CN113686336A - A method for improving underwater navigation accuracy based on iterative optimal loop points based on grid topology - Google Patents

Abstract

The invention discloses a method for improving underwater navigation precision based on a grid topological structure iteration optimal ring domain point, which comprises the following steps: obtaining the optimal matching position of the track starting point of the underwater vehicle through a track starting point small-ring area grid matching positioning strategy; generating lattice points to be matched in a large ring area by a track end point angle-variable three-layer ring area matching positioning strategy according to the optimal matching position of the track starting point of the underwater vehicle; and (3) iteratively calculating the matching efficiency evaluation index of the grid points to be matched in the large ring area, and obtaining the optimal matching position of the underwater vehicle track terminal in the large ring area range according to the optimal principle so as to realize the effective matching and positioning of the underwater vehicle track terminal, further correcting the INS system control parameters and assisting in finishing the navigation target of the underwater vehicle with long voyage and long voyage distance. By the method, the matching precision of gravity-assisted navigation of the underwater vehicle is improved.

Description

A method for improving underwater navigation accuracy based on iterative optimal loop points based on grid topology

technical field

The invention belongs to the cross technical fields of underwater navigation, oceanographic mapping and the like, and in particular relates to a method for improving underwater navigation accuracy based on iterative optimal ring domain points based on a grid topology structure.

Background technique

Inertial Navigation System (INS) is the core navigation system to realize autonomous navigation of underwater submersibles. It has the characteristics of short-term high-precision positioning, but the inherent errors of inertial components and the multiple integration of positioning solutions lead to INS. The error accumulates and diverges with time, and it is difficult to meet the high-precision positioning target of the long-running submersible. Therefore, the INS needs to be calibrated regularly to maintain the navigation accuracy.

As one of the inherent geographical attributes of the earth, gravity field information is not easily affected by uncertain environments such as climate and ocean waves, and exhibits long-term relative stability. Therefore, gravity field information is suitable to be used for auxiliary navigation. At present, gravity-assisted navigation systems are used as underwater aids. INS navigation is an important technology and has become an international hot topic of research by scholars at home and abroad.

The matching algorithm is the core of the gravity-assisted inertial navigation system. At present, the common gravity matching algorithms mainly include the Sandia inertial terrain-assisted navigation algorithm SITAN, the iterative nearest contour point algorithm ICCP and the terrain contour matching algorithm TERCOM. In comparison, the TERCOM algorithm has received extensive attention and research by scholars for its advantages of simple calculation, insensitivity to initial error, strong robustness, and high positioning accuracy.

In improving the matching accuracy of the TERCOM algorithm, Liu et al. proposed a novel INS/TERCOM system optimization structure combining coordinate transformation, mismatch detection and Kalman filtering, and studied the matching of its decentralized and adaptive federated filtering information fusion algorithms performance; Yan et al. proposed a new matching algorithm by integrating TERCOM and ICCP, using TERCOM to obtain the initial position and using ICCP for precise positioning; Yuan et al. proposed a combined underwater assisted navigation algorithm by combining the TERCOM/ICCP algorithm with Kalman filtering, and at the same time The exact matching accuracy uses a sliding window to improve the efficiency of the algorithm; Wang et al. proposed a rotation-splicing gravity matching algorithm based on the TERCOM algorithm; Tong et al. approximated the local gravity reference map with a two-dimensional Gaussian function and solved it with a quasi-Newtonian BFGS nonlinear optimization method. A combined matching algorithm based on local gravity map approximation was proposed by calculating the relevant extreme value matching model. At the same time, the matching accuracy of the algorithm was improved by the rough matching of TERCOM and the preprocessing of measured data by the difference method; Zhao et al. combined the TERCOM algorithm with particle filtering. A new terrain-aided navigation algorithm was proposed to enhance the positioning accuracy of the BITANII algorithm; Wei et al. proposed a correlation SITAN algorithm based on weighted decreasing iteration to solve the initial error and linear error problems, and used TERCOM for the correlation processing of the algorithm. Process; Zhang et al. proposed an online judgment criterion for joint probability mismatch of multiple reference points in the correlation plane to solve the mismatch problem that the TERCOM algorithm is susceptible to elevation measurement errors, terrain similarity, etc.; Wang et al. The TERCOM algorithm realizes sonar-free positioning in the deepest part of the earth's oceans; Zhang et al. used the TERCOM algorithm to simulate and analyze the influence of the main factors such as the speed of the submersible, the sounding accuracy, the initial position deviation, the characteristics of the underwater terrain, and the resolution of the digital map on the matching accuracy. law. In terms of improving the matching efficiency and reliability of the TERCOM algorithm, Han et al. integrated the shortest path algorithm and the new correlation analysis algorithm to construct an improved TERCOM algorithm. At the same time, they proposed a mismatch diagnosis method through the spatial order constraint and the decision criterion constraint integration mechanism. To improve the reliability and matching accuracy of TERCOM; Liu Xianpeng et al. tracked the track of the submersible with the speed and heading information output by INS and proposed a TERPM positioning algorithm based on track line tracking; Li Zhaowei et al. proposed a new hierarchical neighborhood threshold search method to improve the matching efficiency of the point-by-point traversal search of the TERCOM algorithm; Li et al. proposed a method by coupling the shortest arc principle of spherical geometry and attitude control theory in air-sea environment The new geodesic-based method can reduce the size of the matching area and improve the matching efficiency of the algorithm; Zhang et al. used TERCOM as the line matching algorithm to perform surface matching with geometric similarity and proposed a line-surface combination underwater terrain matching algorithm to improve the algorithm. Robustness and positioning accuracy.

To sum up, most scholars mainly focus on the application of TERCOM and the improvement of the navigation performance of underwater submersibles, and there is little research work on changing the topology of TERCOM matching grid. However, due to the fact that the traditional TERCOM algorithm does not rely on the position information of the starting point of the track, it only uses 3 times the INS error at the end of the track as the half length to form a square grid matching lattice, and traverse the search to determine the best match for the position of the underwater vehicle However, this search mechanism has a large amount of computation and is easy to lead to low matching efficiency of the algorithm; in addition, the TERCOM algorithm is difficult to effectively deal with observation noise and process noise, and is sensitive to the angle error of the inertial navigation segment. The design problem of matching grid topology.

SUMMARY OF THE INVENTION

The technical solution of the present invention is to overcome the deficiencies of the prior art, and to provide a method and system for improving the accuracy of underwater navigation based on iterative optimal loop points based on grid topology, aiming to improve the matching accuracy of gravity-assisted navigation of underwater submersibles. .

In order to solve the above technical problems, the present invention discloses a method for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology structure, including:

The optimal matching position of the track starting point of the underwater vehicle is obtained through the grid matching and positioning strategy of the small ring area of the track starting point;

According to the best matching position of the starting point of the track of the underwater vehicle, the grid points to be matched in the large ring area are generated through the variable angle three-layer ring field matching and positioning strategy at the end point of the track;

Iteratively calculate the matching efficiency evaluation index of the grid points to be matched in the large ring domain, and obtain the best matching position of the end point of the track of the underwater vehicle within the scope of the large ring domain according to the optimal principle, so as to realize the effective performance of the end point of the track of the underwater vehicle. Matching and positioning, and then correcting the control parameters of the INS system and assisting the completion of the long-duration and long-distance navigation goal of the underwater submersible.

In the above-mentioned method of improving underwater navigation accuracy based on iterative optimal loop points based on grid topology, the construction process of the small loop grid matching positioning strategy at the starting point of the track is as follows:

Obtain the INS track output inertial navigation sequence {S ₁ , S ₂ ,..., S _L } of the INS track of a certain submersible voyage according to the sampling time interval Δt; where L represents the sampling sequence length of the INS track, S ₁ , S ₂ , ..., _SL represents each track point in the inertial navigation sequence;

From {S ₁ , S ₂ ,…,S _L }, the position coordinates and measured gravity values corresponding to each track point are extracted, and the position coordinate sequence {(x ₁ , y ₁ ), (x ₂ , y ₂ ) is constructed and obtained ,…,(x _L ,y _L )} and the underwater vehicle’s measured gravity value sequence {g ₁ ,g ₂ ,…,g _L };

Taking the position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ in the inertial navigation sequence as the center, and taking 3σ ₀ as the maximum extension search boundary radius, a small ring domain is constructed; among them, σ ₀ indicates that the sampling interval is Δt The standard deviation of the inertial navigation drift error when ;

According to the north-easterly rotation angle θ ₀ and the ring radius proportional sharing coefficient λ, the total number and position coordinates of the grid points to be matched in the small ring domain are determined, and the set of grid points to be matched in the small ring domain is constructed.

