A hierarchical framework for crowd behavior simulation with constraints on different levels

Shilei Li,Fei Peng,Yuewen Fu,Qing Ye,Jiahong Liang,Meng Li,Shi Qu
DOI: https://doi.org/10.1109/ITCOSP.2017.8303090
2017-01-01
Abstract:This paper presents a hierarchical framework to simulate crowd behaviors, which is constructed by dividing the complex task into sub-tasks under the different levels. By specifying different information-domain constraints or requirements into different levels, the diverse behavior characteristics of individual and the cohesive characteristics of the whole group can be generated simultaneously. In addition, the constraints are integrated into the probabilistic roadmaps with different resolutions and thus it is well adapted to unstructured environment by achieving complementary advantages of high-level planning complexity and rapid responses to environmental changes. The paper also briefly describes the design of framework and presents a case study to show some results in a crowd simulation scenario. The experimental results show that our hierarchical behavior model not only well reflects the characteristics of the human behaviors, but also shows plausible crowd behaviors.
What problem does this paper attempt to address?