This article presents an extensive examination of the principles underlying data-driven vehicle autonomy, with a particular emphasis on the significant advancements in sensor technology and data fusion methods, all of which collectively contribute to the enhancement of safety in autonomous driving systems and optimising their navigational capabilities. It emphasises the essential issues pertaining to roadway safety by comparing human mistakes with the potential advantages of automated technologies, enabling reduced collision probabilities and enhanced traffic regulation. This paper delves into a variety of sensors, namely LiDAR, Radar, Cameras, Multispectral Imaging, Ultrasonics, Inertial sensors, and GNSS systems, examines the sensors’ contribution to the perception of the environment and discusses their operating principles, challenges, performance objectives and applications. This paper reviews the importance of sensor fusion techniques towards the enhancement of accuracy and reliability in vehicle localisation and dynamic state estimation. Also, it discusses emerging artificial intelligence and machine learning methods enabling multitask perception and decision-making for autonomous vehicles. Advances in automation tasks, such as path planning, vehicle control, and cooperative localisation, highlight the approaches for trajectory generation and the real-time adjustments using predictive control strategies, which are additionally reviewed. This review discusses the current state of the art in-vehicle autonomy technologies and outlines future directions toward safer and more efficient automated transportation systems.