的计算公式如下：In the above method for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology, the position coordinates of each grid point to be matched in the small ring domain

The calculation formula is as follows:

和

分别表示第i个小环上第j个格点的横纵坐标，r_i表示小环域内第i个环的半径，β_j表示小环域各环上第j个格点的旋转角度。in,

and

respectively represent the abscissa and vertical coordinates of the jth grid point on the _ith ringlet, ri represents the radius of the ith ring in the ringlet domain, and βj represents the rotation angle of the _jth grid point on each ring of the ringlet.

In the above method to improve underwater navigation accuracy based on iterative optimal loop points based on grid topology,

r _i =3σ ₀ λi

β _j =j·θ ₀

且j的最大值

Among them, i=1,2,..., and the maximum value of i

and the maximum value of j

In the above method of improving underwater navigation accuracy based on iterative optimal loop points based on grid topology, the optimal matching position of the track starting point of the underwater vehicle is obtained through the grid matching and positioning strategy of the small loop starting point of the track, including:

作为航迹起点真实位置的待匹配点集；The position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ and the position coordinates of the grid points to be matched in the small ring domain

The set of points to be matched as the true position of the track start point;

和

对应的理论重力值

Map the set of points to be matched at the real position of the track starting point to the gravity reference map point by point, and take the gravity value at the nearest gravity reference grid point as the gravity value of the point to be matched, and obtain the theoretical gravity corresponding to (x ₁ , y ₁ ) value

and

The corresponding theoretical gravity value

Among them, C represents the grid resolution of the gravity reference map, mapt(·,·) represents the gravity value matrix of the gravity reference map according to the grid position, and [·] represents the rounding;

According to the principle of minimizing the absolute value of gravity deviation, the best matching position of the starting point of the track of the underwater vehicle is obtained

即为x₁、

即为y₁。Among them, when i=0, j=0,

is x ₁ ,

That is y ₁ .

In the above-mentioned method for improving underwater navigation accuracy based on iterative optimal loop points based on grid topology, the construction process of the three-layer loop matching positioning strategy with variable angle at the end of the track is as follows:

According to the position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ and the position coordinates (x _L , y _L ) of the track end point S _L in the inertial navigation sequence, it is determined to obtain the heading information and navigation of the underwater vehicle. Distance information:

Among them, d _INS represents the distance metric between the start and end points of the inertial navigation track with the h∈[0, +∞) norm, and set h=2, that is, calculate the Euclidean distance; α _INS represents the abscissa of the coordinate system as a positive The heading radian angle of the direction;

以及水下潜器航行的航向信息和航距信息，估算得到大环域的中心位置坐标(x_O,y_O)：according to

As well as the heading information and distance information of the underwater vehicle navigation, the center position coordinates (x _O , y _O ) of the large ring domain are estimated:

Taking (x _O , y _O ) as the center, and taking R _max as the maximum extension search boundary radius, a variable-angle three-layer topology ring grid point area covering the true end point of the track is constructed, which is denoted as a large ring area;

Taking the grid resolution C of the gravity reference map as the span interval of each ring in the large ring, the total number of rings in the large ring is obtained

的半径之比作为大环域内各格点生成的基准角

按“内倍外半”原则，确定大环域内层环上相邻格点间的跨度角为

外层环上相邻格点间的跨度角为

并结合总环数M，构建得到大环域待匹配格网点集；其中，大环域待匹配格网点集为变角度三环层格点拓扑结构。Compare the grid resolution C of the gravity datum map with the middle ring

The ratio of the radii of , as the reference angle generated by each grid point in the large ring domain

According to the principle of "inner double outer half", the span angle between adjacent lattice points on the inner ring of the large ring domain is determined as

The span angle between adjacent lattice points on the outer ring is

Combined with the total number of rings M, the set of grid points to be matched in the large ring domain is constructed; among them, the set of grid points to be matched in the large ring domain is a variable-angle three-ring layer grid point topology.

In the above-mentioned method for improving underwater navigation accuracy based on iterative optimal loop points based on grid topology, the grid points to be matched in the large loop are generated as follows:

其中，σ表示惯导累积漂移误差标准差；The total number of rings in the macrocyclic domain under the three mechanisms of 1σ-EPMP, 2σ-EPMP and 3σ-EPMP is denoted as

Among them, σ represents the standard deviation of the inertial navigation accumulated drift error;

分别进行修正，得到修正后的三种机制下的大环域的总环数M′₁、M′₂、M′₃：right

Corrected respectively, to obtain the total ring numbers M′ ₁ , M′ ₂ , M′ ₃ of the macroring domain under the three modified mechanisms:

Among them, ξ=1, 2, 3, corresponding to 1σ-EPMP mechanism, 2σ-EPMP mechanism, and 3σ-EPMP mechanism respectively;

According to the total ring numbers M ₁ ′, M ₂ ′ and M ₃ ′ of the macrocyclic domain under the three modified mechanisms, the radii R _1,j , R _2,j , R _3,j :

R _ξ,ζ =ζC,ζ=1,2,...,M' _ξ ...(6)

Among them, R _ξ,ζ represents the radius of the ζth ring of the macrocyclic domain under the ξth mechanism;

According to the revised total ring numbers M′ ₁ , M′ ₂ , M′ ₃ of the macrocyclic domain under the three mechanisms, the intermediate ring radius of the macrocyclic domain under the three mechanisms is obtained

得到三种机制下大环域内各格点生成的基准角

According to the middle ring radius of the macrocyclic domain under the three mechanisms

Obtain the reference angle generated by each lattice point in the large ring domain under the three mechanisms

According to (5)-(8), the grid points to be matched in the macroring domain under different mechanisms are obtained.

In the above method to improve underwater navigation accuracy based on iterative optimal loop points based on grid topology,

的计算公式如下：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the large ring domain under the 1σ-EPMP mechanism

The calculation formula is as follows:

的计算公式如下：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the large ring domain under the 2σ-EPMP mechanism

The calculation formula is as follows:

的计算公式如下：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the macroring domain under the 3σ-EPMP mechanism

The calculation formula is as follows:

In the above method of improving underwater navigation accuracy based on iterative optimal loop points based on grid topology, iteratively calculates the matching efficiency evaluation index of grid points to be matched in the large loop, and obtains the underwater diving within the large loop according to the optimal principle. The best matching position of the end point of the track of the aircraft, including:

Determine the total number N _ξ of grid points to be matched on the inner-middle-outer ring layer of the macroring domain under the ξth mechanism;

其中，r∈{1,2,…,N_ξ}；Take the rth grid point to be matched in the large ring domain under the ξth mechanism as the end point of the track to be evaluated, denoted as a point

where, r∈{1,2,…, _Nξ };

在重力基准图中对应的位置(x_r,y_r)与重力基准图的格网分辨率C相比，并按四舍五入原则得到重力基准图上最近邻于点

的格网点base_L；will point

The corresponding position (x _r , y _r ) in the gravity reference map is compared with the grid resolution C of the gravity reference map, and the nearest adjacent point on the gravity reference map is obtained according to the rounding principle.

The grid point base _L of ;

作为点

重力值的近似；Set the gravity value corresponding to the grid point base _L

as a point

Approximation of gravity value;

对应的重力图航迹序列

及对应的最近邻重力序列

According to the position coordinates of the grid point base _L on the gravity reference map, the speed and heading of the underwater vehicle, the points are extracted.

Corresponding gravity map track sequence

and the corresponding nearest neighbor gravity sequence

与水下潜器实测重力值序列{g₁,g₂,…,g_L}进行对比，并计算匹配效能评价指标，记作MSD_r；Will

Compare with the actual gravity value sequence {g ₁ ,g ₂ ,...,g _L } of the underwater vehicle, and calculate the matching efficiency evaluation index, denoted as MSD _r ;

Calculate the matching efficiency evaluation index of each grid point to be matched on the inner-middle-outer ring layer of the macroring domain under the ξth mechanism in turn, and obtain the inner-middle-outer ring layer of the macroring domain under the ξth mechanism. The matching performance evaluation index set and {MSD _r |r=1,2,...,N _ζ } of all grid points to be matched of ;

Based on {MSD _r |r=1,2,…,N _ζ }, the best matching position of the end point of the track of the underwater vehicle in the large ring domain is obtained by filtering according to the optimal principle.

Correspondingly, the present invention also discloses a system for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology structure, including:

The solving module is used to obtain the best matching position of the starting point of the track of the underwater vehicle by matching the positioning strategy with the small ring grid at the starting point of the track;

The generation module is used to generate the grid points to be matched in the large ring area according to the best matching position of the starting point of the track of the underwater vehicle, through the variable angle three-layer ring matching and positioning strategy at the end point of the track;

The iterative determination module is used to iteratively calculate the matching efficiency evaluation index of the grid points to be matched in the large ring area, and obtain the best matching position of the end point of the track of the underwater vehicle within the large ring area according to the optimal principle, so as to realize underwater diving. It can effectively match and locate the end point of the track of the underwater vehicle, and then correct the control parameters of the INS system and assist in the completion of the long-duration and long-distance navigation goal of the underwater submersible.

The present invention has the following advantages:

In order to break through the limitation of the inherent grid structure of the traditional gravity matching algorithm and improve the navigation accuracy of underwater gravity matching, the present invention proposes a new iterative optimal loop point method (IOAP) for the grid topology structure. The matching and positioning strategy of the small ring grid of the track starting point is constructed by guiding the starting point position, drift error and rotation angle. Through the matching comparison of the matching points of the small ring grid, the best matching positioning of the track starting point is obtained and the algorithm is enhanced. The insensitivity of the initial position error; secondly, using the best matching position of the starting point of the track, combined with the inertial navigation distance information, accumulated drift error, etc. Finally, iteratively calculates the matching index of the matching points in the loop domain and obtains the best matching position of the track end point within the loop domain according to the optimal principle.

Considering the statistical indicators of matching accuracy, average matching time and matching success rate, etc. as the analysis basis for matching quality, it verifies the matching performance difference of the method described in the present invention under different inertial navigation cumulative error multiples or reference angle ring radii. Its good robustness.

This is, by comparing the gravity matching test of the tracks with the starting points and ending points in different gravity zones in different regions, it is proved that the method of the present invention has the advantages of high matching accuracy and strong positioning applicability in different gravity zones. Compared with the TERCOM algorithm, the average matching accuracy and worst matching accuracy are improved by 40.39% and 72.16% respectively.

Description of drawings

1 is a flow chart of the steps of a method for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology structure in an embodiment of the present invention;

2 is a schematic diagram of the distribution of points to be matched in a small ring domain based on an SPMP strategy in an embodiment of the present invention;

3 is a schematic diagram of the distribution of points to be matched in a large ring domain based on an EPMP strategy in an embodiment of the present invention;

4 is a schematic diagram of the distribution of gravity anomalies in a study area satellite remote sensing and a local enlarged area in the embodiment of the present invention; wherein, 4(a) is a satellite remote sensing map, and 4(b) is a gravity anomaly reference map;

5 is a schematic diagram of a comparison of the matching test effects of gravity matching algorithms under different σ criteria in an embodiment of the present invention; wherein, 5(a) is the TERCOM algorithm, 5(b) is the 1σ-IOAP algorithm, and 5(c) is the 2σ algorithm -IOAP algorithm, 5(d) is 3σ-IOAP algorithm;

Fig. 6 is a bar graph comparison of the successful matching probabilities of four algorithms under a different positioning accuracy in an embodiment of the present invention;

7 is a schematic diagram of the comparison between the matching position of a TERCOM algorithm and the real position and the classification of grid points in an embodiment of the present invention; wherein, 7(a) is 100 tests of the TERCOM algorithm, and 7(b) is the matching position point of the TERCOM algorithm Classification;

FIG. 8 is a schematic diagram of comparing the matching position and the real position of a different σ-IOAP algorithm in an embodiment of the present invention; wherein, 8(a) is 100 tests of the 1σ-IOAP algorithm, and 8(b) is 100 times of the 2σ-IOAP algorithm Test, 8(c) is 100 tests of 3σ-IOAP algorithm;

9 is a schematic diagram of the comparison of the matching effects of the IOAP algorithm under a different ring radius reference angle in an embodiment of the present invention; wherein, 9(a) is the TERCOM algorithm, 9(b) is the 1-IOAP algorithm, and 9(c) is 1.5 -IOAP algorithm, 9(d) is 2-IOAP algorithm, 9(e) is 2.5-IOAP algorithm;

Fig. 10 is a columnar comparison chart of the IOAP algorithm matching success probability under a kind of different ring radius reference angles in the embodiment of the present invention;

11 is a schematic diagram of the comparison of the matching and positioning effects of algorithms under different track starting points in the embodiment of the present invention; wherein, 11(a) is the TERCOM algorithm (track starting point A), and 11(b) is the TERCOM algorithm (track starting point A) B), 11(c) is TERCOM algorithm (track starting point C), 11(d) is 1.5-IOAP algorithm (track starting point A), 11(e) is 1.5-IOAP algorithm (track starting point B), 11 (f) is the 1.5-IOAP algorithm (track start point C).

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the embodiments disclosed in the present invention will be described in further detail below with reference to the accompanying drawings.

Based on the TERCOM algorithm matching the structure layout of the lattice and the drift error characteristics of the INS system, the present invention proposes a method for improving the underwater navigation accuracy based on the iterative optimal loop points based on the grid topology structure, which is referred to as the IOAP algorithm, and its realization principle is as follows: In order to reduce the sensitivity of the IOAP algorithm to the initial error, the starting point position of the inertial navigation system is taken as the center, and a small ring field matching lattice is formed with a certain drift error and rotation angle, and a small ring field matching and positioning strategy of the track start point of the ring topology is constructed. , and obtain the optimal matching position of the starting point according to the principle of minimizing the absolute value of the gravity deviation; according to the optimal matching position of the starting point of the track, combined with the information of the inertial navigation direction, the center position of the matching grid in the large ring domain is obtained, and then based on the inertial navigation Accumulate drift error, etc. to determine the number of matching grid rings in the large ring domain, and according to the grid point deflection angle of the intermediate ring radius and the principle of "inside double outer half", obtain the topology structure matching the annular distribution of grid points, and construct the track end point The matching and positioning strategy of the variable-angle three-layer ring domain is to extract the gravity map sequence through grid points, and then calculate the mean square error between the measured gravity sequence and the measured gravity series, and then match and compare, and obtain the best matching position of the end of the track according to the principle of minimizing the mean square error.

As shown in Figure 1, in this embodiment, the method for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology structure includes:

Step 101 , obtain the best matching position of the track starting point of the underwater vehicle through the grid matching and positioning strategy of the track starting point.

In this embodiment, considering the relative sensitivity of gravity matching algorithms such as SITAN and ICCP to the initial error, and at the same time to improve the insensitivity (robustness) of the IOAP algorithm to the initial error of gravity matching to a certain extent, a navigation system is constructed. Matching Positioning strategy of the tracking Starting Point in small ring domain (SPMP). Among them, the SPMP strategy takes the position of the underwater vehicle indicated by the inertial navigation as the center, and uses a certain drift error and rotation angle to form a small annular topology grid point area to be matched that covers the true starting point position of the track according to the probability (denoted as Small ring domain), and then determine the best matching position of the starting point of the track through the optimal principle of the gravity matching evaluation index of the point to be matched.

