Deep Social Force
This work addresses incremental improvements in pedestrian simulation for applications like urban planning and autonomous systems.
The paper tackled the problem of unnatural locking behavior and lack of regional bias in classical Social Force models for pedestrian simulation by introducing a differentiable version with neural networks to relax interaction potential assumptions, resulting in potentials with sharp tips avoiding locking and asymmetric potentials enabling left or right bias.
The Social Force model introduced by Helbing and Molnar in 1995 is a cornerstone of pedestrian simulation. This paper introduces a differentiable simulation of the Social Force model where the assumptions on the shapes of interaction potentials are relaxed with the use of universal function approximators in the form of neural networks. Classical force-based pedestrian simulations suffer from unnatural locking behavior on head-on collision paths. In addition, they cannot model the bias of pedestrians to avoid each other on the right or left depending on the geographic region. My experiments with more general interaction potentials show that potentials with a sharp tip in the front avoid locking. In addition, asymmetric interaction potentials lead to a left or right bias when pedestrians avoid each other.