同一场景中异质布料的动态绘制

Dynamic Drawing of Heterogeneous Fabrics in the Same Scene

在运动仿真中异质布料属性不同,使得姿态变化不一样,统一设定迭代次数又会造成计算资源的不合理分配,因而,在PBD的基础上探索异质布料的动态绘制.首先,引入自由质点约束因子改进LRAs模型,解决自由质点间距离过长问题,并根据属性约束构建异质布料;然后,在XPBD解决同一布料中刚度依赖迭代次数的基础上,采用独立迭代的方式对约束方程进行求解,保证不同布料的迭代独立,并根据场景中不同约束所占比重不同设置不同的迭代次数,优化计算资源分配;之后,为加快求解过程,用Jacobi替代PBD中的Gauss-Seidel方法,以并行方式计算各约束;最后绘制不同平台下异质布料间的碰撞.实验结果表明,该方法能够快速绘制运动过程中的异质布料.

The attitude of heterogeneous fabrics diversification are not the same because of different attribute setting in the motion,and the set of same number of iterations would cause the waste of computing resources,we explored the dynamic drawing method of heterogeneous fabrics based on PBD. First,introduce the free particles constraint to solve the problem of too long distance between free particles in the LRAs model,and build heterogeneous fabric model according to attribute constraints. Then,In the case of XPBD solved the problem of stiffness dependent on the number of iterations,we use our independent iterative method to solve the constraint and set different iterations numbers of constraints according to the importance of constraints in different scenes to optimize the allocation of computing resources. In order to speed up the process,use the Jacobi method to replace the Gauss-Seidel method to solute constraints in parallel. Finally,the collision between heterogeneous fabrics under different platforms is drawn. The experimental results show that our method can quickly draw the heterogeneous fabric during the movement.