Modeling autonomous virtual agent in virtual environment for risk behavior

Owing to the complexity of human behaviors in risk situation and the harsh physical environment of underground mine, current researches on coalmine virtual environment have paid little attention on modeling human risk behaviors. Based on Artificial Life method, an intelligent virtual miner agent model was proposed to yield better understandings of human factors in emergency situation and to assist risk behaviors analysis in underground coalmine. The virtual miner agent is a comprehensive model of 3D virtual human and combines perceptual, behavioral, internal states and cognitive modules. And then, behavioral model of virtual miner was introduced in detail. Behaviors of virtual miner are constructed by a set of parameterized basic actions and are controlled by personalized action selection mechanism, which incorporate miners´ cognitive processes and internal state characteristics to explore human behavioral mechanism of risk accidents. The preliminary results of application case show the effectiveness of our method to simulate human risk behaviors in coalmine virtual environment.