places, the study of modeling crowd evacuation has become a significant research field. However, most previous research ignores to incorporate behavioral heterogeneity of individuals into the modeling framework, making it hard to replicate more realistic evacuation processes. Therefore, a behavioral heterogeneity-based social force model (BHSFM) is proposed to reveal the heterogeneity characteristics from the aspect of individual behavior. Numerical experiments show that the BHSFM provides a general mathematical framework for describing behavioral heterogeneity and forms a more reasonable and elaborate evacuation process. Notably, some interesting evacuation phenomena can emerge by integrating the behavioral heterogeneity coefficient with temporal-spatial dynamic risk indexes. Compared with the social force model (SFM), higher frequencies of small-scale displacements are performed by BHSFM due to more pushing behaviors. Furthermore, the periods and areas of a potential crowd disaster are revealed by our model under different numbers of pedestrians, which has important guiding significance for formulating reasonable evacuation schemes in specific scenarios.