Research On Virtual Environmental Modeling And Path Planning Of Group Animation

In recent years, the rapid development of computer graphics and digital processingtechnology pushes and popularizes the technology and application of computer animation.Group animation is always the main studying field of computer animation, which has greatlyattracted the attention of scholars with its large quantity, powerful movement effects andsimulation results. At present, group animation has played an important role in games,entertainment, education, military, virtual reality and so on.There are two crucial steps for the creation of group animation, environment modelingand path planning, which will determine whether the group animation is successful or not.Theoriginal method for creating group animation is based on software and keyframe, which hassome disadvantage, such as the truth of animation cannot be guaranteed, lack of autonomy andinnovation, time consuming, a large labor to provide, high cost, high demands for theequipments of hardware, these disadvantages cause the original method can’t meet people’sdemands for group animation. Also, the continually development of information technologyand the constant progress of tertiary industry increase the difficulty of creating groupanimation. In that situation, how to design virtual environment, path data, role model and creatgroup animation, with great efficiency, high quality and also at low cost, under the prerequisiteof ensure the truth, independent and intelligence, are urgent problems to be settled in the fieldof group animation production.For the above-mentioned problems, starting with virtual environmental modeling andpath planning, aim for designing methods of environmental modeling, improving independentand interactive of path planning, modeling the roles rapidly and increasing efficiency ofcreationg group animation, this paper presents a thorough study of virtual environmentalmodeling, swarm intelligence, evolutionary computation and applies them to create groupanimation. The main work and innovation of this paper are summarized as follows:1. Exploring the environmental modeling methods and in the case of the existingproblems of virtual environmental modeling of group animation, this paper divides theprocessing of virtual environmental modeling into two parts: environmental modeling and rolemodeling. The multi-level virtual environmental modeling method based on octree (HEM-BO)and a novel role modeling method (NICGA) based on improved Cellular Genetic Algorithmand NURBS intelligence are proposed. They are used to model the virtual environment androles. These methods can raise the efficiency of environmental modeling and detecting,resolve the interaction between environment and roles, and generate group animation quickly. 2. Based on HEM-BO mentioned above, a novel swarm path planning method based onimproved Artificial Bee Colony algorithm (HABC) is proposed. Topology path planning usingA star in outer space and swarm path planning using HABC in inner space are combined inthis method, which can resolve the problems of algorithm convergence and intelligence, andimprove efficiency of path planning.3. Based on the methods of environmental modeling and path planning, multi-modulegroup animation creative method based on data-driven (MDGA) is proposed, which dividesthe process of creating group animation into a four-part cent: model of environmentalmodeling (MEM), model of path planning (MPL), model of role modeling (MPL) and modelof data-driven animation generating (MDD), and imports data of environmental model, rolemodel and path data to animation software, and uses these data to drive and generate keyframeanimations and path animations. The four modules are division as well as cooperation, whichcan effectively improve creating efficiency of group animation.4. Relating scientific projects, applying the above research achievements, I have beeninvolved with projects on related subjects for research and development, such as“Collaborative Animation Innovation Design Platform Based on Technologies of BionicAlgorithms (item number: 2010GGX10106)”etc. I am mainly responsible for virtualenvironmental modeling, path planning, data reading and importing, data-driven animationcreating and so on. A 3D group animation creating system-GA Creator, with ACIS/.NET andMAYA software as platforms, is designed and developed by working with other members ofproject group. This system supports environmental modeling, role modeling, path planningand animation generating, uses the data of prototype system to assist in generating animation,and is also used to be a platform to prove the practicability of the above theories. Experimentalresults in the prototype system show that this method can be used to generate group animationquickly and improve the truth and intelligence of group animation.