Preferably, the construction process of the small ring domain grid matching positioning strategy at the starting point of the track may be as follows:

Obtain the inertial navigation sequence {S ₁ ,S ₂ ,…,S _L } of the INS track output by the sampling time interval Δt of a certain underwater vehicle navigation; extracted from {S ₁ ,S ₂ ,…,S _L } The position coordinates and the measured gravity value corresponding to each track point are constructed to obtain the position coordinate sequence {(x ₁ , y ₁ ), (x ₂ , y ₂ ),...,(x _L , y _L )} and the submersible Sequence of measured gravity values {g ₁ ,g ₂ ,…,g _L }. Among them, L represents the sampling sequence length of the INS track, and S ₁ , S ₂ , ..., _SL represent each track point in the inertial navigation sequence.

Then, under the SPMP strategy, take the position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ in the inertial navigation sequence as the center, and take 3σ ₀ as the maximum extension search boundary radius to determine the small ring domain. Further, according to the north easterly rotation angle θ ₀ and the ring radius proportional sharing coefficient λ, the total number and position coordinates of the grid points to be matched in the small ring domain can be determined, and then the set of grid points to be matched in the small ring domain can be constructed. Among them, σ ₀ represents the standard deviation of the inertial navigation drift error when the sampling interval is Δt. For example, when θ ₀ =45 and λ = 1/3, the distribution of the grid point set to be matched in the small ring domain stretched by the SPMP strategy is shown in FIG. 2 .

的计算公式如下：Further, according to the setting of the above parameters, the position coordinates of each grid point to be matched in the small ring domain can be obtained.

The calculation formula is as follows:

和

分别表示第i个小环上第j个格点的横纵坐标；r_i表示小环域内第i个环的半径，r_i＝3σ₀λi；β_j表示小环域各环上第j个格点的旋转角度，β_j＝j·θ₀；i＝1,2,…，且i的最大值

且j的最大值

in,

and

respectively represent the abscissa and vertical coordinates of the jth lattice point on the _ith ringlet; ri represents the radius of the _ith ring in the ringlet domain, ri = 3σ ₀ λi; β _j represents the jth ring on each ring of the ringlet domain Rotation angle of lattice point, β _j =j·θ ₀ ; i=1,2,..., and the maximum value of i

and the maximum value of j

Then, the process of determining the best matching position of the starting point of the track of the underwater vehicle can be as follows:

作为航迹起点真实位置的待匹配点集；逐点将航迹起点真实位置的待匹配点集映射到重力基准图上，并按最近重力基准格点处的重力值作为待匹配点重力值，得到(x₁,y₁)对应的理论重力值

和

对应的理论重力值

The position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ and the position coordinates of the grid points to be matched in the small ring domain

The set of points to be matched as the real position of the starting point of the track; the set of points to be matched at the real position of the starting point of the track is mapped to the gravity reference map point by point, and the gravity value at the nearest gravity reference grid point is used as the gravity value of the point to be matched, Get the theoretical gravity value corresponding to (x ₁ , y ₁ )

and

The corresponding theoretical gravity value

In order to determine the best matching position of the starting point of the track of the underwater vehicle, the best matching position of the starting point of the track of the underwater vehicle can be obtained according to the principle of minimizing the absolute value of the gravity deviation.

即为x₁、

即为y₁。Among them, C represents the grid resolution of the gravity reference map, mapt(·,·) represents the gravity value matrix of the gravity reference map according to the grid point position, [·] represents the rounding; when i=0, j=0,

is x ₁ ,

That is y ₁ .

It should be noted that the multi-modal phenomenon may occur when the gravity reference map is mapped to a small range of multi-grid points. In this embodiment, the first grid point with the smallest absolute value of the gravity deviation is matched. Of course, the navigation can also be obtained according to the random selection mechanism. The best matching of the starting point of the trace is not limited in this embodiment.

It can be seen from the above that the SPMP strategy can realize the effective matching and positioning of the starting point of the track of the underwater vehicle according to the probability, and weaken the sensitivity of the IOAP algorithm to the initial error. Matching positioning provides the basis for the location information of the origin of the track.

In step 102, according to the best matching position of the starting point of the track of the underwater vehicle, the grid points to be matched in the large ring domain are generated through the variable angle three-layer ring domain matching and positioning strategy at the end point of the track.

可进一步推得航迹终点待匹配域的中心位置O；再基于惯导系统的漂移误差统计特性而构建新的待匹配格网点环形分布的拓扑结构及其匹配定位策略(航迹终点变角度三层环域匹配定位策略/机制，MatchingPositioning mechanism of the tracking Ending Point in three-layer loopdomain，EPMP)，EPMP策略以惯导累积漂移误差标准差σ、重力基准图的格网分辨率C和中间环的半径三者间的相对关系，得到环形覆盖区域的最大环数和匹配格网点间的偏转角并张成一个大的依概率覆盖航迹真实终点位置的待匹配的变角度三层拓扑结构环型格网点区域，记为大环域；再按各格网点坐标位置提取出匹配航迹的重力图序列再与水下潜器的真实重力序列比对并按评价指标最优原则确定出航迹终点的最佳匹配位置。In this embodiment, considering that the inertial navigation track sequence contains better short-term high-precision heading and distance information, the best matching position of the track starting point obtained by combining the SPMP strategy

The center position O of the track end point to be matched can be further deduced; then based on the drift error statistical characteristics of the inertial navigation system, a new topology structure of the grid points to be matched is constructed and its matching positioning strategy (track end point variable angle three) is constructed. Layer loop domain matching positioning strategy/mechanism, MatchingPositioning mechanism of the tracking Ending Point in three-layer loopdomain, EPMP), EPMP strategy is based on the inertial navigation cumulative drift error standard deviation σ, the grid resolution C of the gravity reference map and the intermediate loop The relative relationship between the three radii, the maximum number of loops in the annular coverage area and the deflection angle between the matching grid points are obtained, and a large variable-angle three-layer topology to be matched covering the true end position of the track according to probability is formed. The grid point area is recorded as a large ring area; then the gravity map sequence matching the track is extracted according to the coordinate position of each grid point, and then compared with the real gravity sequence of the underwater vehicle, and the end point of the track is determined according to the principle of optimal evaluation index. Best match location.

Preferably, the construction process of the variable-angle three-layer ring domain matching positioning strategy at the track end point may be as follows:

According to the position coordinates (x ₁ , y ₁ ) of the track starting point S ₁ and the position coordinates (x _L , y _L ) of the track end point S _L in the inertial navigation sequence, it is determined to obtain the heading information and navigation of the underwater vehicle. Distance information:

Among them, d _INS represents the distance metric between the start and end points of the inertial navigation track with the h∈[0, +∞) norm, and set h=2, that is, calculate the Euclidean distance; α _INS represents the abscissa of the coordinate system as a positive The heading angle in radians of the direction.

以及水下潜器航行的航向信息和航距信息，估算得到大环域的中心位置坐标(x_O,y_O)：Then, according to

As well as the heading information and distance information of the underwater vehicle navigation, the center position coordinates (x _O , y _O ) of the large ring domain are estimated:

Then, under the EPMP strategy, with (x _O , y _O ) as the center and R _max as the maximum extension search boundary radius, a variable-angle three-layer topology ring grid point area covering the true end point of the track is constructed, denoted as Large ring domain. At the same time, the grid resolution C of the gravity reference map is used as the span interval of each ring in the large ring, and the total number of rings in the large ring is obtained.

的半径之比作为大环域内各格点生成的基准角

按“内倍外半”原则，确定大环域内层环上相邻格点间的跨度角为

外层环上相邻格点间的跨度角为

并结合总环数M，构建得到大环域待匹配格网点集，为一变角度三环层格点拓扑结构。例如，当

M＝9时，EPMP策略所张成的大环域待匹配格网点集的分布如图3所示。On this basis, given the deflection angle between adjacent grid points on each ring of the large ring domain, a set of grid points to be matched can be formed. If you choose the same method as the SPMP strategy to determine the deflection angle between adjacent lattice points on each ring of the large ring domain, the deflection angle of the large ring domain will keep the same amount of grid points on each ring, which will lead to the expansion of the SPMP strategy. There is a phenomenon of “inner dense and outer sparse” in the grid of , that is, the interval between the inner ring grid points is too small and the outer ring grid point interval is too large. Therefore, for the large ring domain, the embodiment of the present invention proposes a new method for determining the deflection angle between adjacent grid points, that is, the variable-angle three-ring layer grid point topology: the grid resolution C of the gravity reference map is used. with middle ring

The ratio of the radii of , as the reference angle generated by each grid point in the large ring domain

According to the principle of "inner double outer half", the span angle between adjacent lattice points on the inner ring of the large ring domain is determined as

The span angle between adjacent lattice points on the outer ring is

Combined with the total number of rings M, the set of grid points to be matched in the large ring domain is constructed, which is a variable-angle three-ring layer grid point topology. For example, when

When M=9, the distribution of the grid point set to be matched in the large ring domain formed by the EPMP strategy is shown in Fig. 3 .

