A fishing boat rescue method based on a Beidou intelligent navigation positioning system

Abstract

The present application belongs to the technical field of ship rescue, and provides a fishing boat rescue method based on a Beidou intelligent navigation positioning system, which comprises the following steps: comparing and analyzing the Beidou terminal calculated position and the real position of the rescued fishing boat in historical rescue, dividing the historical rescue into positioning error rescue and normal rescue, then analyzing the spatial consistency and periodic consistency of the positioning error vector, identifying target rescue caused by the multi-path effect, fitting the wave height-positioning error linear relationship according to the satellite elevation angle grouping based on the target rescue data, constructing a multi-path error compensation model combined with the wave direction-azimuth angle difference, and constructing a multi-path effect risk prediction model using the sea wave parameters and the fishing boat position parameters of normal / target rescue, if there is a risk, correcting the Beidou positioning result using the compensation model to solve the positioning error problem caused by the sea wave multi-path effect, improving the positioning accuracy to within the nominal accuracy range of Beidou, and being suitable for offshore fishing boat emergency rescue.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of ship rescue, in particular to a fishing boat rescue method based on a Beidou intelligent navigation positioning system. BACKGROUND

[0002] The Beidou satellite navigation system has been widely used in the rescue of fishing boats at sea, and its positioning accuracy directly determines the efficiency and success rate of the rescue. However, in the scenario of sea rescue, satellite signals are easily affected by the reflection of sea waves to form a multipath effect: the direct wave and the reflected wave superimpose and interfere at the antenna, causing pseudorange measurement error, and further causing positioning error, which may cause the rescue ship to miss the target and prolong the rescue time in severe cases.

[0003] Existing fishing boat rescue positioning methods mostly rely on the original output data of the Beidou terminal, and do not construct a targeted risk prediction and error correction mechanism for the multipath effect: first, there is a lack of deep mining of historical rescue data, and the correlation between factors such as sea waves and satellite geometry and multipath error cannot be quantified; second, a real-time risk prediction model has not been established, making it difficult to identify multipath hazards in advance; third, the positioning correction mostly uses a fixed filtering algorithm, which has limited correction accuracy.

[0004] Therefore, the application provides a fishing boat rescue method based on a Beidou intelligent navigation positioning system. SUMMARY

[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem raised in the background.

[0006] The technical scheme adopted by the application to solve its technical problems is: a fishing boat rescue method based on a Beidou intelligent navigation positioning system, comprising: Step 1: comparing and analyzing the Beidou terminal calculated position and the real position of the rescued fishing boat in historical multiple rescues, and dividing the historical rescue into positioning error rescue and normal rescue; Step 2: for positioning error rescue, performing fluctuation period coincidence analysis and error consistency analysis on the positioning error of the rescued fishing boat in the target sea area and the wave period, to determine whether the positioning error is caused by the multipath effect, and if so, marking the positioning error rescue as target rescue; Step 3: correlating the wave height and positioning error in different satellite elevation angle intervals in the target rescue to construct a multipath error compensation model; Step 4: constructing a multipath effect risk prediction model according to the wave parameters and the position of the rescued fishing boat in normal rescue and target rescue, and combining the current wave parameters and the position of the fishing boat to be rescued to determine whether there is a multipath effect risk in the current rescue; Step five: If exists, input the current sea wave parameter and the position of the fishing boat to be rescued into the multipath error compensation model to dynamically correct the Beidou positioning result, and obtain the corrected positioning coordinates of the fishing boat to be rescued.

[0007] Further, the historical rescue is divided into positioning error rescue and normal rescue in the following way: For each rescued fishing boat in any historical rescue; The horizontal error of the Beidou terminal position solution is calculated using the geographic coordinate distance formula; Compare the horizontal error with the error threshold. If there is a rescued fishing boat with a horizontal error greater than the error threshold, mark the historical rescue as a positioning error rescue; Otherwise, if there is no rescued fishing boat with a horizontal error greater than the error threshold in the historical rescue, it is a normal rescue.

[0008] Further, the way of fluctuation period coincidence analysis of the positioning error of the fishing boat to be rescued in the target sea area and the sea wave period is as follows: Extract the sea wave period, i.e. the wave peak interval time, during the rescue period from the historical meteorological database, and record the wave peak time; Extract the continuous positioning points of the rescued fishing boat from the Beidou terminal log, calculate the horizontal error with the true position, and integrate it into an error sequence; For any rescued fishing boat in each positioning error rescue: Segment the error sequence according to the sea wave period, calculate the Pearson correlation coefficient of the error peak value and the wave peak time; Statistically, the proportion of the number of rescued fishing boats with a Pearson correlation coefficient greater than the preset correlation coefficient in all rescued fishing boats is compared with the preset proportion. If it is greater than the preset proportion, the error rescue has periodic consistency.

[0009] Further, the process of error consistency analysis of the positioning error of the fishing boat to be rescued in the target sea area and the sea wave period is as follows: For each rescued fishing boat in the target sea area; Take the Beidou positioning point of the rescued fishing boat as the starting point and the true position as the ending point to calculate the positioning error vector, which includes the error direction and the error size. The error direction is the horizontal angle from the Beidou terminal position solution to the true position; Decompose the error direction of the rescued fishing boat into x and y components in the horizontal plane, and take the arithmetic mean of the x and y components of all rescued fishing boats to obtain the average direction; Calculate the error of the error direction and the average direction, take the absolute value and then average to obtain the direction concentration degree; If the direction concentration degree is less than or equal to the threshold, the error direction is concentrated; Calculate the deviation of the error direction of the rescued fishing boat and the satellite reflection direction; Calculate the mean of the absolute values of the deviations of the error directions of all rescued fishing boats and the satellite reflection direction to obtain an absolute mean error; If the absolute mean error is less than or equal to a preset absolute mean value, the error direction is consistent with the satellite reflection direction; If the error directions are concentrated and consistent with the satellite reflection direction, there is spatial consistency.

[0010] Further, the way of judging whether the positioning error is caused by multipath effect is: For any once positioning error rescue: If there is periodic consistency and spatial consistency, it is determined that the positioning error is caused by multipath effect, and the error rescue is marked as a target rescue.

[0011] Further, the way of constructing the multipath error compensation model is: Real-time wave height, wave direction-azimuth angle difference, and satellite elevation angle are obtained from historical target rescue data as input parameters, and positioning error size and error direction are obtained as output parameters, wherein the satellite elevation angle is the horizontal angle between the satellite and the fishing boat antenna connection line; Based on the sliding window method, the wave height and positioning error under different satellite elevation angles are linearly fitted, and the fitting equation slopes of the wave height and positioning error in each sliding window are compared and analyzed to identify the satellite elevation angle grouping points; The satellite elevation angle is divided into two elevation angle groups according to the satellite elevation angle grouping points, linear regression is performed on each group of data, and the wave height interval of each satellite elevation angle group is recorded; The error direction is consistent with the satellite reflection direction, and the reflection direction = satellite azimuth angle , is negative when, is positive when, wherein, is the wave direction-azimuth angle difference; Integrate the relationship between the positioning error and the wave height and the relationship between the error direction and the satellite elevation angle to form a multipath error compensation model of error size + error direction.

[0012] Further, the way of identifying the satellite elevation angle grouping is: The wave height and error in the continuous satellite elevation angle interval are fitted by the sliding window; The regression coefficients are solved by the least square method to obtain the linear regression equation in each sliding window; The central satellite elevation angle, wave height coefficient, and constant term of each sliding window are recorded, wherein the wave height coefficient is the slope of the linear regression equation; Calculate the difference of adjacent sliding window k, and calculate the mean μ and standard deviation σ of all differences, set the sudden change threshold as μ + 2σ; Traverse all the differences of adjacent sliding window k, and compare them with the sudden change threshold, the upper limit of the previous sliding window corresponding to the first difference exceeding the sudden change threshold is the satellite elevation grouping point.

