Collaborative Research: NeTS: JUNO3: Towards an Internet of Federated Digital Twins (IoFDT) for Society 5.0: Fundamentals and Experimentation

点击翻译按钮获取中文摘要

The concept of a digital twin (DT) that enables the creation of a programmable, digital representation of physical systems (e.g., smart vehicles) will revolutionize future industries. DTs lie at the heart of the vision of a future smart society, dubbed Society 5.0 by Japan, in which a high integration between cyber and physical spaces is exploited to bring forth economic and societal advancement across industries ranging from intelligent transportation to robotics. To realize this vision of a new DT-driven Society 5.0, this project envisions a novel concept of an Internet of Federated Digital Twins (IoFDT) that holistically integrates heterogeneous and physically separated DTs representing different Society 5.0 services within a single system. The goal of the research is therefore to create the scientific foundations of the IoFDT through a close collaboration between a team of US and Japanese researchers with complementary expertise in wireless networking, DTs, and artificial intelligence (AI). These foundations include a new, fundamental framework for allowing effective design, analysis, and optimization of an IoFDT system, in presence of heterogeneous applications. The proposed research will have a tangible societal impact since it contributes towards enabling diverse Society 5.0 services (from transportation to factory automation) thus potentially improving the quality of life. The research is coupled with collaborative US-Japan DT-centric education, outreach, and dissemination plans that will help broaden the impact of the research. To design the proposed framework, this joint US-Japan project will merge ideas from wireless communications, machine learning, and programmable networking to contribute several innovations: 1) A cross-layer wireless and computing framework that introduces new methods for dynamic network slicing for the IoFDT system, 2) Efficient methods for DT-centric proactive resource optimization as well as collaborative computing and learning techniques that can deal with unprecedented dynamics of the cyber (wireless) and physical systems of an IoFDT, 3) A holistic AI framework, based on continual learning, that can faithfully build programmable DTs from a continuous stream of data while exploiting complex relationships and patterns across the physical systems and their coordinated twins, and 4) Realistic implementation in a programmable IoFDT platform that provides a meaningful proof-of-concept of an IoFDT system deployment in Japan.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.