Modeling and Path Generation Approaches for Crowd Simulation Based on Computational Intelligence

Modeling and simulating group behaviours have been an active research topic in the field of computer animation and game. This paper presents some novel approaches for supporting entity modeling and path generation in crowd simulation. It analyses related work about crowd simulation first. Then, an entity modeling approach based on CGA (Cellular genetic algorithm) and NURBS (Non uniform relational B splines) technologies is presented. Next, following the analysis to PSO (Particle swarm optimization) and ABC (Artificial bee colony) algorithms, a crowd path generative approach based on ABCPSO is put forward. After that, a simulating example of crowd cohesion and performance comparison are exhibited for showing the efficiency of the algorithms. Finally, the current work is summarized and an outlook for the future work is given.