The growing global population demands sophisticated transportation infrastructure to combat their travel needs. While the majority of people are expected to live in cities, traffic congestion is inevitable. This also results in unavoidable environmental impacts, causing restructuring of existing transportation systems. Emerging transportation technologies can address these challenges. The intelligent transportation system (ITS), a smart transport system, includes a wide range of innovations encompassing communication networks, automation, data analytics, and real-time data collection, which work all together towards transport system optimisation, from managing traffic flow to vehicle operations. This interconnected system can overcome the problems of traffic congestion, safety concerns, and environmental impacts. Federated learning (FL) signifies a decentralised machine learning approach, which offers a promising solution for advancing smart transportation by training machine learning models on various decentralised devices while maintaining data privacy. FL applications are increasingly deployed in multiple Internet of Things (IoT) domains, including the Internet of Vehicular Things (IoVT) and smart cities. This chapter aims to comprehensively review FL for IoVT networks in ITS aimed at sustainable routing. In addition, this survey also explores how FL in ITS contributes to sustainable development goals, thus developing a sustainable transportation system. The research employs a systematic literature review mechanism called the PRISMA framework (Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA)) to assess the studies included. A narrative approach is used to integrate the findings of the incorporated studies. The study outcomes portray recent trends, challenges, and pathways for further research. The practical implications of the study include promoting sustainable smart transport systems by adopting greener practices and encouraging inclusive industrialisation towards realising Sustainable Development Goals (SDGs).

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Systematic Survey on Federated Learning for Sustainable Routing in Intelligent Transportation Systems: Insights into SDGs

  • A. Katheeja Naseeha,
  • Manu Shukla,
  • Purvi Pujari,
  • Anuj Kumar

摘要

The growing global population demands sophisticated transportation infrastructure to combat their travel needs. While the majority of people are expected to live in cities, traffic congestion is inevitable. This also results in unavoidable environmental impacts, causing restructuring of existing transportation systems. Emerging transportation technologies can address these challenges. The intelligent transportation system (ITS), a smart transport system, includes a wide range of innovations encompassing communication networks, automation, data analytics, and real-time data collection, which work all together towards transport system optimisation, from managing traffic flow to vehicle operations. This interconnected system can overcome the problems of traffic congestion, safety concerns, and environmental impacts. Federated learning (FL) signifies a decentralised machine learning approach, which offers a promising solution for advancing smart transportation by training machine learning models on various decentralised devices while maintaining data privacy. FL applications are increasingly deployed in multiple Internet of Things (IoT) domains, including the Internet of Vehicular Things (IoVT) and smart cities. This chapter aims to comprehensively review FL for IoVT networks in ITS aimed at sustainable routing. In addition, this survey also explores how FL in ITS contributes to sustainable development goals, thus developing a sustainable transportation system. The research employs a systematic literature review mechanism called the PRISMA framework (Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA)) to assess the studies included. A narrative approach is used to integrate the findings of the incorporated studies. The study outcomes portray recent trends, challenges, and pathways for further research. The practical implications of the study include promoting sustainable smart transport systems by adopting greener practices and encouraging inclusive industrialisation towards realising Sustainable Development Goals (SDGs).