Interactive large-scale crowd simulation

Abstract

We survey some recent work on interactive modeling, simulation, and control of large-scale crowds. Our primary focus is on interactive algorithms that can handle a large number of autonomous agents. This includes techniques for automatically computing collision-free trajectories for each agent as well as generating emergent crowd behaviors including lane formation, edge effects, vortices, congestion avoidance, swirling and modeling varying crowd density. Some of these methods map well to current multi-core and many-core processors and we highlight their performance in different urban scenarios. © 2012 Springer-Verlag Berlin Heidelberg.