Simulation of Social Behaviors in Virtual Crowd

This paper presents a novel communication model to simulate various crowd behaviors such as riot. Our communication model is heavily based on the results from sociology research. Collective behaviors can emerge out of social processes such as emotion contagion and conformity effect among individual agents. The communication model has been implemented in our crowd simulation system, IMCrowd, in which each agent has a local perception and autonomous abilities to improvise their actions. Simulations on riot formation and riot control are demonstrated as an application example of IMCrowd.