In this embodiment, considering the good applicability of normal distribution to any systematic error under natural conditions and the high probability coverage of 99.73% of the 3σ criterion, at the same time, different σ(σ) is used for the subsequent tests, indicating the standard deviation of the inertial navigation cumulative drift error. ) principle, three IOAP algorithms based on 1σ-EPMP, 2σ-EPMP, and 3σ-EPMP mechanisms are constructed according to different σ principles, which are abbreviated as 1σ-IOAP, 2σ- IOAP, 3σ-IOAP. Among them, ξ=1, 2, 3 are used to represent the three mechanisms of 1σ-EPMP, 2σ-EPMP and 3σ-EPMP, that is, ξ=1, representing the 1σ-EPMP mechanism; ξ=2, representing the 2σ-EPMP mechanism; ξ =3, indicating the 3σ-EPMP mechanism.

Then the EPMP policy can be constructed in the following new way:

The total number of rings in the macrocyclic domain under the three mechanisms of 1σ-EPMP, 2σ-EPMP and 3σ-EPMP is denoted as

分别进行修正，得到修正后的三种机制下的大环域的总环数M₁′、M₂′、M₃′，以便于后续对大环域的内中外三环层的划分：Considering that the total number of loops in the large loop is not necessarily a multiple of 3 under different σ principles, and at the same time to further enhance the good coverage effect of the EPMP strategy on the real position of the submersible by the large loop grid, the following formula can be used to calculate:

Correction is made respectively to obtain the total number of rings M ₁ ′, M ₂ ′ and M ₃ ′ of the macrocyclic domain under the revised three mechanisms, so as to facilitate the subsequent division of the inner, middle and outer three-ring layers of the macrocyclic domain:

Then, according to the total number of rings M ₁ ′, M ₂ ′ and M ₃ ′ of the macrocyclic domain under the revised three mechanisms, the radii R _1,j , R ₂ of each ring of the macrocyclic domain under the three mechanisms can be obtained _,j , R _3,j , and the intermediate ring radius of the macroring domain under the three mechanisms

R _ξ,ζ =ζC,ζ=1,2,...,M' _ξ ...(6)

where R _ξ,ζ denotes the radius of the ζth ring of the macrocyclic domain under the ξth mechanism.

Furthermore, the reference angles generated by each lattice point in the macrocyclic domain under the three mechanisms can be obtained

Finally, according to (5)-(8), the grid points to be matched in the macroring domain under different mechanisms are obtained. Among them, the calculation formula of the grid points to be matched in the large ring domain under each mechanism is as follows:

的计算式为：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the large ring domain under the 1σ-EPMP mechanism

The calculation formula is:

的计算式为：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the large ring domain under the 2σ-EPMP mechanism

The calculation formula is:

的的计算式为：The position coordinates of each grid point to be matched on the inner-middle-outer ring layer of the macroring domain under the 3σ-EPMP mechanism

The calculation formula is:

Step 103, iteratively calculate the matching performance evaluation index of the grid points to be matched in the large ring domain, and obtain the best matching position of the end point of the track of the underwater vehicle within the scope of the large ring domain according to the optimal principle, so as to realize the track of the underwater vehicle. Effective matching and positioning of the end point, and then correcting the INS system control parameters and assisting the completion of the long-duration and long-distance navigation goal of the underwater submersible.

In this embodiment, considering the high accuracy of the gravity value at the grid resolution in the gravity reference map, the gravity value calculated by the interpolation method may not truly reflect the actual gravity at the matching point. The matching process of the algorithm is to determine the best matching positioning of the end point of the underwater vehicle in the set of points to be matched in the EPMP large ring.

Preferably, a feasible way to iteratively calculate the matching efficiency evaluation index of the grid points to be matched in the large ring area, and obtain the best matching position of the end point of the track of the underwater vehicle in the large ring area according to the optimal principle is as follows:

Determine the total number N _ξ of grid points to be matched on the inner-middle-outer ring layer of the macroring domain under the ξth mechanism.

将点

在重力基准图中对应的位置(x_r,y_r)与重力基准图的格网分辨率C相比，并按四舍五入原则得到重力基准图上最近邻于点

的格网点base_L。将格网点base_L对应的重力值

作为点

重力值的近似；根据格网点base_L在重力基准图上的位置坐标、水下潜器航行的航速和航向，提取得到点

对应的重力图航迹序列

及对应的最近邻重力序列

将

与水下潜器实测重力值序列{g₁,g₂,…,g_L}进行对比，并计算匹配效能评价指标，记作MSD_r。其中，需要说明的是，在计算匹配效能评价指标时，可以选择任意一种适当效能评价指标进行计算匹配，本实施例中是以均方差MSD为例进行的说明。Take the r (r∈{1,2,…,N _ξ })th grid point to be matched in the large ring domain under the ξth mechanism as the end point of the track to be evaluated, denoted as a point

will point

The corresponding position (x _r , y _r ) in the gravity reference map is compared with the grid resolution C of the gravity reference map, and the nearest adjacent point on the gravity reference map is obtained according to the rounding principle.

The grid point base _L . Set the gravity value corresponding to the grid point base _L

as a point

The approximation of the gravity value; according to the position coordinates of the grid point base _L on the gravity reference map, the speed and course of the underwater vehicle, the points are extracted and obtained.

Corresponding gravity map track sequence

and the corresponding nearest neighbor gravity sequence

Will

Compare with the measured gravity value sequence {g ₁ ,g ₂ ,...,g _L } of the submersible, and calculate the matching efficiency evaluation index, denoted as MSD _r . It should be noted that, when calculating the matching efficiency evaluation index, any appropriate efficiency evaluation index may be selected for calculation and matching. In this embodiment, the mean square error MSD is used as an example for description.

According to the above calculation process of the rth grid point to be matched, the matching efficiency evaluation index of each grid point to be matched on the inner-middle-outer ring layer of the large ring domain under the ξth mechanism can be calculated in turn, and the ξth mechanism can be obtained. The matching performance evaluation index set of all grid points to be matched on the inner-middle-outer ring layer of the large ring domain under the mechanism is {MSD _r |r=1,2,...,N _ζ }.

Finally, based on {MSD _r |r=1,2,...,N _ζ }, the best matching position of the end point of the track of the underwater vehicle within the large ring domain is obtained by filtering according to the optimal principle.

In summary, through the above steps 101 to 103, the effective matching and positioning of the end point of the underwater vehicle is realized, and the best matching position of the end point of the track of the underwater vehicle is used as the end point of the track of the underwater vehicle to correct the INS system. Control parameters and assist in completing the long-duration and long-range navigation goals of the underwater submersible.

On the basis of the above embodiment, the method for improving the underwater navigation accuracy based on the iterative optimal loop domain point based on the grid topology structure described in the above embodiment is verified below.

In order to verify the effectiveness and superiority of the IOAP algorithm in the application of underwater vehicle gravity navigation, a total of 3 sets of tests are designed:

Test 1, to verify the matching performance difference of the IOAP algorithm under different σ criteria;

Test 2, verify the different influences on the algorithm matching performance with reference angles under different ring radii;

In Test 3, the good applicability of the proposed IOAP algorithm to underwater gravity matching navigation is verified with the starting points of different regional tracks.

The example data is derived from the website of the University of California, San Diego (http://topex.ucsd.edu/), and its resolution is 1′×1′ gravity anomaly data. As shown in Fig. 4(a), the present invention selects the gravity anomaly data in the South China Sea for research, and the longitude and latitude of the data ranges from (113°E–115°E longitude, 10°N–12°N latitude). The present invention converts the gravity anomaly reference data into 100m×100m grid resolution gravity data through the bilinear interpolation method, as shown in Figure 4(b), the maximum gravity anomaly in this area is 130.57mGal, and the minimum value is -33.53 mGal, with an average of 15.43 mGal.

