Human behaviors modeling in multi-agent virtual environment

PubDate: March 2017

Teams: Chongqing University of Posts and Telecommunications

Writers: Linqin Cai, Binbin Liu, Jimin Yu & Jianrong Zhang

PDF: Human behaviors modeling in multi-agent virtual environment

Abstract

Human behavior Modeling and simulation has become one of the most challenging research topics in many sectors where safety and organizational complexity are key issues. This paper involves constructing an integrative behavior framework for autonomous virtual miner agents that populate believably virtual coalmine environment to simulate human behaviors of underground coalmine. In this study, we present an emotional behavior model for virtual miner based on cognitive appraisal theories. The proposed models can generate more believable behaviors in virtual environment based on the personalized emotion states, internal motivation needs, and behavior selection thresholds of virtual miners. In addition, we take into account the impacts of underground mine workers’ personality straits on the intensity threshold of emotions, the emotion’s decay, and the motivation update process. Finally, we implement an interactive virtual environment for underground human behavior simulation. The behavior believability of virtual miner was evaluated with user assessment method. Experimental results show that the proposed models can create more realistic real-time virtual coalmine environments to simulate human behavior resulting in underground accidents.