Framework Design of a Digital Twin of an XY Compliant Parallel Manipulator Based on Non-Negative Matrix Factorization

Xueguan Song,Kunpeng Li,Shuo Wang,Ziyun Kan,Haiyang Li,Jiaxiang Zhu,Guangbo Hao

DOI: https://doi.org/10.1115/detc2022-89187

2022-01-01

Abstract:A promising multi- layer mirror-symmetry XY compliant parallel manipulator (CPM) has been recently reported to address the tradeoff between a small compact footprint and a minimized parasitic rotation. In order to scientifically ensure the healthy operation of equipment and make maintenance decisions reasonably, there is a need to depict its physical mechanical characteristics in a virtual space instantaneously. The digital twin, an emerging technology, can be used to address this need by achieving a seamless convergence of physical and virtual spaces for this XY CPM. However, the high accuracy and instantaneousness requirements have hindered the application and popularization of the digital twin. This article presents a framework to build an accurate and lightweight digital twin, and in the meanwhile significantly reduces the computational budget (i.e. high computation efficiency). The framework is validated by an XY CPM test apparatus. The results demonstrate that the proposed framework is an effective tool to build an accurate and lightweight digital twin for the XY CPM, which is also promising for other compliant mechanisms or parallel manipulators.