Gray Correlation-Based Algorithm for Fusion of Shore-based Radar and AIS Targets in Inland Waters

Traffic flow in inland river waters is dense, and the phenomenon of AIS communication congestion and delay is very serious. The traditional AIS and radar fusion algorithm mainly considers the position of the ship, and matches the AIS and radar tracks in time and space through coordinate conversion and time alignment, which is difficult to meet the business needs of ship supervision in special areas. In view of the actual needs of complex inland waters, based on the consideration of the ship's latitude and longitude, the ship's scale, as well as the heading and speed are introduced to construct the multivariate matrix-type correlation feature vectors, calculate the gray correlation of different information sources in the same waters, and at the same time, introduce the average correlation degree of correlation attribute feature vectors in order to improve the correlation correctness rate and increase the correlation degree stability. The ship data are exported from the ferry navigation collision avoidance intelligent monitoring system, and the ship scale data measured by radar are corrected to improve the robustness of the fusion algorithm. Experiments show that: the fusion success rate of this algorithm is 74.6% for ships with length greater than 80m, and the probability of fusion stabilization in the successfully fused data is 80%; the fusion success rate for ships with length less than 80m is 68.6%, and the probability of fusion stabilization in the successfully fused data is 72 %. The algorithm has good applicability to radar and AIS target information matching and fusion.