Objectives Automated driving provides an effective solution to facilitate travel, reduce pollution and improve transportation. High-definition maps have the role of location positioning, path planning, assisted decision-making, enhanced perception, etc. In supporting the process of automated driving, especially in the stage of high-level automated driving above L3 level, high-definition maps can greatly reduce the requirements for the arithmetic power of the vehicles, and help solve the problem of realizing better automated driving under the environment of the high arithmetic power hardware “necklace”. However, the contradiction between the urgent demand for high-definition maps for automated driving and the safety risks in practical applications is becoming more and more prominent.

Methods Based on the urgent demand for high-definition maps for automated driving and the safety risks in applications, we focus on safety supervivion of high-definition map for intelligent connected vehicle and aim to provide effective solution ideas for the safe application. First, the current situation of high-definition map safety supervision at home and abroad is introduced. Based on the systematic analysis of the safety risks throughout the whole lifecycle of high-definition map, from data collection, production, processing, release and application, a high-definition map safety supervision framework is proposed. Then, the geographic information decryption and desensitization, map safety scrutiny, spatial-temporal data provenance and other key technologies involved in the framework are further sorted out.

Results Taking the high-definition map safety supervision pilot in Beijing as a case study, we introduce the current status of its application and exploration experience. After a series of exploration, Beijing has established an advanced mechanism for dynamic supervision of the safety of high-definition maps for intelligent connected vehicles by pre-registration, mid-monitoring, and post-tracing. A platform for the safety supervision of maps is also set up, which comprehensively utilizes geographic information, big data, 5G, artificial intelligence and other technology.

Conclusions High-definition maps, as important basic resources and data elements, are an important link in the automated driving industry and the spatial-temporal information industry chain. To better secure high-definition maps, on the one hand, the government should empower policy propaganda and also solidify the supervision, on the other hand, enterprises and institutions engaged in surveying and mapping activities should also strengthen safety awareness and protection.