[0013] Further, the construction process of the multi-path effect risk prediction model is: Extract the continuous wave data of the normal rescue and target rescue positioning error period; Identify the wave peak by the extreme value method, that is, the highest point between two adjacent troughs, and record the wave height of each wave, that is, the vertical distance between the wave peak and the previous trough; Arrange the wave height in descending order, calculate the average of the top third wave height, and get the effective wave height; Construct a multi-path effect risk prediction model, input the current sea wave parameters and the position parameters of the fishing boat to be rescued, and output the probability of positioning error caused by multi-path effect; The dependent variable Y is whether the multi-path effect occurs, Y=1 represents the target rescue, and Y=0 represents the normal rescue; Extract the sea wave parameters and fishing boat position parameters from historical data, wherein the sea wave parameters include effective wave height, sea wave period, and wave direction-azimuth angle difference, and the fishing boat position parameters include satellite elevation angle; Construct the interaction term of sea wave parameters and fishing boat position parameters, and the interaction term is the wave height-elevation angle coordination term: X=H×(30°-θ 卫 ), wherein H is the wave height, θ 卫 is the satellite elevation angle; Take the sea wave parameters, fishing boat position parameters and wave height-elevation angle coordination term as input features Calculate the Pearson correlation coefficient of the input features and the dependent variable Y, retain the features with Pearson correlation coefficient greater than or equal to the preset correlation coefficient, and get the feature set [x1, x2,..., x i ], wherein x i is the i-th feature; Take the feature set [x1, x2,..., x i ] as the independent variable, and construct the probability prediction formula: ; Wherein β0 is a constant term, and β1~β4 are regression coefficients; Use the training set to fit the regression coefficients, and finally get the multi-path effect risk prediction model.

[0014] Further, the judgment method of whether the current rescue has multi-path effect risk is: Input the current rescue sea wave parameter and the position of the fishing boat to be rescued into the risk prediction model of multipath effect, and output the probability of the current positioning of the fishing boat to be rescued appearing multipath effect; If the probability is greater than or equal to the preset probability, the current rescue has a risk of multipath effect.

[0015] Further, the dynamic correction method for the Beidou positioning result is: Determine the satellite elevation angle according to the current Beidou positioning position of the fishing boat to be rescued, judge the satellite elevation angle group, input the wave height into the multipath error compensation model, and obtain the positioning error size; Calculate the wave direction-azimuth angle, input the multipath error compensation model, and obtain the error direction; After converting the error direction into radians, input the positioning error size and the error direction into the azimuth-distance conversion latitude and longitude formula, and obtain the corrected positioning coordinates of the fishing boat to be rescued.

[0016] The beneficial effects of the present application are as follows: through spatiotemporal consistency analysis, the deviation caused by multipath effect is accurately screened from the positioning error, other interference factors such as equipment failure are excluded, the attribution accuracy is improved, the error compensation model constructed based on the satellite elevation angle group can output a quantitative error vector according to the real-time wave height and satellite parameters, the dynamic correction of the positioning result is realized, the positioning accuracy is improved, the multipath effect risk prediction model can judge the risk probability of the rescue scene in advance, and provides a basis for rescue decision-making, avoids the hysteresis of passive correction, and combines environmental parameters such as sea wave period and wave direction, so that the model can adapt to the positioning error law under different sea conditions, and has a wide application range. BRIEF DESCRIPTION OF DRAWINGS

[0017] The present application will be further described below with reference to the drawings.

[0018] Figure 1 is a step flow chart of a fishing boat rescue method based on a Beidou intelligent navigation positioning system according to the present application; Figure 2 is a logic judgment chart of a fishing boat rescue method based on a Beidou intelligent navigation positioning system according to the present application. DETAILED DESCRIPTION

[0019] In order to make the technical means, creative features, purposes and effects realized by the present application easy to understand, the present application will be further described below with reference to the specific embodiments.

