A Review of Urban Path Planning Algorithms in Intelligent Transportation Systems

With the accelerating pace of urbanization and the increasing complexity of traffic systems, urban transportation faces growing challenges such as congestion, inefficiency, and environmental strain. Path planning algorithms—key components in intelligent transportation systems—have evolved from classical graph-based methods like Dijkstra and A* to modern approaches leveraging metaheuristics and deep learning. This paper systematically reviews the development of urban path planning algorithms, tracing their progression from foundational methods to state-of-the-art techniques such as Ant Colony Optimization, Probabilistic Roadmaps, and Rapidly Exploring Random Trees. Recent innovations, including improved genetic algorithms, hybrid A* variants, and reinforcement learning models, are analyzed in terms of adaptability, efficiency, and real-time performance. Furthermore, the review highlights ongoing challenges in scalability, dynamic adaptation, and algorithmic fairness, while discussing future directions that integrate technical innovation with policy and ethical considerations to support sustainable and equitable urban mobility.