Smart transportation systems are rapidly evolving, driven by advancements in technology and the growing demand for efficient, safe, and sustainable mobility solutions. Federated learning, a novel approach that enables decentralized machine learning across multiple devices without compromising data privacy, is emerging as a powerful tool to revolutionize smart transportation. This chapter explores the applications of federated learning in various facets of smart transportation, including traffic management, autonomous vehicle navigation, predictive maintenance, and public transport optimization. By leveraging data from diverse sources while maintaining data security and privacy, federated learning can enhance decision-making processes, improve operational efficiency, and reduce costs. However, implementing federated learning in smart transportation presents several challenges, such as data heterogeneity, scalability, communication costs, and regulatory concerns. This chapter discusses these challenges in detail and proposes potential solutions and strategies to overcome them. Furthermore, it outlines future directions for research and development, highlighting the integration of federated learning with other emerging technologies, such as the Internet of Things (IoT) and edge computing, to create more robust and adaptive smart transportation systems. Ultimately, this chapter aims to provide a comprehensive overview of how federated learning can transform the future of smart transportation, addressing current limitations and paving the way for innovative applications and improved urban mobility.