Self-adaptive fusion method for ocean multi-source gravity data

Abstract

The invention relates to the field of positioning data processing methods, in particular to a self-adaptive fusion method for ocean multi-source gravity data, which comprises the following steps: step 1, arranging obtained gravity data of a target area; step 2, carrying out mass classification on the ship measurement gravity data; step 3, determining weight coefficients of the ship measurement gravity data of different mass categories for correcting the satellite height measurement gravity data, circulating the weight coefficients of different mass categories according to constraint conditions, and stopping circulating when an end condition is reached; 4, correcting and fusing the satellite altimetry gravity data according to a preset rule by using the ship-measured gravity data and the weight coefficient; and step 5, calculating and correcting a root-mean-square error between the gravity anomaly reference map obtained by fusion and the high-precision ship measurement gravity data which does not participate in the fusion process, taking a weight coefficient when the root-mean-square error is minimum as a fusion parameter combination in the step 3, and finally obtaining a high-fineness gravity reference map. According to the invention, the accuracy of underwater navigation can be improved.

Description

Technical Field

[0001] The present invention relates to the field of positioning data processing methods, and in particular to an adaptive fusion method for ocean multi-source gravity data. Background Art

[0002] In the ocean, the primary technical means of underwater navigation is inertial navigation, a dead reckoning navigation system that outputs an approximate position of the vehicle, but its accuracy decreases over time. A matching positioning algorithm can be used to match the measured gravity data collected by ocean gravimeters with an ocean gravity reference map. The output of the matching position result can be used to correct the position data containing accumulated errors in the inertial navigation system, thereby achieving long-duration, high-precision passive autonomous navigation underwater.

[0003] The marine gravity reference map is a uniformly distributed, highly accurate gravity dataset. Currently, the only marine gravity reference map that covers the entire global ocean area with high accuracy is the DTU-17 satellite altimetry gravity data. In gravity-matching-assisted navigation systems, the high-resolution marine gravity reference map is the primary constraint on improving navigation accuracy. However, the accuracy of navigation using DTU-17 satellite altimetry gravity data in China's coastal waters is insufficient. This is because the DTU-17 satellite altimetry gravity data is based on a European and American model. Its ship-based gravity data is primarily derived from ship-based gravity data in waters near these countries, and it does not fully integrate ship-based gravity data from China's coastal waters. This results in low underwater positioning accuracy. Summary of the Invention

[0004] The present invention aims to provide an adaptive fusion method for ocean multi-source gravity data to solve the problem of low positioning accuracy of existing underwater navigation based on satellite altimetry gravity data.

[0005] The adaptive fusion method of multi-source ocean gravity data in this scheme includes: Step 1: obtain satellite altimetry gravity data and ship-measured gravity data of the target area, perform preprocessing, and organize the gravity data of the target area; Also includes: Step 2: classify the ship-measured gravity data according to a plurality of evaluation indicators; Step 3: Determine the weight coefficients for correcting the satellite altimetry gravity data using the ship-measured gravity data of different mass categories, and loop through the weight coefficients of the different mass categories according to the constraint conditions, and stop the loop when the end condition is met; Step 4: Based on the adaptive combination of multi-source ship-measured gravity data and weight coefficients, and under the constraints of preset rules, the satellite altimetry gravity data is corrected and fused according to the net function interpolation method and the Manhattan interpolation method; In step 5, the root mean square error (RMS) between the fused gravity anomaly reference map and the high-precision ship-measured gravity data that did not participate in the fusion process is calculated. The weight coefficient with the minimum RMS error is used as the fusion parameter combination in step 3, and finally a high-precision gravity reference map is obtained.

[0006] The beneficial effects of this program are: Before using ship-surveyed gravity data to correct satellite altimetry gravity data, the data must be quality-classified to select high-precision ship-surveyed gravity data for fusion. Using this high-precision ship-surveyed gravity data, a combination of net function interpolation and Manhattan interpolation is employed to correct the satellite altimetry gravity data. Using this fused gravity data for underwater navigation effectively improves navigation accuracy. Furthermore, an adaptive fusion method, which compares the corrections made by unfused ship-surveyed gravity data with the fused gravity data, allows for adaptive fusion parameters based on the different gravity data sources, minimizing the impact of human factors on the fusion process.

