A unified structure for crowd simulation

Abstract

Realistic simulation of crowds is an important issue for the production of virtual worlds for games, entertainment or architectural and urban planning. Difficult issues need to be addressed such as collision avoidance and the handling of dynamic environments. In this paper, we present a unified structure for the simulation of crowds in complex urban environments. We propose a topological multiresolution model supporting different levels of details, allowing efficient proximity querying and compatible with real‐time rendering and hierarchical path planning. A fine exploitation of the multiscale aspect of the underlying model allows to achieve the same efficiency as the fastest existing methods. The generality of the approach allows the simulation to be executed on any two manifold, and the unified approach eases the handling of dynamic environments. Copyright © 2012 John Wiley & Sons, Ltd.

Proximity Queries for Crowd Simulation Using Truncated Voronoi Diagrams

“…A topological multi-resolution data structure to find neighbors, which encodes the information for every agent in a cellular structure was presented by Jund et al [Jund et al 2012]. Passos et al store entities inside a sorted matrix and use an extended Moore neighborhood to find neighbors [Passos et al 2009].…”

Deformable polygonal agents in crowd simulation

“…A multiresolution approach that adapts the size of the cells to the density of agents is a good trade‐off between these two choices. The framework presented in offers good performances and is flexible enough to support polygonal agents. As our works rely on this structure, we describe it further in section 4.…”

“…The work presented here enriches the approach of with the support of polygonal agents. Without this enrichment our framework would show similar performances than that has already been compared to the k d‐tree and spatial hashing approaches.…”

“…The work presented here enriches the approach of with the support of polygonal agents. Without this enrichment our framework would show similar performances than that has already been compared to the k d‐tree and spatial hashing approaches. Thus we present a formal study of the time complexity of the different approaches, followed by benchmarks to evaluate the cost of polygonal agents compared to simulations with point‐like agents only.…”

“…These latter agents can be complex non‐planar urban environments filled with fixed obstacles (such as buildings, bridges, roads or trees). This work is an extension of that introduced a unified structure for crowd simulation. This multiresolution data structure is used for the representation and rendering of the environment, for the tracking of point‐like agents and to optimize proximity queries.…”