Abstract: In ADAS navigation maps, multi-source data fusion and ADAS navigation map data collection methods often cause elevations in the ADAS data that do not conform to actual anomalies. The data collected in different periods are different because of various systematic errors in the collection process (collection mode, collection environment, and personnel). This has led to the problem of the discontinuity of LINK level elevation that is currently prevalent in ADAS navigation map data. Due to space limitations, this article only discusses control layer processing issues. The ADAS elevation anomaly constraint correction algorithm based on the idea of global level network adjustment is proposed. Experiments show that this method can realize the processing of elevation discontinuity and elevation abnormality of the whole ADAS data. It has the advantages of good overall conformance, small change of original data elevation, and reasonable overall (all LINK nodes are not abnormal).

Key words: ADAS, abnormal elevation, minimum independent closed loop, overall best estimate, data reconstruction