Improved Group Search Algorithm And Its Application Research In The Group Animation

The computer graphics is a study based on the laws of physics, empirical methods, and cognitive principle, processing the two-dimensional or three-dimensional graphics data to generate a visual data performance scientific. It is a branch of the field of computer science and application direction, focused on digital synthesis and operation of visual graphics content. Computer graphics after nearly40years of development, has entered a more mature development phase, and widely been applicated in computer-aided design and processing, film and television animation, military simulation, medical image processing, meteorology, geology, finance and electromagnetic science visual field. Computer graphics has been successfully applied in these areas, especially in the rapid development of the animation industry. Currently, most animation traditional keyframe animation division has created a lot of good works, but with the expansion of applications, the expansion of the scale and the improvement of people’s needs, the lack of keyframe-based technology has become increasingly conspicuous, traditional animation techniques in every movement of the animated characters and mobile detail by animation division control, the role of the increase in the number of longer time with the animation, scene complexity rises, animator, a significant increase in the amount of labor, at the same time, exist in the movement of groups of individuals influence each other, like the mass movement has individual characteristics and group characteristics, simulation groups animation performance and realism improve, the animation division is very challenging.The group search optimization algorithm is a new swarm intelligence optimization algorithm, the simulation of the algorithm derived from the foraging behavior of social animals, such as birds, lions and fish. The algorithm is based on the the producer-scrounger model, and introduced a rogue strategy to avoid faster into local extremum. At the same time, the algorithm uses the the animal visual search mechanism to expand the scope of the search. The algorithm is simple, and has good global search capability in dealing with high-dimensional problems. But it has a common problem of most of the optimization algorithm:easily falls into local minima, impacts the convergence of the algorithm, reduces the optimization of the performance of the algorithm.In this paper,according to the limitations of the traditional groups animation, algorithm is improved search algorithm to group, it showed better global search capability in handling high-dimensional function in dealing with low-dimensional problems can be demonstratedsuperior performance, and appled to the group animation to enhance the authenticity of the application of the algorithm animation. There are four aspects in this paper:1. An improved search optimization algorithm is proposed.By improving the convergence strategy and group intelligence on the group search optimization algorithm, differential evolution algorithm is introduced groups optimal stalled, and in accordance with the algorithm to its own characteristics, differential planning algorithm variation it to get rid of local polar value point of bondage, simulated annealing mechanism to improve the global search capability. 2.The improved group search optimization algorithm is applied in crowd animations.Build a simulation system on VS2003ACIS platform and under the Windows XP operating system,and crowd simulation groups animation phenomenon.By avoiding collisions between the groups and between groups with an obstacle, the groups showed better intelligence. At the same time, the simulation experiment applied to the Maya3D animation, and the algorithm has better animation application.3. A new group path planning method based on improved group search algorithm by step search is presented.According with the the Group Animation traditional path planning algorithm searches for a long time and low efficiency, optimizing capacity and poor, presented the group path planning method based on improved group search algorithm. Firstly, to improve for the group search optimization algorithm limitations, use the simulated annealing algorithm to abandon visual search mode step search, the group search algorithm is efficient and simple and easy to implement. The same time, in order to avoid the long path, complex environment, it is simple to use algorithm planning starting point to the path between the target point resulting in computationally intensive, time-consuming and long, hierarchical path planning to introduce multi-threaded and path random splicing technology. The outer layer of the use of grid area multithreaded local path planning based on the A*algorithm for global path planning method,and the inner layer of the improved group search optimization algorithm, to achieve the parallelism between the grid area. According to the traditional path splicing technology of "the elongated line","gathering"and"jump"phenomenon, this article uses a random path splicing technology, effectively stitching path within the grid area, and then planning the whole path.4.Groups path planning method based on improved group search algorithm is applied to chemical plants escape.The evacuation is the behavior of people in the face of danger, quickly fled the scene. Application of chemical plants escape path planning method based on improved group search algorithm groups not only be able to the crowd escape the more realistic simulation of chemical plants dangerous path, and has a good application in computing speed and analog effects.