CrowdMixer: Multiple Agent Types in Situation-Based Crowd Simulations

This paper presents a scalable approach to crowd simulation that can generate complex and varied simulations by using multiple types of individuals in a crowd. Efficiency is attained by using a situation-based approach where an individual agent adopts behaviors according to its situation, which corresponds to a subspace of the universe.

• Type: Conference
• Year: 2006
• Where: AIIDE
• Authors: Shannon Blyth, Howard J. Hamilton
• Keywords: Agent Adopts Behaviors, AIIDE 2006, Artificial Intelligence, Crowd Simulation, Scalable Approach