ODEs learn to walk: ODE-Net based data-driven modeling for crowd dynamics
This work addresses crowd dynamics prediction for urban planning and safety applications, but appears incremental as it combines existing ODE-Net and social-force methods.
The authors tackled the problem of predicting pedestrian crowd behaviors by developing a data-driven modeling approach based on the ODE-Net framework, incorporating social-force concepts to address dimensionality challenges, and demonstrated its performance through application examples.
Predicting the behaviors of pedestrian crowds is of critical importance for a variety of real-world problems. Data driven modeling, which aims to learn the mathematical models from observed data, is a promising tool to construct models that can make accurate predictions of such systems. In this work, we present a data-driven modeling approach based on the ODE-Net framework, for constructing continuous-time models of crowd dynamics. We discuss some challenging issues in applying the ODE-Net method to such problems, which are primarily associated with the dimensionality of the underlying crowd system, and we propose to address these issues by incorporating the social-force concept in the ODE-Net framework. Finally application examples are provided to demonstrate the performance of the proposed method.