Research On Modeling Method Of Autonomous Model

Virtual environment is primarily used to simulate real-world scenarios, therefore, the properties of the model in the virtual scene is determined by the corresponding object in the real world. However, the modeling process focuses on geometrical features and surface features in the current virtual reality and simulation systems, of which there aren't any physical properties, perception, independent reasoning and other aspects of information, so the models in the scene are static and passive. In order to enhance the reality and immersion of the virtual environment, a modeling method of autonomous model is presented against the deficiencies of existing modeling system.Modeling method and related technologies of autonomous model in the virtual environment are studied in this paper from several aspects, such as geometric modeling, physical property, perception mechanism, communication component, reasoning mechanism, and behavioral modeling.1. Geometric modeling. Considering the accuracy and display speed of the model, using polygonal representation method builds the geometric model. A variety of model formats are supported to load, such as 3ds, flt, ive, osg, obj, etc.2. Physical property. In order to achieve the reality, physical properties are added in autonomous model, such as mass and inertia. Its implementation depends on ODE physics engine which is integrated by Delta3D and physical property editing in BitMod system.3. Perception mechanism. In order to obtain the relevant information of other models and environment, moreover, to detect the changes of external environment, autonomous model use blackboard system supported by information layer and information space to achieve the perception mechanism in virtual environment.4. Communication component. Autonomous model has the ability to improve the communication efficiency between collaboration models and strengthen effective links between members. Point to point and publish/subscribe mode are supported to achieve natural and reasonable information exchange in the system.5. Reasoning mechanism. In order to ensure certain degree of autonomy and intelligence, autonomous model in the virtual environment must have the ability to make real-time response to a variety of events. Using BDI dynamic inference mechanism to achieve the selection process of optimal behavior is important reflection of autonomous model's adaptability and controlling.6. Behavior modeling. Using MD2 animation model, Python scripting language, equations, finite state machines and other technologies to achieve the simulation and editing of simple behavior, computable behavior and intelligence behavior of autonomous model.This paper present an extended model format, define a valid framework of model perception, communication and reasoning, establish a reasonable behavior selection mechanism and achieve the model properties'editing and model files'accessing in BitMod system. The modeling result of autonomous model is shown to offer technical support to runtime platform. Compare to other modeling methods, autonomous model is more likely to describe their own nature of the object.