Research Of Dynamic Cloth Simulation Based On Position Constraint Method

Cloth simulation is an important part of computer animation. It is widely used in film, television, games, and the textile industry. Cloth simulation is committed to simulate the movement of real cloth deformation effects, and the research covers computer science, physics, mathematics, and art design disciplines. The cloth has delicate geometric modeling and complex physical characteristics, which causes the cloth simulation become as the most challenging research topic in the area of Computer Graphics.There have been numerous studies of dynamic Cloth Simulation, however, the stability of building stable cloth system in the explicit integration framework. In this paper, a cloth simulation algorithm based on position constraints has been proposed with the theory of kinetic constraints. The algorithm does not involve calculating the internal forces during the entire calculation process; it achieves updating particle positions with directly moving the particle. The main work in this paper as follow:Firstly, this paper presents a comprehensive detailed introduction of fabric physical modeling, numerical integration methods and collision detection technology.Then, this paper describes the cloth simulation algorithm based on position constraint. The main innovation of the algorithm is to improve the traditional particle-spring model, remove the spring internal forces in the model:(1) Remove the effect of the structure spring, define the location of the distance between two particles adjacent as structural constraints;(2) Cancel the force of the bending spring, and define the dihedral between the adjacent triangular faces as bending constraint;(3) Define the fabric collision as the location constraint which between the particle and the collision surface, and realize collision response with projection constraint. In every calculation, the particle position witch driven by external force in the next moment with explicit calculation, and solve the constraint equation with the predicted position, then iterative updating the particle position which meets the constraints. And update the particle velocity with Vrelet integration at last. Finally, this paper presents the algorithm simulation results and analysis. It is improved by experiment that the sports deformation and slitting of cloth could both be stably simulated with this algorithm.