Overview of Object Detection and Recognition in Vehicle Autonomous Driving

Abstract

In recent years, autonomous vehicles have garnered significant attention. As a core technology, object detection and recognition enable vehicles to perceive surrounding obstacles and traffic information, facilitating safer and more convenient operations. This paper offers a comprehensive overview of object detection and recognition methods in autonomous driving, encompassing key sensors (cameras, LiDAR, radar), traditional techniques, fusion approaches, and uncertainty estimation. It analyzes the strengths and limitations of methods like VoxelNet and CenterNet, and discusses future challenges from complex, diverse environments. Finally, it explores current deficiencies, difficulties, and potential research directions. This review aims to inform researchers and practitioners in autonomous vehicle perception systems.