(1) The matching performance difference verification of IOAP algorithm under different σ criteria

同时记录N次测试匹配定位精度的平均值(mean)、标准差(std)和最差值(max)以及航迹平均匹配时间T(不含环境配置时间)作为重力匹配算法性能评价指标。The resolution of the gravity anomaly grid in the simulated sample block is 100m×100m, the accelerometer constant zero bias is 10 ^-3 m/s ² (the root mean square error of the inertial navigation follows a normal distribution), the speed is 10m/s ¹ , and the heading north 70° east, initial position error 0m, velocity error 0.04m/s ¹ , heading error 0.05°, the real-time measurement data of the gravimeter is the sampling value of the real track in the gravity anomaly database superimposed with random noise with a standard deviation of 1mGal. The number of points is 110, and the sampling period is 20s. Among them, the present invention defines the matching positioning accuracy as 1, then the difference between the absolute value of the matching position and the real position is an effective matching within the closed interval [0, 1], so the effective matching times n of the algorithm under N experimental tests can be obtained. and the matching success rate of the algorithm

At the same time, the mean (mean), standard deviation (std) and worst value (max) of the matching positioning accuracy of N tests and the average track matching time T (excluding the environmental configuration time) were recorded as the performance evaluation indicators of the gravity matching algorithm.

In order to test and analyze the application performance of IOAP algorithms with different σ criteria in gravity matching navigation of underwater vehicles, 100 independent experiments were conducted with 1σ-IOAP, 2σ-IOAP and 3σ-IOAP algorithms, and the TERCOM algorithm was used as a comparison algorithm. The grid point coordinates (1400, 1500) of the gravity reference map are used as the simulation starting point of the underwater submersible navigation, and the comparison effect of intuitive matching and positioning accuracy is shown in Figure 5.

The matching and positioning performance of the IOAP algorithm under different σ effects is different, and the matching effect of the 3σ-IOAP algorithm is the best and significantly better than the traditional TERCOM algorithm. On the T index, although the average running time of the 1σ-IOAP algorithm is the smallest, the matching effect is poor and it is difficult to be effectively applied to the navigation of actual underwater vehicles; the average running time of the 2σ-IOAP algorithm is about half of that of the TERCOM algorithm, and the average matching accuracy is less than 1 grid resolution, and its matching probability is also better than that of the TERCOM algorithm (88%>82%) under the condition of positioning accuracy l=100, indicating that the 2σ-IOAP algorithm has certain matching efficiency and matching accuracy under the dual target condition. According to the actual navigation application value, the gravity navigation mechanism can be appropriately selected according to the actual navigation requirements; the 3σ-IOAP algorithm is not much different from the T index of TERCOM, and the mean value, std value and max value of the matching accuracy and the matching success probability are not different. It is better than most of the index values of TERCOM algorithm and other algorithms in terms of indicators, which fully shows that the proposed IOAP algorithm has better matching performance and good potential practical value in gravity-assisted navigation of underwater vehicles.

的约束条件下统计100次测试的成功匹配概率对比结果的直观对比柱状图，如图6所示。In order to further analyze the difference of the successful matching effect of the four algorithms under the constraint condition of different positioning accuracy L, take l as 20, 40, 60, 80, 100 and

The intuitive comparison histogram of the comparison results of the successful matching probability of the statistics for 100 tests under the constraint conditions is shown in Figure 6.

The successful matching probabilities of different σ-IOAP algorithms under different positioning accuracy constraints are significantly different, and the three σ-IOAP algorithms have a certain number of successful matching when l≤40, but all TERCOM matching fails; comprehensive analysis of different positioning accuracy l For the successful matching probability of the algorithm, the matching performance of the 3σ-IOAP algorithm is the best, followed by the 2σ-IOAP algorithm, which is better than the traditional TERCOM algorithm, but the 1σ-IOAP algorithm is weaker than the TERCOM algorithm under the constraint of positioning accuracy; Figure 6 shows it further intuitively. The excellent and successful matching performance of the 3σ-IOAP algorithm is shown, and the above conclusions still effectively verify the excellent performance of the proposed 3σ-IOAP algorithm in underwater gravity matching navigation.

In order to further explore and analyze the reasons why the 3σ-IOAP algorithm is superior to TERCOM and other σ-IOAP algorithms in terms of matching efficiency and accuracy, the scatter comparison between the matching position of the TERCOM algorithm and the actual position of the underwater vehicle is drawn as shown in Figure 7 (in the usual way. The guide position is used as the origin of the image coordinates to ensure that 100 test results can be drawn on the same image).

From the analysis in Figure 7(a), it can be seen that the actual positions of the underwater vehicles tested by the TERCOM algorithm for 100 times are almost all within the 3σ error grid range of the inertial navigation position (the solid line frame outside the dotted circle in Figure 7(a)), At the same time, all the position points are almost all located inside the dashed line of the 3σ ring boundary centered on the inertial navigation position (the dashed circle of the inner solid line frame in Figure 7(a)), which indicates to a certain extent that the rectangular grid of the TERCOM algorithm needs to be matched. There are a certain number of points to be matched with a small probability, as shown in Figure 7(b), the points between the solid line box and the dotted circle, that is, the points outside the 3σ circle have a small probability of being matched, but they obviously affect the matching efficiency of TERCOM. , therefore, the gravity matching algorithm that deletes the matching points outside the 3σ ring domain can effectively improve the matching efficiency without significantly affecting the matching accuracy of the algorithm. Feasibility and effectiveness.

In order to further analyze the reason why the 3σ-IOAP algorithm is superior to other σ-IOAP matching accuracy, a schematic diagram of the matching effect of different σ-IOAP algorithms for 100 tests is shown in Figure 8. From the analysis in Figure 8, it can be seen that the IOAP algorithms based on different σ ring domains have different coverage and matching effects on the real position of the underwater vehicle: the 3σ-IOAP algorithm realizes the high precision of the underwater vehicle with the grid coverage of the large ring domain. Matching positioning; while 2σ-IOAP and 1σ-IOAP only achieve good matching of real positions in the large ring domain, but the optimal matching of real positions outside the domain is often only scattered on its boundary ring, and it is difficult to find better matching positions. It affects the matching accuracy of the algorithm, and then explains the difference in the successful matching probability of different σ-IOAP algorithms and the good matching accuracy of the 3σ-IOAP algorithm to a certain extent. important potential application value.

(2) Differential verification of the impact of the matching performance of the IOAP algorithm under different ring radius reference angles

对3σ-IOAP算法匹配性能的影响效果，分别以3σ-IOAP算法的内环层最大环半径R¹＝R_M/3、“内中”环层中间环半径

中间环层最大环半径R²＝R_2M/3、“中外”环层中间环半径

来确定3σ-IOAP算法的基准角

并张成变角度三环层待匹配格点集，相应算法分别记为1-IOAP、1.5-IOAP(即2.1节的3σ-IOAP)、2-IOAP和2.5-IOAP算法。根据式(6)分析可知，环半径R越大则3σ-IOAP算法的大环域基准角

越小，生成的待匹配格网点总数目N越多，故理论上1-IOAP算法的匹配效率最快而2.5-IOAP算法的执行最慢。In order to further explore the reference angle based on different ring radii R

The effect on the matching performance of the 3σ-IOAP algorithm is based on the maximum ring radius R ¹ = _RM/3 of the inner ring layer of the 3σ-IOAP algorithm, and the middle ring radius of the “inner middle” ring layer.

The maximum ring radius of the middle ring layer R ² =R _2M/3 , the middle ring radius of the “intermediate” ring layer

to determine the reference angle of the 3σ-IOAP algorithm

And the grid points to be matched in the variable-angle three-ring layer are denoted as 1-IOAP, 1.5-IOAP (ie 3σ-IOAP in Section 2.1), 2-IOAP and 2.5-IOAP algorithms respectively. According to the analysis of formula (6), it can be seen that the larger the ring radius R is, the larger the reference angle of the large ring domain of the 3σ-IOAP algorithm is.

