A Study On The Key Techniques Of Emotional Virtual Human With Learning Ability

Virtual human is the expression of the geometric and behavioral characters of human in the space generated by computer (virtual environment)]. The ultimate reason for developing realistic-looking virtual humans is to be able to use them in any virtual scene that simulates the real world . Researches that are relative to virtual human have begun since 70s . The requirements for the simulation technology in human's appearance and behavior are increasing rapidly in some new application fields, such as: three-dimensional game, the computer cartoon, film production, etc. These demands play a powerful role to the development of virtual human technology. Furthermore, autonomous virtual human is a very important component of virtual environment.There searches in virtual human technology involve computer graphics, kinematics and dynamics, multi-functional perception, artificial intelligence, virtual reality, etc. Based on the new method of combination of virtual human and intelligent agent, this dissertation presents a novel prototype of virtual human which has desire, emotion and learning ability. The results of either new technologies or the improvements of previous works in the field of virtual human are behavior modeling, affection modeling and motion learning modeling. The works and evolvements that author has done is introduced in this dissertation, including:To realize the believable existence of human being in the virtual environment, we should not only represent virtual human'body through geometry modeling, but also simulate their behavior through behavior modeling. As the fast development of the techniques of virtual reality, it becomes a new research focus to simulate virtual human's behavior and assess its performance. This dissertation models autonomous virtual human's behavior model as fuzzy logical system. Fuzzy logic enhances our capabilities of capturing the intrinsic uncertainty of human common-sense reasoning and decision-making during their communications. Autonomous virtual humans play an important role in dynamic virtual environment.Recent research in psychology and neuroscience has identified both the critical role of emotion in decision-making and social interaction. Due to this, to simulate virtual human's believable social communication, a personality-emotion model should be constructed based on artificial psychology first. In this paper, Dynamic Bayesian Network is used to simulate the human's affection model. Nonverbal communication is chosen to present our virtual human's emotion change. Nonverbal communication is crucial to increase the believability of virtual humans. During which, they respond not only to the presence of other virtual humans but also to their postures.