An adaptive multi-beam survey line layout method based on kriging interpolation prediction

Abstract

The application discloses a kind of self-adapting multi-beam survey line layout method based on kriging interpolation prediction, belong to underwater surveying and mapping technical field, for multi-beam survey line layout, including using extreme value weighted depth model to predict and layout based on strip coverage ratio survey line, according to the measured water depth abrupt change situation adaptive adjustment extreme value weighting weight, using kriging interpolation model based on measured water depth water area to unmeasured water depth water area carries out water depth space interpolation prediction, according to the terrain difference of water area, switch variogram model to optimize interpolation prediction result, based on strip coverage ratio and the water depth space of interpolation prediction, the spatial position of next survey line is planned, according to the boundary of to-be-measured deep water area, the two end points of new survey line are determined, and complete survey line is planned.The application can effectively capture local water depth variation characteristics, and then provide reliable basis for subsequent survey line strip width estimation and interval control, and automatically generate multi-beam survey line with high coverage and continuity.

Description

TECHNICAL FIELD

[0001] The application discloses an adaptive multi-beam survey line layout method based on Kriging interpolation prediction and belongs to the technical field of underwater surveying and mapping. BACKGROUND

[0002] In the fields of underwater topographic mapping, water conservancy projects, channel dredging and environmental investigation, high-precision water depth data acquisition is a basic work. With the wide application of unmanned surveying ships and multi-beam sounding systems, the sounding efficiency and precision have been significantly improved. However, when sounding in unknown or complex waters, how to automatically plan survey lines to achieve high coverage rate, low redundancy and high measurement efficiency is still a key technical problem. Traditional survey line planning mostly adopts a regular layout strategy and relies on the prior known survey area boundary and average water depth, and lacks the adaptive adjustment capability for actual terrain features, which often leads to insufficient strip rate coverage, excessive survey line overlap, and inability to adapt to irregular water area boundaries and complex bottom shapes. SUMMARY

[0003] The application aims to provide an adaptive multi-beam survey line layout method based on Kriging interpolation prediction, so as to solve the problems of insufficient strip rate coverage, excessive survey line overlap, and inability to adapt to irregular water area boundaries and complex bottom shapes in the prior art.

[0004] An adaptive multi-beam survey line layout method based on Kriging interpolation prediction comprises the following steps:

[0005] S1. According to the range of the water area to be sounded, the convex hull boundary of the water area is extracted to ensure that the laid survey lines cover the entire water area to be sounded.

[0006] S2. The multi-beam opening angle value and the expected strip coverage rate are set.

[0007] S3. The main survey line direction is determined.

[0008] S4. The survey line based on the strip coverage rate is predicted and laid out using an extreme value weighted water depth model, and the extreme value weighting weight is adaptively adjusted according to the water depth mutation of the surveyed water area.

[0009] S5. The water depth space of the unsurveyed water area is predicted by using a Kriging interpolation model based on the surveyed water depth, and the variogram model is switched according to the water area terrain difference to optimize the interpolation prediction result.

[0010] S6. The spatial position of the next survey line is planned based on the water depth space of the strip coverage rate and the interpolation prediction, and the two end points of the new survey line are determined according to the boundary of the water area to be sounded.

[0011] S7. The complete survey line is planned, which comprises the main survey line, the Z-shaped connecting survey line between adjacent main survey lines and the check survey line.

[0012] S1 includes S1.1, using a multi-beam carrying platform to perform obstacle avoidance and following in a to-be-measured deep water area, moving the depth measurement along the natural boundary of the water area, obtaining a set of boundary points of the to-be-measured water area, and forming a range of the to-be-measured deep water area;

[0013] S1.2, projecting the set of boundary points of the to-be-measured water area in two dimensions, extracting a convex hull vertex based on a two-dimensional convex hull algorithm Quickhull, screening a convex hull boundary of the set of boundary points of the to-be-measured water area, and outputting the coordinates of the convex hull vertex arranged in a counterclockwise order, the convex hull boundary is:

[0014] ;

[0015] In the formula, is the m-th convex hull coordinate, is the number of convex hull coordinates;

[0016] S1.3, calling a convex hull in-point judgment function to judge whether a point in the point set is within the boundary of the to-be-measured water area:

[0017] ;

[0018] In the formula, is a set of to-be-measured water area boundaries, the point of is removed from the survey line, is the coordinate of the point in the to-be-measured water area.

[0019] The multi-beam opening angle value is 120°, and the expected strip coverage rate is 25%;

[0020] The multi-beam strip width is:

[0021] ;

[0022] In the formula, is the water depth of the current survey line layout area, is the multi-beam opening angle.

[0023] S3 includes approximately fitting a quadrilateral to the survey area based on the convex hull shape of the water area boundary, selecting the long side direction of the fitted quadrilateral as the main survey line direction, the ship width distance deviating from the main survey line direction corresponding to the long side as the initial survey line position, and taking the intersection point of the initial survey line and the water area boundary as the survey line starting and ending points.

[0024] The extreme value weighted depth model includes introducing a definition of a weighting coefficient :

[0025] ;

[0026] ;​

[0027] wherein, is the weight factor of extreme outliers, is the weight factor of mild outliers, is the first quartile, is the third quartile, is the interquartile range, is the water depth of the i-th known point;

[0028] extreme value weighted water depth is:

[0029] ;

[0030] wherein, is the total number of known points.

[0031] S5 comprises constructing a polygonal strip area for the known points, the polygonal strip area constituting a water depth point set is:

[0032] ;

[0033] wherein, is the spatial coordinate of the i-th known point, is the water depth of the i-th known point;

[0034] Based on the polygonal strip area, a multi-span extrapolation sampling prediction is performed to calculate the predicted water depth value of the unknown point :

[0035] ;

[0036] ;

[0037] wherein, is the weight of the i-th known point; The variogram function of Kriging interpolation and the covariance relationship are:

[0038]

[0039] ; wherein,

[0040] is the Kriging interpolation variogram function, is the maximum covariance, is the covariance function, is the spatial distance between two points, which is minimized when solving the weight , ​​​​It is the water depth at an unknown point;

[0041] The weights are to be determined as follows:

[0042] ;

[0043] In the formula, It is the first The known point and the first The variogram values ​​between known points Is the prediction point and the first The variogram values ​​between known points It is a Lagrange multiplier. It is the first The weights of the known points.

[0044] S6 includes obtaining the predicted water depth value of the unknown point, calculating the water depth of the adjacent area, and determining the spacing between the next survey line and the previous survey line based on the multibeam strip width and strip coverage.

[0045] After determining the spatial location of the next survey line, the next survey line is truncated according to the convex hull boundary conditions of the water area to be measured. The part of the next survey line that intersects with the water area is retained, and the two intersection points of the truncated next survey line on the convex hull boundary are set as the starting point and ending point of the next survey line.

[0046] S7 includes the following: when generating the main survey line, if the midpoint of the main survey line exceeds the range of the deep water area to be measured, only the valid part of the main survey line within the range of the deep water area to be measured is retained, and the subsequent generation and prediction of the main survey line is terminated.

[0047] A Z-shaped connection strategy is introduced during the generation of the main survey lines. Z-shaped survey lines are dynamically added between adjacent main survey lines, so that the main survey lines are connected end to end to form a continuous path structure.

[0048] Record the main survey line and generate inspection survey lines according to the proportion of the total length of the main survey line.

[0049] Compared with existing technologies, the present invention has the following advantages: The present invention can effectively capture local water depth variation characteristics, thereby providing a reliable basis for subsequent measurement line strip width estimation and spacing control, and automatically generating high coverage and continuous multibeam measurement lines; it effectively suppresses the interference of abnormal water depth on strip width calculation and predicts the water depth of unmeasured areas in real time, overcoming the problems of low deployment accuracy and poor adaptability of traditional methods; combined with adaptive strip coverage strategy and connection measurement line planning, it enhances the path planning accuracy and operation efficiency of unmanned vessels in unknown waters. Attached Figure Description

[0050] Figure 1 It is a linear model predictive planning survey map;

[0051] Figure 2 is a spherical model prediction planning line graph;

[0052] Figure 3 is a spherical model strip coverage graph;

[0053] Figure 4 is an exponential model prediction planning line graph;

[0054] Figure 5 is an exponential model strip coverage graph;

[0055] Figure 6 is a Gaussian model prediction planning line graph. DETAILED DESCRIPTION

[0056] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0057] An adaptive multi-beam line layout method based on Kriging interpolation prediction, comprising:

[0058] S1. According to the range of the deep water area to be measured, the convex hull boundary of the water area is extracted to ensure that the laid line covers the entire deep water area to be measured;

[0059] S2. Set the multi-beam opening angle value and the expected strip coverage rate;

[0060] S3. Determine the main line direction;

[0061] S4. Use the extreme value weighted water depth model to predict and layout the line based on the strip coverage rate, and adaptively adjust the extreme value weighting weight according to the measured water area water depth mutation;

[0062] S5. Use the Kriging interpolation model to perform spatial interpolation prediction on the unmeasured water depth area based on the measured water depth area, and switch the variogram model according to the water area topographic difference to optimize the interpolation prediction result;

[0063] S6. Based on the spatial water depth of the strip coverage rate and the interpolation prediction, the spatial position of the next line is planned, and the two end points of the new line are determined according to the boundary of the deep water area to be measured;

[0064] S7. Plan the complete line, which includes the main line, the Z-shaped connecting line between adjacent main lines, and the inspection line.

[0065] S1 includes S1.1, using a multi-beam carrying platform to perform obstacle avoidance and following in a to-be-measured deep water area, moving the depth measurement along the natural boundary of the water area, obtaining a boundary point set of the to-be-measured water area, and forming a to-be-measured deep water area range;

[0066] S1.2, two-dimensional projection is performed on the boundary point set of the to-be-measured water area, and a convex hull vertex is extracted based on a two-dimensional convex hull algorithm Quickhull, a convex hull boundary of the boundary point set of the to-be-measured water area is screened out, and a convex hull vertex coordinate arranged in a counterclockwise order is output.

[0067] ;

[0068] In the formula, is the mth convex hull coordinate, is the number of convex hull coordinates;

[0069] S1.3, a convex hull in-point judgment function is called to judge whether a point in the point set is in the to-be-measured water area boundary:

[0070] ;

[0071] In the formula, is the to-be-measured water area boundary set, and a point of the to-be-measured water area boundary set is removed from the survey line, is the coordinate of the point in the to-be-measured water area. The multi-beam opening angle value is 120°, and the expected strip coverage rate is 25%.

[0072] The multi-beam strip width

[0073] is:

[0074] ;

[0075] In the formula, is the water depth of the current survey line layout area, is the multi-beam opening angle.

[0076] S3 includes approximately quadrilateral fitting of the survey area based on the convex hull shape of the water area boundary, selection of the long side direction of the fitted quadrilateral as the main survey line direction, selection of the ship width distance deviating from the main survey line direction as the initial survey line position, and selection of the intersection point of the initial survey line and the water area boundary as the survey line starting and ending endpoints.

[0077] The extreme value weighted depth model includes introducing a definition weighted coefficient :

[0078] ;

[0079] ; ​​​

[0080] wherein, is the weight factor of extreme outliers, is the weight factor of mild outliers, is the first quartile, is the third quartile, is the interquartile range, is the water depth of the th known point;

[0081] extreme value weighted water depth is:

[0082] ;

[0083] wherein, is the total number of known points.

