Digital Twins and Federated Learning for Smart Cities and Their Applications

Digital Twins and Federated Learning for Smart Cities and Their Applications

Surabhi Shanker
DOI: 10.4018/978-1-6684-3733-9.ch009
(Individual Chapters)
No Current Special Offers


This chapter is written with the intent to explore the history, architecture, applications, and challenges in the implementation of digital twin with IoT competences. Digital twins are considered to be a fundamental starting point for today's smart city construction. The chapter initiates with a brief description of the concepts of digital twins and digital twin for cities and smart homes, discusses the relationship between digital twins and smart cities, analyses the characteristics of smart cities and homes based on digital twins, and focuses on the main applications of smart cities based on digital twins. This chapter sheds light on the future development of smart cities and smart homes based on digital twins.
Chapter Preview


The concept of “twin” was introduced by NASA's Apollo program, which designed and manufactured two real identical space vehicles. One of them was launched into the air space to perform the mission, while the other remained on Earth, agreeing engineers to mirror the conditions of the launched one (Boschert & Rosen, 2016). In the white paper written by Grieves (Glaegen & Stargel, 2012), the Digital Twins was first projected at his executive course on product lifecycle management (PLM). With technical enhancements, the DT was introduced into the aerospace industry by National Aeronautics and Space Administration (NASA) and U.S. Air Force (Tuegel et al., 2011). Consequently, the space vehicle remaining on Earth could be replaced by a digital mirror model to provide more insights through high-fidelity simulation. Nowadays, the DT has been applied to more fields and has become a most demanding technology. In the figure:1 the complete development history of DT can be divided into three stages.

Figure 1.

The complete development history of DT can be divided into three stages


In the inception stage i.e. first stage, the concept of Digital Twin was introduced by Grieves in 2003 according to the Whitepaper written by him (Glaegen & Stargel, 2012). It was explained in three dimensions, including a physical entity, a digital counterpart, and a linking that connect the two parts together (Glaegen & Stargel, 2012).

In 2005, Grieves set forth another idea and told that the Digital Twin could be classified into three subtypes, including: DT prototype, DT instance, and DT aggregate (Tac, 2012). However, due to technical and perceptive limitations, we are having very few related reports in the following 5 years. But fortunately, during this period, the New IT has arisen and developed, which established the frameworks for the future development of the Digital Twins.

In 2010, NASA established the definition and function of Digital Twins for space vehicles in detail in the Draft Modelling, Simulation, Information Technology and Processing Roadmap (Allare, 2014). In 2011, the U.S. Air Force discovered the application of DT in the mechanical health management of aircrafts (Reifinider & Majumdar, 2013). In 2012, NASA and the U.S. Air Force jointly issued a paper about the Digital Twins, which expressed that the Digital Twin was the critical innovation for future vehicles (Tuegel et al., 2011). After that, the number of research studies on the DT in aerospace has improved gradually. For example, Tuegel proposed the Airframe Digital Twins for design and maintenance, and deliberated the challenges for its development (Swedberg, n.d.). Allaire et al. researched a dynamic data-driven application system, which was portrayed as the execution establishment of the DT for aerospace vehicles (Menard, n.d.). Reifsnider and Majumdar introduced a multidisciplinary physics-based approach for the DT in taskforce management (Science Service Dr. Hempel Digital Health Network, n.d.).

Key Terms in this Chapter

Smart City Operation Brain (SCOB): The SCOC is under the authority of the COO, and between them the Chief Information Officer (CIO) Joint Conference Committee is responsible for management and supervision. SCOC manages four main sectors for the urban information, including Urban IT Operation and Maintenance Centre, Big Data Centre, Urban Operation Monitoring and Command Centre, and Smart Service Centre.

DT Prototype: Digital Twin Prototype offers the user with the panorama to modify parameters and operating conditions of different components and allows observing the response of the entire system in real time. Digital twin prototype is a great way to ease communication between product developers and end users from planning and design phase to authorizing and forms the backbone and platform for the digital twin coupled to the physical entity through IoT and live sensor data.

Smart Supply Chain: Blockchains track the complete product information, avert the entry of bogus products into the market, and share information amongst various entities to optimize the decision-making process. In the process of doing this, IoT sensors gather and monitor the information of entities, such as logistics, specifications, affiliation, and value. The distributed network nodes monitor the composed data and produce a real-time package for communicating. The received package will be used to analyse the behaviours of vehicles. The process of sharing will be recorded on a “CSP.” Retailers, distributors, transporters, and suppliers can retrieve the “CSP” for access to the information associated with them. After successful authentication, entities will get the information they need.

Smart City Traffic Brain: The Smart City Traffic Brain is one more application of smart city created on digital twin. Trusting on technologies such as holographic perception, time-space analysis, and data mining, the Wuhan Road Traffic Smart Emergency System is established, which is a significant part of Wuhan Smart City Traffic Brain. The system is intended to deeply assimilate multi-network resources and real-time dynamic traffic information, while linking various emergency platform resources such as city alarm system, the police, road condition system, accident emergency system and traffic video system, and presenting them on the same interface.

Digital Twins: Digital twin is the concept of virtual replica of physical object. Digital twins duplicate the physical model for remote monitoring, viewing, and controlling based on the digital format. It is actually the alive model of the physical systems which uninterruptedly adapts to operational changes based on the real-time data from various loT sensors and devices and predicts the future of the matching physical counterparts with the help of machine learning/artificial intelligence.

Artificial Intelligence: Artificial intelligence (AI) is spreading like a fire in our world. With intelligent machines enabling high level cognitive processes like thinking, perceiving, learning, problem solving and decision making, with the facility of high-level data collection and aggregation, analytics and computer processing power, Artificial Intelligence is empowering and increment human intelligence and enrich the way people be in this world and work.

Federated Learning: Federated learning is a machine learning technique that allows machine learning models to obtain experience from different data sets placed in different sites (e.g., local data centers, a central server) deprived of sharing training data. This allows personal data to remain in local sites, reducing possibility of personal data breaches. Federated learning is used to train other machine learning algorithms by using multiple local datasets without exchanging data. This allows companies to create a shared global model without putting training data in a central location.

Complete Chapter List

Search this Book: