CN-116595430-B - Inland ship track correlation method based on radar and AIS data fusion

Abstract

The invention discloses a inland ship track association method based on radar and AIS data fusion, wherein in the track association method in radar and AIS fusion, the distance between the radar track of a target ship and the AIS track is used as a coefficient matrix, and the optimal solution of track matching is finally found out through operations such as matrix transformation, trial assignment, niger coverage theory and the like. According to the invention, the optimal matching scheme of the ship radar track and the AIS track is obtained by exploring the similarity between different tracks in a multi-target complex scene, the interconnection corresponding relation between the AIS and the radar in the sailing process of the ship is found, the dynamic information fusion of the ship is completed, and the dynamic perception and resolution capability of the target ship in inland navigation is improved.

Inventors

• WANG YANXIA
• GAN SHAOJUN
• LIANG SHAN
• CHEN YANYAN

Assignees

• 北京工业大学

Dates

Publication Date

20260505

Application Date

20221227

Claims (1)

1. 1. The inland ship track association method based on radar and AIS data fusion is characterized by comprising the following implementation steps: S1 supposedly has n pieces of radar data respectively And n AIS data Calculating the distance between each piece of radar data and AIS data through a distance formula : ; Wherein the method comprises the steps of ; Longitude and latitude of the nth radar data respectively; longitude and latitude of the nth AIS data, respectively; S2 distance to Matrix of components As the coefficient matrix of the global track matching problem; S3, the ship to be matched in the radar data and each candidate ship in the AIS data are called a candidate pair, and a binary function is defined for each candidate pair : ; The mathematical model of trajectory matching is described as: ; It satisfies the following conditions: ; ; S4 pair coefficient matrix Firstly, subtracting the minimum value of the row from each element in each row, and then subtracting the minimum value of the column from each column element, wherein if 0 exists in a certain row or column in the matrix, subtraction is not needed; S5, performing trial assignment, namely, matrix coefficients Each row represents a different ship track in the radar data and each column represents a different ship track in the AIS data, the 0 element in a row or column is marked as @ from the row or column with only one 0 element, namely the radar ship track represented by the row is only related to the AIS track represented by the column or row, then the other 0 in the row or column with @ is deleted, recorded as @ and the AIS track represented by the row or column is related by the radar, and a coefficient matrix is obtained The same operation is performed on other rows and corresponding columns of the matrix, wherein the rows only contain one 0 element, until all 0 elements in the matrix are deleted or marked; S6 after S4, if the coefficient matrix still contains 0 element, then starting from the row with the least 0 element, comparing the number of 0 elements in the row or column, marking the row or column with the less 0 element as @, and deleting other 0 elements in the same row and column, repeating step S6 until the coefficient matrix All 0 elements in (1) are marked or deleted; S7 if coefficient matrix If the number of the steps is equal to the number of the marked @ elements, an optimal solution is obtained, and if the number of the steps is not equal, a step S8 is executed; s8, utilizing the Niger cover theory, performing the following steps: a) Marking the row without @ element with an x number; b) In the x-numbered row, all columns containing 0 element or including & are x-numbered; c) In the column marked with the x number, the x number is given to the row containing the @ element; d) Repeating the steps b) and c) until no row or column in the matrix is marked with a x number; e) Finally, drawing vertical lines for columns with the number of X in the coefficient matrix, and drawing horizontal lines for rows without the number of X; f) Setting k straight lines, returning to S5 if k is the same as the matrix order, and continuing the following steps if k is smaller than the matrix order; s9, finding out the minimum value in the elements which are not covered by the straight line in the matrix, subtracting the value from each element in the row with the X number, adding the value to each element in the column with the X number to obtain a new matrix, carrying out the same processing on the new matrix according to S5-S7, obtaining the optimal solution if n independent 0 elements are obtained, otherwise, returning to S8, repeating the processing to obtain the optimal solution, namely that AIS representing the 0 elements is matched with the ship represented by the radar data.

Description

Inland ship track correlation method based on radar and AIS data fusion Technical Field The invention relates to a method for fusing radar and AIS data and correlating a flight path, and belongs to the technical field of inland navigation digital technology. Background In inland shipping, radar and AIS are important means for identifying inland shipping ship targets, and information has complementarity. The AIS data can provide accurate ship position and attribute information, is high in navigation accuracy and is not easily influenced by factors such as position, weather and the like, but the AIS can only work at very high frequency, and a ship without AIS equipment cannot be identified. The radar can track and survey ships moving and stationary in a detection area at the same time, but the radar has a blind area, and echoes are easily influenced by weather, sea conditions, terrain shielding and the like, so that the problems of low resolution and the like are caused. Therefore, it is necessary to fully utilize the fusion information of the radar and the AIS. The key of the radar and AIS data fusion is track association of two target ships, and the correlation of the position and the motion state of the target ships is mainly based. At present, more track association methods are applied, such as a weighting method, a correction method, a nearest neighbor method and the like, and detection methods such as likelihood ratios, multiple assumptions and the like are also used, but under the conditions of dense ships, crossing or strong maneuverability, the methods often lead to a plurality of missed associated tracks. Disclosure of Invention The invention provides a inland ship track association method based on radar and AIS data fusion, which aims to solve the technical problems in the technical background and solve the track association problem in radar and AIS data fusion. In the track association method in the fusion of the radar and the AIS, the distance between the radar track of the target ship and the AIS track is mainly used as a coefficient matrix, and the optimal solution of track matching is finally found out through operations such as transformation, trial assignment, niger coverage theory and the like of the matrix, so that the method has the advantages of high accuracy, high reliability and the like, and the accurate and reliable association of the radar track of the ship and the AIS track is realized under a complex scene of multiple targets. In order to achieve the purpose, the invention provides the technical scheme that the inland ship track association method based on the fusion of radar and AIS data comprises the following implementation steps: S1 assuming that there are n pieces of radar data { (x r1,yr1),(xr2,yr2),…(xrn,yrn) } and n pieces of AIS data { (x a1,ya1),(xa2,ya2),…(xan,yan) }, respectively, a distance Q ij between each piece of radar data and AIS data is calculated by a distance formula: Qij＝6371·arcos[cos(yri)·cos(yaj)·cos(xri-xaj)+sin(yri)·sin(yaj)] Wherein i=1 to n, j=1:n, x rn,yrn is the longitude and latitude of the nth radar data, and x an,yan is the longitude and latitude of the nth AIS data. S2 uses a matrix q= (Qij) n ×n composed of distances Qij as a coefficient matrix of the global track matching problem. S3, each candidate ship in the ship to be matched in the radar data and the AIS data is called a candidate pair, and a binary function Xij is defined for each candidate pair: The mathematical model of trajectory matching is described as: It satisfies the following conditions: S4, transforming the coefficient matrix Q, firstly subtracting the minimum value of the row from each element in each row, and then subtracting the minimum value of the column from each column element. If there is already a 0 in a row or column in the matrix, then no subtraction is necessary. S5, performing trial assignment. For convenience of description, each row of coefficient matrix Q represents a different ship track in the radar data, and each column represents a different ship track in the AIS data. First, starting with a row or column having only one 0 element, the 0 element in that row or column is marked @, i.e. the radar vessel track represented by that row is only associated with the AIS track represented by that column or row. The other 0s in the column or row where @ is located are then deleted and denoted @ as @ and indicate that the AIS trace represented by this column or row has been radar correlated without regard to other radar vessel traces. Similarly, the same operation is performed on the other rows and their corresponding columns in the coefficient matrix Q that contain only one 0 element until all 0 elements in the matrix are deleted or marked. S6, after S4, if the coefficient matrix still contains 0 element, starting from the row with the least 0 element, comparing the number of 0 elements in the row or column, marking the row or column with the less 0 element as @, and