[0084] S5 comprises constructing a polygonal strip area for the known points, the polygonal strip area constituting a water depth point set is:

[0085] ;

[0086] wherein, is the spatial coordinate of the th known point, is the water depth of the th known point;

[0087] Based on the polygonal strip area, a multi-span extrapolation sampling prediction is performed to calculate the predicted water depth value of the unknown point :

[0088] ;

[0089] ;

[0090] wherein, is the weight of the th known point;

[0091] The relationship between the Kriging interpolation variogram function and the covariance is:

[0092] ;

[0093] wherein, is the Kriging interpolation variogram function, is the maximum covariance, is the covariance function, is the spatial distance between two points, which is minimized when solving the weight , is the water depth of the unknown point;

[0094] Solving the weight is:

[0095]

[0096] In the formula, is the variogram value between the first known point and the first known point, is the variogram value between the prediction point and the first known point, is the Lagrange multiplier, is the weight of the first known point.

[0097] S6 includes that after obtaining the predicted water depth value of the unknown point, the water depth of the adjacent area is calculated, and the spacing between the next measuring line and the previous measuring line is determined according to the strip width and the strip coverage of the multi-beam.

[0098] After determining the spatial position of the next measuring line, the next measuring line is intercepted according to the convex boundary condition of the water area to be measured, the part of the next measuring line intersecting the water area range is reserved, and the two intersection points of the next measuring line on the convex boundary after interception are set as the starting point and the ending point of the next measuring line.

[0099] S7 includes that when the main measuring line is generated, if the midpoint of the main measuring line is out of the range of the deep water area to be measured, only the effective part of the main measuring line in the range of the deep water area to be measured is reserved, and the subsequent main measuring line is terminated.

[0100] In the process of generating the main measuring line, a zigzag connection strategy is introduced, and a zigzag measuring line is dynamically added between adjacent main measuring lines, so that the main measuring lines are connected end to end to form a continuous path structure.

[0101] The main measuring line is recorded, and the inspection measuring line is generated according to the total length proportion of the main measuring line.

[0102] In the embodiment of the application, a plurality of Kriging interpolation variogram functions are used, including Linear linear model, Spherical spherical model, Exponential exponential model and Gaussian Gaussian model. In the weight calculation, the three values of refer to Table 1.

[0103] Table 1, three values

[0104]

[0105] The linear model is used for predicting the planning measuring line as shown in Figure 1 , and the spherical model is used for predicting the planning measuring line as shown in Figure 2 ​The spherical model strip coverage map is shown as Figure 3 The exponential model prediction planning survey line is shown as Figure 4 The exponential model strip coverage map is shown as Figure 5 The Gaussian model prediction planning survey line is shown as Figure 6 According to the actual underwater topographic features of the survey area, the Kriging interpolation variation function is selected and adjusted to achieve more reasonable fitting and prediction of the water depth distribution under different topographic conditions, as shown in Table 2.

[0106] Table 2, selection of Kriging interpolation variation function

[0107] .