[0007] Furthermore, in step 2, the evaluation indicators include gross error, internal accuracy, root mean square error and average value.

[0008] The beneficial effect is that the ship-measured gravity data can be classified into multiple evaluation indicators to distinguish the quality of the ship-measured gravity data. In the multi-source gravity data fusion process, different fusion weights are assigned according to the quality classification results of the ship-measured gravity data, so as to perform calculation corrections with different weights.

[0009] Furthermore, in step 3, the constraint conditions are: four weight coefficients are set, the starting values ​​of the weight coefficients are 0.4, 0.3, 0.2 and 0.1 respectively, the loop interval is 0 to 1, the step size during the loop is set to 0.05, and the sum of the weight coefficients is 1.

[0010] The beneficial effect is that by setting different starting values ​​for the weight coefficients and cycling them within a fixed interval with a fixed step size, different combinations of weight coefficients can be obtained, thereby adapting to the fusion of gravity data from different sources and different regions.

[0011] Furthermore, in step 3, the end condition is that the root mean square error between the satellite altimetry gravity data corrected in step 4 and the ship-measured gravity data not involved in the fusion is the smallest.

[0012] The beneficial effect is that when the deviation between the ship-measured gravity data and the satellite-measured gravity data is minimized, the cycle of the weight coefficients is ended, which can ensure that the most appropriate combination of weight coefficients is found, and a more accurate gravity reference map is obtained by fusion, which can reduce positioning errors when used for underwater navigation.

[0013] Furthermore, in step 4, the preset rule is: dividing the target area for the fusion of ship-measured gravity data and satellite altimetry gravity data into small areas of 1°×1°, and in each small area of ​​1°×1°, counting the positive and negative ratios of the correction values ​​of the multi-source ship-measured gravity data points to the same satellite altimetry gravity data point, and determining whether to perform a positive correction or a negative correction based on the positive and negative ratios.

[0014] The beneficial effect is that, by counting the positive and negative correction results of multiple ship-measured gravity data at the same satellite altimetry data point within a relatively small area, the influence of a few low-quality ship-measured gravity data on the correction results of the satellite altimetry data point can be avoided to the greatest extent, thereby improving the accuracy of the correction results.

[0015] Furthermore, in step 4, if the proportion of positive numbers is greater than 50%, only positive numbers are corrected; otherwise, only negative numbers are corrected.

[0016] The beneficial effect is that the corresponding data are corrected according to the proportion of the correction data results, which can improve the situation where the correction results offset each other when the multi-source ship-measured gravity data are corrected for the satellite altimetry gravity data, thereby improving the accuracy of the fused gravity reference map data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flowchart of an adaptive fusion method for ocean multi-source gravity data according to an embodiment of the present invention; Figure 2 This is a gravity reference map after fusion in an adaptive fusion method for ocean multi-source gravity data in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following is further explained in detail through specific implementation methods.

[0019] Example 1 An adaptive fusion method for ocean multi-source gravity data, such as Figure 1 Shown, including: Step 1: Obtain DTU 17 satellite altimetry gravity data and ship-derived gravity data for the target area. The target area is selected based on the actual application, for example, the target area is the coastal area of ​​China.

[0020] Step 2: Classify the ship-measured gravity data according to their quality. The quality classification is carried out according to multiple evaluation indicators, including gross error, internal conformity precision, root mean square error and average value. The quality classification includes four categories: excellent, good, medium and poor. The ship-measured gravity data is classified under a single indicator to obtain a score, and then the quality classification is performed based on the sum of the scores of the ship-measured gravity data under the four indicators.

[0021] (1) Gross errors: The number of gross errors and their proportion to the total number of points in the ship-surveyed gravity data are counted (unit: %). The smaller the proportion of gross errors, the better the quality of the ship-surveyed gravity data.

[0022] (2) Internal Accuracy: Check whether there are any intersections in the ship-surveyed gravity data and calculate the internal accuracy of the intersections (unit: mGal). The smaller the internal accuracy, the better the quality of the ship-surveyed gravity data.

[0023] (3) RMS: Based on the latitude and longitude information of the ship-measured gravity data points, the satellite altimetry gravity data at the track is interpolated using the DTU-17 satellite altimetry gravity data. The RMS of the ship-measured gravity data and the satellite altimetry gravity data at the track is calculated. The smaller the RMS value, the better the quality of the ship-measured gravity data.

