Exploring Sensor Solutions for Autonomous Vehicles:A Systematic Literature Review

Sammanfattning

Self-driving vehicles, also called Autonomous vehicles (AV), represent a rapidly growing technol- ogy with the potential to transform transportation. However, the public adoption of AVs is not only dependent on technical reliability, but also on how passengers experience safety during autonomous rides. This systematic research review examines 16 peer-reviewed studies to answer two research ques- tions. RQ1: How does sensing and perception in AVs influence perceived safety and public trust among passengers in real-world scenarios with fully autonomous vehicles? RQ2: Which sensor configura- tions and fusion strategies are most effective for improving both technical robustness and perceived safety? Studies were selected through systematic database searches and selected based on their rele- vance to real-world AV operation and passenger experience. The findings revealed that perceived safety is shaped mainly by predictable and smooth driving behavior, human supervision, HMI (Human Ma- chine Interface) design, and prior experience with AVs. Based on a synthesis of technical findings, the most effective sensor configuration combines LiDAR (Light Detection and Ranging), camera and mil- limeter radar with GNSS (Global Navigation Satellite System) and IMU (Internal Measurement Unit), fused through hierarchical architecture using SLAM-based (Simultaneous Localization and Mapping) localization and deep learning models such as YOLOv8 (You Only Look Once) for real-time object detection. This research concludes that technical robustness and perceived safety are inseparable. Ac- curacy and stable sensor fusion produce smooth and predictable driving behavior, which passengers will associate with safety and trust, making technical performance the foundation of public acceptance.