Meshless physical deformation simulation method based on moving least squares

Abstract

The invention provides a meshless physical deformation simulation method based on moving least squares. The method comprises the following steps: firstly, the algorithm carries out particle sampling for a simulation material geometric model according to a given sampling solution, and initializes the attribute of the sampled particle material; then, the algorithm enters the physical simulation cycle, and resolves an accelerated physical control equation by decomposing the particle stress calculation into a constant quantity and a variable quantity based on jacobian matrix; finally, high-efficiency simulation is executed by decomposing the surface normal vector calculation of an object into a variable quantity and a constant quantity based on a meshless shape function gradient when driving the object to produce surface mesh deformation. The meshless physical deformation simulation method provided by the invention is more stable in simulation result, can obtain higher simulation accuracy, and is comparable to the SPH (smooth particles hydrodynamics) in calculation efficiency for scenes containing simulation particles of similar scale.

Description

technical field [0001] The invention relates to the field of virtual reality, in particular to a gridless physical deformation simulation method based on least moving squares. Background technique [0002] In the field of virtual reality, the accuracy and real-time performance of physical simulation of virtual scenes are getting higher and higher. Traditional non-physical methods, such as free-form models and semi-physical method spring particle models, are difficult to meet certain requirements due to low simulation accuracy. For applications with high requirements, mesh algorithms such as finite element have high simulation accuracy, but the calculation overhead is high, especially when the topology changes, it is difficult to update the physical model accurately in real time. The meshless physical deformation method that has appeared in recent years is very suitable for deformation simulation of objects with variable topology and can guarantee high simulation accuracy. T...

AI Technical Summary

Problems solved by technology

Martin et al. introduced the concept of elastic elements (Elastons) to solve the stability problem of MLS, but this method can only meet the needs of non-real-time physical simulation

Images