[0108] The above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that the technical solutions recorded in the foregoing examples can still be modified, or some or all of the technical features can be replaced by equivalents, without changing the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An adaptive multi-beam survey line layout method based on Kriging interpolation prediction, characterized in that, Comprise: S1, according to the range of deep water area to be measured, extract the water area convex hull boundary, ensure that the layout of the survey line covers the entire deep water area to be measured; S2, set the multi-beam opening angle value and the expected strip coverage; S3, determine the main survey line direction; S4, use the extreme value weighted depth model to predict and layout the survey line based on strip coverage, and adaptively adjust the extreme value weighting weight according to the measured water area depth mutation; S5, use the Kriging interpolation model to predict the depth of the unmeasured water area based on the measured water depth, and switch the variogram function model according to the water area topographic difference to optimize the interpolation prediction result; S6, based on the strip coverage and the water depth space of the interpolation prediction, plan the spatial position of the next survey line, and determine the two end points of the new survey line according to the boundary of the deep water area to be measured; S7, plan the complete survey line, which includes the main survey line, the Z-shaped connecting survey line between adjacent main survey lines and the check survey line; The extreme value weighted water depth model includes introducing a definition of a weighting coefficient : ； ； wherein is a weight factor for extreme outliers, is a weight factor for mild outliers, is the first quartile, is the third quartile, is the interquartile range, is the depth of the th known point. Extreme value weighted water depth is: ； wherein is the total number of known points; S5 comprises constructing a polygonal strip area from the known points, the polygonal strip area constituting a set of water depth points is: ； wherein is the spatial coordinate of the th known point, is the water depth of the th known point; Based on the polygon belt area, multi-span extension sampling prediction is performed to calculate the predicted water depth value of the unknown point :​ ； ； In the formula, For the first The weights of the known points; The relationship between the Kriging interpolation variogram function and the covariance is: ； where, is the kriging interpolation variogram, is the maximum covariance, is the covariance function, is the spatial distance between two points, minimizing the prediction variance when solving the weights , is the water depth at the unknown point; The weight is solved as: ； wherein is the variogram value between the th known point and the th known point, is the variogram value between the prediction point and the th known point, is the Lagrange multiplier, is the weight of the th known point.

2. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 1, characterized in that, S1 includes, S1.1, using a multi-beam carrying platform to execute obstacle avoidance and follow-up in the deep water area to be measured, moving along the natural boundary of the water area to obtain the boundary point set of the deep water area to be measured, and forming the range of the deep water area to be measured; S1.2, project the to-be-tested water area boundary point set in two dimensions, extract the convex hull vertex based on the two-dimensional convex hull algorithm Quickhull, screen out the convex hull boundary of the to-be-tested water area boundary point set, output the convex hull vertex coordinates arranged in counterclockwise order, the convex hull boundary is: ； wherein is the th convex hull coordinate, is the number of convex hull coordinates. S1.3, call the convex hull inner point judgment function to judge whether the points in the point set are within the boundary of the deep water area to be measured: ； In the formula, is a set of water area boundaries to be measured, and is a set of points to be measured in the water area. is the coordinates of the points to be measured in the water area.

3. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 2, characterized in that, The multi-beam opening angle value is 120°, and the expected strip coverage is 25%; Multi-beam strip width Is: ； In the formula, is the water depth of the current survey line layout area, is the multi-beam opening angle.

4. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 3, characterized in that, S3 includes, based on the convex hull shape of the water area boundary, approximate quadrilateral fitting is performed on the survey area, the long side direction of the fitted quadrilateral is selected as the main survey line direction, the ship width distance deviating from the main survey line direction is the initial survey line position, and the intersection point of the initial survey line and the water area boundary is taken as the starting and ending end points of the survey line.

5. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 4, characterized in that, S6 includes, after obtaining the predicted water depth value of the unknown point, the water depth of the adjacent area is calculated, and according to the multi-beam strip width and the strip coverage, the distance between the next survey line and the previous survey line is determined; After determining the spatial position of the next survey line, according to the convex hull boundary condition of the deep water area to be measured, the next survey line is intercepted, the part of the next survey line intersecting with the water area range is reserved, and the two intersection points of the next survey line on the convex hull boundary are set as the starting point and the ending point of the next survey line.

6. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 5, characterized in that, S7 includes, when generating the main survey line, if the midpoint of the main survey line is out of the range of the deep water area to be measured, only the effective part of the main survey line within the range of the deep water area to be measured is reserved, and the prediction of the subsequent main survey line is terminated.

7. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 6, characterized in that, In the process of generating the main survey line, a Z-shaped connection strategy is introduced, and a Z-shaped survey line is dynamically added between adjacent main survey lines, so that each main survey line is connected end to end to form a continuous path structure.

8. The adaptive multi-beam survey line layout method based on Kriging interpolation prediction according to claim 7, characterized in that, Record the main survey line and generate the check survey line according to the total length proportion of the main survey line.

Images