Perception Model Of Autonomous Virtual Characters In Intelligent Virtual Environments

Autonomous behavior is supported by perceptual data of the environment, so it is very important to build perception model for the autonomous virtual character. The character can operate without the direct intervention of the external controllers or other virtual characters, and it also implement simple perceptual function in order to control its own behavior and internal state. The perceptual function include the perceptual information of the scence and the the virtual character. The character can respond according to the information of the environment, as well as changes its behavior.We studied perception modeling technology and its relevant technologies, especially the visual perception. First it introduces some correlative conceptions and applications of the intelligent virtual environments and autonomous virtual character. Then did some research on relevant technology for constructing autonomous virtual character model in intelligent virtual environment and discussed relevant methods. Secondly, we builded a simple 3D virtual environment and a virtual human model, and we selected the appropriate action animation for the virtual character based on the motion capture data. At last, we builded the perception modle for the autonomous virtual character. We take the method of false colouring to code for visual images and identify objects according to colour from the perspect of the virtual character; when building the vision sensor, we take forward a kind of collision detection method based on the z_buffer depth value; and we builded three kinds of visual memory storage modules: ultra short-term memory module, short-term memory module and long-term memory module.In this paper, a perception model is presented. And by building the perception modeling of autonomous virtual characters in intelligent virtual environments, we make the character become an autonomous character who can decide what to do and how to do by itself. Besides that we represent the environment using perceptual data structures which efficiently support the autonomous character.