[0024] (4) Average value. Calculate the average value of the difference between the ship-measured gravity data and the satellite altimetry gravity data at the track. The gravity data at the satellite altimetry point is the average value of the local sea area. Therefore, statistically, the average value of the difference between the ship-measured gravity data and the surrounding satellite altimetry gravity data should be 0. Unit: mGal. The closer the average value is to 0, the better the quality of the ship-measured gravity data.

[0025] When classifying data, the data is divided into four categories based on a single metric. For example, for gross errors, the thresholds for these four categories are 0-0.005 (1 point), 0.005-0.015 (2 points), 0.015-0.025 (3 points), and greater than 0.025 (4 points). This same approach is used for classification under the internal accuracy, RMS, and average values. Each gravity data set is evaluated using these four metrics and assigned a corresponding score. Lower scores indicate higher accuracy. Based on the score for each data set, scores below 7 are considered excellent, 7-10 are considered good, 10-14 are considered fair, and the remainder are considered poor.

[0026] Step 3: Determine the weight coefficients for correcting the satellite altimetry gravity data using ship-based gravity data of different quality categories, and loop through the weight coefficients of each quality category according to the constraints, stopping the loop when the end condition is met. The constraints are: four weight coefficients are set, with starting values ​​of 0.4, 0.3, 0.2, and 0.1, respectively; the loop interval is 0 to 1; the step size during the loop is set to 0.05; and the weight coefficients add up to 1. The end condition is: the root mean square error between the satellite altimetry gravity data corrected in step 4 and the ship-based gravity data not involved in the fusion is minimized. The gravity data not involved in the fusion is selected from the high-quality gravity data as the test data after the scores of the initial four evaluation indicators are determined. These four gravity data are evenly distributed in different areas.

[0027] Step 4: Adaptively combine the multi-source ship-measured gravity data and the weight coefficients obtained in step 3. Under the constraints of preset rules, correct and fuse the satellite altimetry gravity data according to the net function interpolation method and the Manhattan interpolation method. The preset rules are as follows: divide the target area for the fusion of ship-measured gravity data and satellite altimetry gravity data into small areas of 1°×1°. In each small area of ​​1°×1°, count the positive and negative ratios of the correction values ​​of the multi-source ship-measured gravity data points to the same satellite altimetry gravity data point. The positive and negative ratios are used to determine whether to make a positive correction or a negative correction. If the positive ratio is greater than 50%, only positive corrections are made; otherwise, if the negative ratio is greater than 50%, only negative corrections are made.

[0028] Corrections to satellite altimetry gravity data using multi-source ship-derived gravity data are applied only to satellite altimetry gravity data points located less than 1 nautical mile from the ship-derived gravity data point. The correction process involves interpolating the ship-derived gravity data at the latitude and longitude of the satellite altimetry gravity data point based on distance. This interpolated value is then subtracted from the satellite altimetry gravity data value at that point. The resulting difference is the correction. This correction can be positive or negative. By storing this correction in a structure variable, we can statistically determine whether it is positive or negative.

[0029] Step 5: Calculate the root mean square error between the gravity anomaly reference map obtained by correction and fusion and the high-precision ship-measured gravity data that does not participate in the fusion process. The weight coefficient with the minimum root mean square error is used as the fusion parameter combination in step 3, and finally obtain a high-precision gravity reference map, such as Figure 2 As shown in the figure, visually, the baseline images before and after fusion are the same, with no difference. This is because the values ​​of the correction results are smaller than the values ​​of the gravity data points themselves, and the numerical differences in the data themselves cannot be reflected in the image. Therefore, only one fused gravity baseline image is shown here.

[0030] The comparison of the gravity reference map generated by the adaptive fusion algorithm for multi-source gravity data with the DTU-17 satellite altimetry gravity data was a two-step process. The first step involved selecting three high-precision ship-measured gravity data sources (cook08mv, rc2610, and ht880719) distributed in different sea areas and not involved in the fusion process. The RMS values ​​between each reference map and the ship-measured gravity data were compared to evaluate the accuracy of the reference map, as shown in Table 1. The second step involved selecting the measured gravity data L1 from the South China Sea to evaluate the navigation accuracy of each reference map in underwater navigation applications, as shown in Table 2. The results in Tables 1 and 2 were calculated using MATLAB. Table 1 shows the RMS values ​​between the reference map and the ship-measured gravity data. This is the root mean square error (RMS) between the reference map data interpolated from the ship-measured gravity data points and the ship-measured gravity data.

