The Key Technology of High-Definition Maps Distribution Based on Edge Computing

Abstract

The High-Definition map (HD map) is a key technology to achieve automatic driving above the grade of L3, with the amount of data comes to more than 105 times that of traditional navigation map. With the arrival of 5G communication and the rapid development of Internet of Things, each autonomous vehicle will request HD map service by accessing the Internet. The service of traditional navigation map under “cloud-end” mode may not be well adapted to future HD map application, so the paper proposes a MEC proposal for HD map application, deploying HD map server under “cloud-edge-end” mode to mitigate the high latency and improve low reliability caused by the faraway physical distance rather than “cloud-end” mode. The HD map is divided according to the latitude and longitude regions, being pushed to local edge computing node servers respectively to realize the interaction between autonomous vehicles and edge servers, which makes more convenient and reliable HD map services available. The experimental results show that the proposed proposal can cut off the communication delay effectively, ensuring the reliability of the HD map service, and providing high-quality HD map service for autonomous vehicles.