High-Definition Maps: Comprehensive Survey, Challenges and Future Perspectives

In cooperative, connected, and automated mobility (CCAM), the more vehicles can perceive, model, analyze surrounding environment, they become aware capable of understanding, making decisions, as well safely efficiently executing complex driving scenarios. High-definition (HD) maps represent road environment with unprecedented centimetre-level precision lane-level semantic information, them a core component in smart systems, key enabler for CCAM technology. These provide strong prior to understand environment. An HD map is also considered hidden or virtual sensor, since it aggregates knowledge (mapping) from physical sensors, i.e. LiDAR, camera, GPS IMU build model Maps are quickly evolving towards holistic representation digital infrastructure cities include not only geometry but live perception participants, updates on weather conditions, work zones accidents. Deployment autonomous at large scale necessitates building maintaining these by fleet which cooperatively continuously keep up-to-date function properly. This article provides an extensive review various applications highly (AD) systems. We state-of-the-art different approaches algorithms maintain maps. Furthermore, we discuss synthesise data, communication requirements distribution Finally, current challenges future research directions next generation mapping