Understanding complex crowd dynamics with generative neural simulators
This work addresses the problem of designing efficient urban infrastructure and ensuring safe crowd management, offering a novel paradigm for data-driven scientific discovery in crowd dynamics.
The authors tackled the challenge of understanding pedestrian crowd dynamics by developing a data-driven Neural Crowd Simulator (NeCS) that combines laboratory-like controllability with statistical resolution from large-scale data, enabling the reproduction of known experimental results and uncovering new insights into vision-guided and topological N-body interactions.
Understanding the dynamics of pedestrian crowds is an outstanding challenge crucial for designing efficient urban infrastructure and ensuring safe crowd management. To this end, both small-scale laboratory and large-scale real-world measurements have been used. However, these approaches respectively lack statistical resolution and parametric controllability, both essential to discovering physical relationships underlying the complex stochastic dynamics of crowds. Here, we establish an investigation paradigm that offers laboratory-like controllability, while ensuring the statistical resolution of large-scale real-world datasets. Using our data-driven Neural Crowd Simulator (NeCS), which we train on large-scale data and validate against key statistical features of crowd dynamics, we show that we can perform effective surrogate crowd dynamics experiments without training on specific scenarios. We not only reproduce known experimental results on pairwise avoidance, but also uncover the vision-guided and topological nature of N-body interactions. These findings show how virtual experiments based on neural simulation enable data-driven scientific discovery.