Cluster based partitioning for agent-based crowd simulations

Abstract

Simulating crowds is a challenging but important problem. There are various methodologies in the literature ranging from macroscopic numerical flow simulations to detailed, microscopic agent simulations. One key issue for all crowd simulations is scalability. Some methods address this issue through abstraction, describing global properties of homogeneous crowds. However, ideally a modeler should be able to simulate large heterogeneous crowds at fine levels of detail. We are attempting to achieve scalability through the application of distributed simulation techniques to agent-based crowd simulation. Distributed simulation, however, introduces its own challenges, in particular how to efficiently partition the load between a number of machines. In this paper we introduce a method of partitioning agents onto machines using an adapted k-means clustering algorithm. We present, validate and use an analysis tool to compare the proposed clustered partitioning approach with a series of existing methods.

A Task Partition Algorithm Based on Grid and Graph Partition for Distributed Crowd Simulation

“…In the experiment, we compare our algorithm with proposed RBP/LP/CRP algorithm by Lui and Chan [8] and cluster-based algorithm proposed by Yongwei Wang [6] in partition cost and execution performance. The computation of partition cost is according to the cost formula mentioned by Lui and Chan [8].…”

“…Yongwei Wang [6] proposed a k-means clustering crowd partition algorithm. In the paper, the individual is supposed to be an agent and each agent has its own area of interest (AOI).…”

“…When there is load imbalance after one clustering step, it needs one more clustering step to get better effects of load balance. Guillermo Vigueras [7] raises an enhanced load balance algorithm based on Yongwei Wang's work [6]. These methods are based on clustering algorithm, so they both have the same weakness.…”

Towards a conceptual framework to support adaptative agent-based systems partitioning

“…Also many algorithms have been either adopted or proposed to properly partition MAS in many application domains. For the crowd simulations, many algorithms were evaluated like the k-mean clustering approach in [5], the grid-based partitioning method in [6], as well as the convex Hull algorithm in [4]. Also, in [7][8], the authors presents respectively a cluster-based approach and a proximity load balancing method to partition individual-oriented fish school simulations.…”

Uma Abordagem Baseada Em Gerenciamento De Interesses Para O Particionamento Dinâmico De Simulações Distribuídas

“…Entre alguns dos principais trabalhos relacionados destacam-se: [Wang et al 2009], [Wang, Lees e Cai 2012], [Logan e Theodoropoulos 2001], ] e [Oguara et al 2005].…”