[0031] The data in Table 2 compares the navigation accuracy results obtained using DTU 17 reference chart data and ship-measured gravity data L1 with the navigation accuracy results obtained by fusing the reference chart data and ship-measured gravity data L1. This primarily demonstrates that the fused reference chart data improves navigation accuracy. The navigation accuracy calculations were performed using a matching route length of 100 nautical miles and a random noise level of 2 milligal.

[0032] Table 1 RMS results of different reference maps compared with ship-measured gravity data

[0033] Note: The unit of RMS is milligal.

[0034] Table 2 Accuracy results of underwater navigation applications using different reference maps

[0035] Note: Navigation accuracy is in nautical miles. Compared with the existing technology, the solution of this embodiment can provide a reasonable fusion coefficient based on multi-source gravity data in different sea areas, reduce the influence of human factors, thereby improving the fineness of the gravity anomaly reference map and enhancing the accuracy of underwater navigation.

[0036] New gravity data is generated by fusing DTU-17 satellite altimetry gravity data with ship-derived gravity data from China's offshore areas. Compared to DTU-17 satellite altimetry gravity data, this new gravity data can improve underwater navigation accuracy. The adaptive fusion method of this embodiment can adaptively set fusion parameters based on gravity data from different sources, improving the efficiency of multi-source gravity data fusion.

[0037] The above is only an embodiment of the present invention, and the common knowledge such as the specific structure and characteristics of the scheme is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An adaptive fusion method for multi-source ocean gravity data, comprising: Step 1: obtain satellite altimetry gravity data and ship-measured gravity data of the target area, perform preprocessing, and organize the gravity data of the target area; It is characterized by further comprising: Step 2: classify the ship-measured gravity data according to a plurality of evaluation indicators; Step 3: Determine the weight coefficients for correcting the satellite altimetry gravity data using the ship-measured gravity data of different mass categories, and loop through the weight coefficients of the different mass categories according to the constraint conditions, and stop the loop when the end condition is met; Step 4: Based on the adaptive combination of multi-source ship-measured gravity data and weight coefficients, and under the constraints of preset rules, the satellite altimetry gravity data is corrected and fused according to the net function interpolation method and the Manhattan interpolation method; In step 5, the root mean square error (RMS) between the corrected and fused gravity anomaly reference map and the high-precision ship-measured gravity data that did not participate in the fusion process is calculated. The weight coefficient with the minimum RMS error is used as the fusion parameter combination for step 3, and finally a high-precision gravity reference map is obtained.

2. The adaptive fusion method for ocean multi-source gravity data according to claim 1, characterized in that: In step 2, the evaluation indicators include gross error, internal accuracy, root mean square error and average value.

3. The adaptive fusion method for ocean multi-source gravity data according to claim 1, characterized in that: In step 3, the constraint conditions are: four weight coefficients are set, the starting values ​​of the weight coefficients are 0.4, 0.3, 0.2 and 0.1 respectively, the loop interval is 0 to 1, the step size during the loop is set to 0.05, and the weight coefficients add up to 1.

4. The adaptive fusion method for ocean multi-source gravity data according to claim 3, characterized in that: In step 3, the end condition is that the root mean square error between the satellite altimetry gravity data corrected in step 4 and the ship-measured gravity data not involved in the fusion is the smallest.

5. The adaptive fusion method for ocean multi-source gravity data according to claim 1, characterized in that: In step 4, the preset rule is: divide the target area for the fusion of ship-measured gravity data and satellite altimetry gravity data into small areas of 1°×1°, and in each small area of ​​1°×1°, calculate the positive and negative ratios of the correction values ​​of the multi-source ship-measured gravity data points to the same satellite altimetry gravity data point, and use the positive and negative ratios to determine whether to perform a positive correction or a negative correction.

6. The adaptive fusion method for ocean multi-source gravity data according to claim 5, characterized in that: In step 4, if the proportion of positive numbers is greater than 50%, only the positive numbers are corrected; Otherwise, only negative numbers will be corrected.

Images