A Systematic Review on IoT and AI Powered Smart Parking Technologies for Urban Environments

Abstract

The number of cars in cities has expanded dramatically, making it more difficult to manage parking effectively and contributing to traffic jams, fuel waste, and environmental damage. Up until now, traditional parking systems—which are either sensor-based or manually operated—have had trouble adapting to changing urban mobility and dynamic traffic circumstances. Unprecedented developments in computer vision and the Internet of Things (loT) have created opportunities for intelligent, reasonably priced, and highly scalable parking solutions that can instantly monitor, assess, and manage spots. This paper's goal is to review the literature on state-of-the-art research in smart parking systems using computer vision and loT technology, highlighting technological developments that have improved functionality and performance as well as implementation issues and future development opportunities. In particular, this research employs a systematic review approach to investigate how the development of parking management techniques for smart cities is being driven by current models incorporated with lightweight YOLO architecture on edge computing and adaptive control algorithm technology. Limitations resulting from system scalability, data security, and environmental conditions are also examined, along with recommendations for potential future enhancements using sensor fusion methods and Al-driven analytics.