Path planning methods for autonomous vehicles at intersections: A review

Autonomous vehicles (AVs) rapidly transform transportation, potentially enhancing safety, efficiency, and traffic flow. However, intersections remain a critical challenge for AVs due to the complex interactions with other vehicles, pedestrians, and dynamic traffic signals. Effective path planning at intersections is essential for AVs to navigate these environments safely and efficiently. Although numerous reviews have been published on path planning, limited attention has been devoted specifically to intersections. This review paper presents a comprehensive analysis of major path-planning methods used in Autonomous Vehicle (AV) navigation at intersections, including graph-based, sampling-based, curve-based, optimization-based, and machine learning–based approaches, while also examining emerging AI-driven path planners to better understand their capabilities. Each method is analysed in terms of its strengths, limitations, and applicability to real-world scenarios, focusing on the specific demands of intersection navigation. Furthermore, the review highlights key challenges such as handling dynamic multi-agent environments, managing interactions with human-driven vehicles, and balancing computational efficiency with path optimality and discusses potential solutions through adaptive, real-time algorithms, cooperative planning, and predictive modelling. Overall, this review aims to support the development of AV path planning, ultimately contributing to safer and more efficient autonomous systems in urban environments.