Soft-body simulation is currently widely used in animation, prostheses, organs, and so on. The most common way to implement the simulation is to use 3D software. However, the rendering speed of common commercial simulation software is limited. For example, the rendering time per frame of SolidWorks (on a DELL XPS 8600 desktop computer) for approximately 1 million points would reach 30 minutes. Therefore, one model based on the mass-spring mechanism is proposed by some researchers who use a mass-spring mechanism to simulate the behavior of soft bodies such as living tissues. The basic configuration of the mass-spring model is: all the mass points are allocated the corners of the cubes (morphable after loading) and are connected by springs. More specifically, every mass is connected to all of its 26 neighboring masses. The inherent problem of such simulations of large-scale mass-spring models with millions of particles and springs is still the rendering speed. In order to improve the rendering efficiency and to realize real-time rendering, a parallel computing architecture based on the CUDA architecture is introduced here. In addition, in order to increase the accuracy of the simulation by reducing the truncated error, the Verlet integration is employed to fabricate the dynamic model. At the same time, to construct an appropriate data structure for CUDA and the GPU hardware architecture, a parallel mass-spring algorithm that explicitly represents the position of the mass is usually employed. The positions of the mass elements are computed with the Verlet integration scheme. The main goal of this work was to check whether the massively parallel computing method based on the CUDA architecture improves the rendering performance of a mass-spring system model. In order to meet the minimum requirement for the refreshing rate to make the human eye comfortable (60 Hz for progressive scans and 30 Hz for interlaced scans), all the tests were locked with a 60 Hz refreshing rate. In addition, the simulation implemented here is a soft body composed of mass-particles and springs. This soft body has a uniform width and depth and a height that is small compared with the width and depth. It was modeled to fall down under the influence of gravity, and then, to impact on a rigid object which is fixed in midair. Subsequently, this soft object slipped around the object and finally disappeared into space. Although the performance difference between the serial method and the parallel method was not significant when rendering less than 2,500 points, it became very apparent when the number of points was increased to 10,000. Therefore, it can be proven that the efficiency of the simulation of a soft body modeled by the parallel mass-spring system can be improved significantly with parallel computing based on the CUDA architecture.


Soft-Body Simulation With CUDA Based on Mass-Spring Model and Verlet Integration Scheme

Division / Sponsors: 
Aerospace, Bioengineering, Applied Mechanics, Design Engineering Division, Advanced Energy Systems, Manufacturing Engineering Division, Noise Control and Acoustics, Nondestructive Evaluation, Diagnosis, and Prognosis (NDPD), Nuclear Engineering Division, Process Industries Division, Fluids Engineering, Heat Transfer, Materials, Microelectromechanical Systems, Safety Engineering and Risk Analysis, Computers & Information in Engineering


Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2020/volume/84478

IMECE2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.