Review on Status and Challenges of Crowdsourced Updating of Highly Automated Driving Maps

This study focuses on the challenges associated with real-time dynamic updates in the high-precision, high-reliability, and high-security aspects of highly automated driving (HAD) maps. The aim is to summarize the existing difficulties and challenges encountered in high-definition map updates, accelerate the large-scale commercialization of autonomous driving maps, enhance the safety and stability of intelligent vehicles, and provide important support for high-level autonomous driving systems. Therefore, this article first describes the definition and connotation of HAD maps, highlighting their data characteristics and functional applications. Next, it summarizes the development status and trends of high-definition map updates and evaluates the advantages and disadvantages of a centralized map and multisource updates, indicating that crowdsourced updates have become a new trend in map update development. Furthermore, it summarizes the basic architecture and core modules of crowdsourced updates, analyzes the core technologies involved, and summarizes the current situation and trends of key technologies involved in crowdsourced updates. Results of the analysis of the crowdsourced update technology for HAD maps show that the current technology still faces seven significant challenges, including map modeling, high-precision positioning, 3D reconstruction, fusion updates, data security, fast review, and standardization of laws and regulations. In response to these challenges, the study suggests that addressing the technical, policy, and legal issues of crowdsourced update technology requires the cooperation of multiple departments from governments, industry, academia, and research. The collective effort will accelerate the development and application of crowdsourced update technology for HAD maps. © 2023 Xi'an Highway University. All rights reserved.