The smaller the value, the more the total number N of grid points to be matched will be generated. Therefore, in theory, the matching efficiency of the 1-IOAP algorithm is the fastest and the execution of the 2.5-IOAP algorithm is the slowest.

The parameter settings such as the simulated sample block and the grid coordinates of the track starting point are the same as above (1). The algorithm matching and positioning statistical results of 100 independent tests are shown in Figure 9 and Figure 10 respectively. The matching accuracy and success probability of the 3σ-IOAP algorithm almost show the improvement of the matching effect with the increase of the reference angle ring radius R, and the performance of 2.5-IOAP is the best, followed by 2-IOAP, and the matching effect of 1-IOAP is relatively better. It is worse but still better than the TERCTOM algorithm in 6/7 indicators, indicating that the reference angles based on different ring radii will affect the matching performance of the 3σ-IOAP algorithm. The increase of R shows the phenomenon of "decrease of matching efficiency", and this result is consistent with its theoretical analysis conclusion, indicating that the total number of grid points to be matched corresponding to different ring radius reference angles is different, which leads to the difference of matching efficiency of the algorithm. Therefore, in the actual underwater vehicle navigation application, the 3σ-IOAP algorithm corresponding to the appropriate ring radius R can be appropriately selected according to the specific matching target requirements, and the specific navigation matching task can be completed, illustrating the 3σ-IOAP proposed in the present invention. The algorithm has high algorithm robustness and good application value in underwater gravity matching navigation.

2.5- The matching accuracy and matching success probability of the IOAP algorithm at different scales are almost superior to other IOAP algorithms and TERCOM algorithms, but its matching T index value is the highest and almost twice the matching time of TERCOM, so the matching efficiency requirements are not high and Under the situation of underwater navigation that pays more attention to matching accuracy, the 2.5-IOAP algorithm is the best choice for the gravity matching algorithm; while in the situation where matching efficiency and matching accuracy are required for dual targets, the 1-IOAP algorithm is the best choice for the gravity matching algorithm. , therefore, it further proves that the 3σ-IOAP algorithm has good underwater gravity matching navigation robustness. In addition, the 1.5-IOAP algorithm (that is, the 3σ-IOAP algorithm in Section 2.1) shows a good compromise between matching efficiency and matching accuracy, and considering that its matching efficiency is not much different from the TERCOM algorithm, the next section is based on The 1.5-IOAP algorithm is further tested to verify its good matching performance under different starting points.

(3) Verification of good matching performance of IOAP algorithm under different regional track starting points

In order to further verify the good matching applicability of the 1.5-IOAP algorithm in the matching and navigation of underwater vehicles, the present invention uses the regional grid coordinates A (1660, 1410), B (1550, 740) and C (1400, 350) as The location of the start of the submersible track. On the premise that the matching efficiency (T index) is not much different, the 1.5-IOAP algorithm is significantly better than the average matching accuracy (mean), the standard deviation of the matching accuracy (std) and the worst matching accuracy (max). Traditional TERCOM algorithm; compared with TERCOM, the worst matching accuracy of 1.5-IOAP in the three scenarios is relatively improved by 47.24%, 63.96% and 72.16%, while the average matching accuracy is relatively improved by 20.37%, 40.39% and 13.88%, It shows that the IOAP algorithm proposed in the present invention has high matching accuracy and good synchronous optimization performance in multiple tests; at the same time, under different successful matching scales l, the algorithm of the present invention shows relatively better matching success than the TERCOM algorithm. Probability, especially when the matching accuracy is less than 40, the 1.5-IOAP algorithm still has a certain probability to successfully match the position of the underwater submersible, which effectively verifies the good matching applicability of the proposed IOAP algorithm for underwater gravity matching navigation under the conditions of different starting points. .

In order to visually show the difference in matching effect between the 1.5-IOAP algorithm and the TERCOM algorithm under three test positions, a comparison diagram of the worst matching positioning in 100 tests is drawn, as shown in Figure 11. From the analysis in Figure 11, it can be seen that in the underwater gravity matching positioning under the starting points of the tracks in different regions, the 1.5-IOAP algorithm has higher matching positioning accuracy than the TERCOM algorithm, and the end points of the three tracks fall in different gravity intervals. To a certain extent, it effectively verifies that the proposed IOAP algorithm has good matching adaptability in different gravity sections, and further effectively proves that the iterative optimal ring domain point algorithm based on the new grid topology structure of the present invention can improve the water quality. Validity and reliability of submersible gravity matching accuracy.

On the basis of the above-mentioned embodiment, the present invention also discloses a system for improving underwater navigation accuracy based on iterative optimal ring domain points based on grid topology structure, comprising: a solving module for passing a small ring domain grid at the track starting point Match the positioning strategy to obtain the best matching position of the starting point of the track of the underwater vehicle; the generation module is used to match the positioning strategy through the three-layer ring domain of the track end point according to the best matching position of the starting point of the track of the underwater vehicle. , generate the grid points to be matched in the large ring domain; the iterative determination module is used to iteratively calculate the matching efficiency evaluation index of the grid points to be matched in the large ring area, and obtain the end point of the track of the underwater vehicle within the large ring area according to the optimal principle. The best matching position is to realize the effective matching and positioning of the end point of the track of the underwater submersible, and then correct the control parameters of the INS system and assist in the completion of the long-duration and long-distance navigation of the underwater submersible.

As for the system embodiment, since it corresponds to the method embodiment, the description is relatively simple, and for related parts, please refer to the description of the method embodiment part.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can use the methods and technical contents disclosed above to improve the present invention without departing from the spirit and scope of the present invention. The technical solutions are subject to possible changes and modifications. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solutions of the present invention belong to the technical solutions of the present invention. protected range.

Contents that are not described in detail in the specification of the present invention belong to the well-known technology of those skilled in the art.

Claims (10)

1. A method for improving underwater navigation precision based on a grid topological structure iteration optimal ring domain point is characterized by comprising the following steps:

obtaining the optimal matching position of the track starting point of the underwater vehicle through a track starting point small-ring area grid matching positioning strategy;

generating lattice points to be matched in a large ring area by a track end point angle-variable three-layer ring area matching positioning strategy according to the optimal matching position of the track starting point of the underwater vehicle;

and (3) iteratively calculating the matching efficiency evaluation index of the grid points to be matched in the large ring area, and obtaining the optimal matching position of the underwater vehicle track terminal in the large ring area range according to the optimal principle so as to realize the effective matching and positioning of the underwater vehicle track terminal, further correcting the INS system control parameters and assisting in finishing the navigation target of the underwater vehicle with long voyage and long voyage distance.

2. The method for improving underwater navigation accuracy based on the iterative optimal ring domain points of the grid topology structure as claimed in claim 1, wherein the construction process of the track starting point small ring domain grid matching positioning strategy is as follows:

obtaining INS track output inertia of a certain section of underwater vehicle navigation according to sampling time interval delta tLeader sequence S₁,S₂,…,S_L}; where L represents the sample sequence length of the INS flight path, S₁、S₂、…、S_LRepresenting each track point in the inertial navigation sequence;

from { S₁,S₂,…,S_LExtracting to obtain position coordinates and actually measured gravity values corresponding to each track point, and constructing to obtain a position coordinate sequence { (x)₁,y₁),(x₂,y₂),…,(x_L,y_L) The measured gravity value sequence of the underwater vehicle and the g₁,g₂,…,g_L}；

Starting point S with track in inertial navigation sequence₁Position coordinates (x)₁,y₁) Centered at 3 σ₀Constructing a small ring domain as the maximum extended search boundary radius; wherein σ₀Representing the standard deviation of inertial navigation drift errors when the sampling interval is delta t;

according to the north east rotation angle theta₀And determining the total number and position coordinates of grid points to be matched in the small ring domain by using a ring radius proportional average coefficient lambda, and constructing to obtain a grid point set to be matched in the small ring domain.

3. The method for improving underwater navigation accuracy based on iterative optimal ring domain points of grid topology structure as claimed in claim 1, wherein the position coordinates of each grid point to be matched in the small ring domain

The calculation formula of (a) is as follows:

wherein,

and

respectively represents the horizontal and vertical coordinates r of the jth lattice point on the ith small ring_iDenotes the radius, β, of the ith ring in the small ring_jThe rotation angle of the jth grid point on each ring of the small ring domain is shown.

