Cybersecurity challenges in AI-enabled smart transportation systems

Abstract

As an essential component of sustainable cities, smart transport plays a pivotal role in moving people and commodities efficiently and enhancing the quality of services for the entire community. The rapid advancements of the Internet of Things (IoT), machine learning (ML), and artificial intelligence (AI) have catalyzed the development of smart transportation systems. Various applications, such as personalized route guidance and traffic control systems, have been extensively studied and widely deployed over the globe. By heavily relying on real-time, multi-source, and accurate information, ML and AI-based system solutions are smart and efficient. However, ML and AI can be a double-bladed sword, as many recent studies revealed the vulnerability issues of ML and AI models under falsified information or adversarial attacks. This presents cybersecurity challenges in smart transportation systems. Available research suggests that few studies have investigated this issue. In this chapter, cybersecurity challenges are discussed in the context of different smart mobility applications such as traffic prediction systems and intelligent traffic signals. The information and analysis in this chapter will assist stakeholders to improve the reliability and robustness of ML and AI-based applications and better protect smart transportation systems.