A Comprehensive Survey on AI-Driven IoT Traffic Control Systems with Collaborative Drones and Federated Intelligence

Abstract

 The rapid rise in urban traffic coupled with the inability of traditional traffic control systems to adapt and expand to meet the needs of modern smart cities has created congestion, increased emissions, and reduced overall system efficiency. Advances in Artificial Intelligence (AI), the Internet of Things (IoT), and Unmanned Aerial Vehicles (UAV) have provided the opportunity to develop intelligent, decentralized, and adaptive solutions to the Urban Traffic Management (UTM) issues that cities face today. This survey provides an in-depth look at the development of IoT-based traffic control systems using AI and collaborative UAV networks, using the Distributed Learning (FL) framework. We examined cutting-edge solutions that have incorporated edge intelligence with real-time aerial monitoring and privacy-preserving learning methods from 2021 to 2025 to optimize traffic management and increase user safety. The paper categorizes the existing research into three primary technology layers—IoT Sensor, UAV Coordination, and Federated Learning—and provides comparative data on each layer with respect to performance metrics (latency, throughput, model accuracy, and energy efficiency). In addition, the paper identifies some of the main research challenges that remain, including heterogeneous data, communication constraints associated with UAVs, and protective measures for information security within a federated model. The overall results indicate that collaborative UAV networks operating on Federated Learning-based framework architecture will provide a scalable, secure, and resilient Intelligent Transportation Systems (ITS) for the future