4. The mesh topology iterative optimal ring domain point-based method for improving accuracy of underwater navigation of claim 3,

r_i＝3σ₀λi

β_j＝j·θ₀

wherein i is 1,2, …, and the maximum value of i

j is 1,2, …, and the maximum value of j

θ₀∈(0,360]；λ∈(0,1]。

5. The method for improving the accuracy of underwater navigation based on the iterative optimal ring area points of the grid topological structure as claimed in claim 4, wherein the optimal matching position of the track starting point of the underwater vehicle is obtained by a track starting point small ring area grid matching positioning strategy, and the method comprises the following steps:

starting the track S₁Position coordinates (x)₁,y₁) And the position coordinates of each lattice point to be matched in the small ring area

A point set to be matched is used as the true position of the starting point of the flight path;

mapping a point set to be matched of the true position of the track starting point to a gravity reference graph point by point, and taking the gravity value at the nearest gravity reference grid point as the gravity value of the point to be matched to obtain (x)₁,y₁) Corresponding theoretical gravity value

And

corresponding theoretical gravity value

Wherein, C represents the grid resolution of the gravity reference map, mapt (·,) represents the gravity value matrix of the gravity reference map according to the grid point position, [ · ] represents rounding;

obtaining the best matching position of the track starting point of the underwater vehicle according to the principle of minimizing the absolute value of the gravity deviation

Wherein when i is 0 and j is 0,

is x₁、

Is y₁。

6. The method for improving underwater navigation accuracy based on the optimal iterative loop points of the grid topology structure as claimed in claim 5, wherein the track end point angle-variable three-layer loop matching positioning strategy is constructed as follows:

according to a track starting point S in an inertial navigation sequence₁Position coordinates (x)₁,y₁) And track end point S_LPosition coordinates of(x_L,y_L) Determining course information and range information of the underwater vehicle navigation:

wherein d is_INSRepresenting that the distance measurement between the start point and the end point of the inertial navigation track is described by h epsilon [0, + ∞) norm, and calculating the Euclidean distance by setting h to be 2; alpha is alpha_INSRepresenting a course arc angle taking the abscissa of the coordinate system as a positive direction;

according to

And estimating to obtain the central position coordinate (x) of the large ring area according to the course information and the range information of the underwater vehicle navigation_O,y_O)：

With (x)_O,y_O) Centered on R_maxConstructing and obtaining an angle-variable three-layer topological structure annular grid point region covering a true end point of a flight path as a search boundary radius of the maximum extension, and recording the region as a large annular region;

taking the grid resolution C of the gravity reference graph as the span interval of each ring of the large ring area to obtain the total ring number of the large ring area

The grid resolution C and the middle ring of the gravity reference map

Radius ofReference angle generated as each lattice point in large domain

According to the principle of 'inner times and outer halves', determining the span angle between adjacent lattice points on the inner ring of the large ring area to be

The span angle between adjacent lattice points on the outer ring is

Constructing and obtaining a large-ring domain grid point set to be matched by combining the total ring number M; wherein, the large-ring domain grid point set to be matched is a variable-angle three-ring layer grid point topological structure.

7. The method for improving underwater navigation accuracy based on the iterative optimal ring domain points of the grid topology structure as claimed in claim 6, wherein the generation mode of the grid points to be matched in the large ring domain is as follows:

the total number of the large-loop domain under the three mechanisms of 1 sigma-EPMP, 2 sigma-EPMP and 3 sigma-EPMP is respectively recorded as

Wherein, sigma represents the standard deviation of inertial navigation accumulated drift errors;

to pair

Respectively corrected to obtain the total ring number M 'of the large ring domain in the three corrected mechanisms'₁、M′₂、M′₃：

Wherein ξ ═ 1,2 and 3 respectively correspond to a 1 σ -EPMP mechanism, a 2 σ -EPMP mechanism and a 3 σ -EPMP mechanism;

according to the corrected three mechanismsTotal Ring number M 'of Large Ring Domain'₁、M′₂、M′₃Obtaining the radius R of each ring of the large ring domain under three mechanisms_1,j、R_2,j、R_3,j：

R_ξ,ζ＝ζC，ζ＝1,2,…,M′_ξ…(6)

Wherein R is_ξ,ζA radius of a zeta-th ring representing a large ring domain under a zeta-th mechanism;

according to the total ring number M 'of the large ring domain under the three corrected mechanisms'₁、M′₂、M′₃Obtaining the intermediate ring radius of the large ring domain under three mechanisms

Intermediate ring radius of large ring domain according to three mechanisms

Obtaining the reference angle generated by each lattice point in the large domain under three mechanisms

And (5) obtaining the lattice points to be matched of the large ring area under different mechanisms according to the steps (5) to (8).

8. The mesh topology iterative optimal ring domain point-based method for improving accuracy of underwater navigation of claim 7,

position coordinates of lattice points to be matched on inner-middle-outer ring layer of large ring domain under 1 sigma-EPMP mechanism

The calculation formula of (a) is as follows:

position coordinates of lattice points to be matched on inner-middle-outer ring layer of large ring domain under 2 sigma-EPMP mechanism

The calculation formula of (a) is as follows:

position coordinates of lattice points to be matched on inner-middle-outer ring layer of large ring domain under 3 sigma-EPMP mechanism

The calculation formula of (a) is as follows:

9. the method for improving underwater navigation accuracy based on the iterative optimal ring domain points of the grid topological structure as claimed in claim 8, wherein the step of iteratively calculating the matching efficiency evaluation index of the grid points to be matched in the large ring domain and obtaining the optimal matching position of the track end point of the underwater vehicle in the large ring domain according to the optimal principle comprises the following steps:

determining the total number N of lattice points to be matched on an inner-middle-outer ring layer of a large ring domain under the xi mechanism_ξ；

Taking the r-th grid point to be matched of the large ring domain under the xi mechanism as the end point of the flight path to be evaluated and recording the r-th grid point as a point

Wherein r ∈ {1,2, …, N_ξ}；

Will be dotted

Corresponding position (x) in the gravity reference map_r,y_r) Comparing with the grid resolution C of the gravity reference diagram, and obtaining the nearest neighbor point on the gravity reference diagram according to the rounding principle

Grid point base_L；

Base the grid points_LCorresponding gravity value

As a point

An approximation of the value of gravity;

according to the grid point base_LExtracting the position coordinates on the gravity reference map, the navigation speed and the navigation course of the underwater vehicle to obtain points

Corresponding gravity map track sequence

And corresponding nearest neighbor gravity sequence

Will be provided with

Sequence of gravity values { g actually measured with underwater vehicle₁,g₂,…,g_LComparing, calculating the evaluation index of matching efficiency, and recording as MSD_r；

Sequentially calculating to obtain the matching efficiency evaluation indexes of all lattice points to be matched on the inner-middle-outer ring layer of the large ring area under the xi mechanism, and obtaining the matching efficiency evaluation index set and { MSD (maximum data set) of all lattice points to be matched on the inner-middle-outer ring layer of the large ring area under the xi mechanism_r|r＝1,2,…,N_ζ}；

Based on { MSD_r|r＝1,2,…,N_ζAnd (6) screening according to an optimal principle to obtain an optimal matching position of the track terminal of the underwater vehicle in the large-ring-area range

10. A system for improving underwater navigation precision based on a grid topological structure iteration optimal ring domain point is characterized by comprising:

the resolving module is used for obtaining the optimal matching position of the track starting point of the underwater vehicle through a track starting point small-ring area grid matching positioning strategy;

the generating module is used for generating lattice points to be matched in a large ring area through a track end point angle-variable three-layer ring area matching positioning strategy according to the optimal matching position of the track starting point of the underwater vehicle;

and the iteration determination module is used for iteratively calculating the matching efficiency evaluation index of the lattice points to be matched in the large ring area, and obtaining the optimal matching position of the underwater vehicle track terminal in the large ring area according to the optimal principle so as to realize the effective matching and positioning of the underwater vehicle track terminal, further correct the control parameters of the INS system and assist in finishing the long-endurance long-range navigation target of the underwater vehicle.

Images