GPU-Accelerated Subsystem-Based ADMM for Large-Scale Interactive Simulation

Harim Ji, Hyunsu Kim, Jeongmin Lee, Somang Lee, Seoki An, Jinuk Heo, Youngseon Lee, Yongseok Lee, Dongjun Lee

Keywords: Simulation, Hand object manipulation, GPU

TL;DR: This paper introduces a new rigid body simulation framework using ADMM based optimization accelerated by GPU, showing promising results in use cases for large scale interactive scenarios.

Abstract:

Abstract—In this paper, we implement the GPU-accelerated
subsystem-based Alternating Direction Method of Multipliers
(SubADMM) for interactive simulation. The challenging ob
jective for interactive simulations is to deliver realistic results
under tight performance, even for large-scale scenarios. We
aim to achieve this by exploiting the parallelizable nature of
SubADMM to the fullest extent. We introduce a new subsystem
division strategy to make SubADMM ‘GPU friendly’ along
with custom kernel designs and optimization regarding efficient
memory access patterns. We successfully implement the GPU
accelerated SubADMM and show the accuracy and speed of
the framework for large-scale scenarios, highlighted with an
interactive ‘Hand demo’ scenario. We also show improved
robustness and accuracy compared to other state-of-the-art
interactive simulators with several challenging scenarios that
introduce large-scale ill-conditioned dynamics problems.

Submission Number: 4