[0020] Please refer to Figure 1 The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to the present application embodiment includes the following steps: Step 1: Compare and analyze the Beidou terminal calculated position and the real position of the rescued fishing boat in historical multiple rescues, and divide the historical rescue into positioning error rescue and normal rescue; In step one, the process of dividing historical rescues into error rescues and normal rescues includes: From the historical log of Beidou terminal or the rescue platform archive data, obtain the Beidou positioning data of the rescued fishing boat at the key nodes of historical rescue, including longitude and latitude, positioning timestamp, and positioning mode; For each rescued fishing boat in any historical rescue: Record the Beidou terminal calculated position as P 测 : (Lon 测 , Lat 测 , H 测 ) , and the true position as P 真 : (Lon 真 , Lat 真 , H 真 ) , where Lon is the longitude and Lat is the latitude; Calculate the horizontal error, which is the distance difference in the plane coordinate system, considering the influence of the earth's curvature, using the geographic coordinate distance formula: ; Where, (longitudinal difference, unit: radian), (latitude difference, unit: radian), and R is the average radius of the earth; Set the error threshold by combining the nominal accuracy of the Beidou system and the actual needs of the rescue scene, specifically: Obtain the horizontal nominal accuracy of the corresponding positioning mode from the technical specifications of the Beidou terminal, including: Single-point positioning: The nominal horizontal accuracy of the third-generation Beidou terminal is usually 5-10 meters; Differential positioning: After correction by the base station differential signal, the nominal horizontal accuracy is usually 1-3 meters; According to the actual accuracy requirements of the rescue scene, determine the acceptable maximum normal error, including: Offshore rescue: The radar accuracy of the rescue ship is greater than or equal to 5 meters, and the fishing boat density is high, requiring accurate positioning to avoid missing the target, with an acceptable maximum error of 10 meters (more than 10 meters may cause the search range of the rescue ship to expand by 1 times); Offshore rescue: The fishing boat density is low, and the search range of the rescue ship can be appropriately expanded, with an acceptable maximum error of 15 meters (more than 15 meters will cause the search time to be extended by more than 30%); Special scene (such as night, bad weather): Visual search is difficult, and a more stringent threshold is required, with an acceptable maximum error of 8 meters; Set the error threshold T0 as the intersection of the horizontal nominal accuracy and the upper limit of the acceptable maximum normal error, i.e., T0 = max (1.5 times the horizontal nominal accuracy, 0.8 times the acceptable maximum normal error); It can be understood that the setting logic of the error threshold is as follows: 1.5 times the horizontal nominal accuracy: to avoid normal rescue misjudgment caused by individual differences of equipment (such as the accuracy of some terminals being slightly lower than the nominal value); 0.8 times the acceptable maximum normal error: to reserve identification space for multipath deviation (multipath deviation is usually 10-50 meters, much larger than the value); For any historical rescue, For each rescued fishing boat, compare the horizontal error with the error threshold. If there is a rescued fishing boat with a horizontal error greater than the error threshold, mark the historical rescue as a positioning error rescue; Otherwise, if there is no rescued fishing boat with a horizontal error greater than the error threshold in the historical rescue, it is a normal rescue; The effect of dividing the historical rescue into positioning error rescue and normal rescue is to filter out positioning error rescue from historical rescue data, which is irrelevant interference for subsequent focus on multipath effect analysis, and is the data preprocessing and classification entrance of the whole process; Step two: for positioning error rescue, perform fluctuation period coincidence analysis and error consistency analysis on the positioning error of the fishing boat to be rescued in the target sea area and the wave period, to determine whether the positioning error is caused by multipath effect. If so, mark the positioning error rescue as target rescue; Please refer to Figure 2 In step two, the process of performing error consistency analysis on the fishing boat to be rescued in the target sea area includes: For any positioning error rescue: Collect the positioning data, true position data, satellite parameters and wave parameters of all rescued fishing boats in the target sea area; Among them, the positioning data is the Beidou terminal solution position P 测 = (Lon 测 , Lat 测 ) of each fishing boat, the true position data is the true position P 真 = (Lon 真 , Lat 真 ) of each fishing boat, the satellite parameter is the azimuth of the main satellite (the horizontal direction of the satellite relative to the fishing boat), and the wave parameter is the wave direction (the propagation direction of the wave) of the target sea area; For each rescued fishing boat, calculate the positioning error vector from the Beidou positioning point to the true position, including the error direction and the error size; The geographic coordinate distance formula is used to calculate the horizontal error between the Beidou terminal calculated position P 测 = (Lon 测 , Lat 测 ) and the true position P 真 = (Lon 真 , Lat 真 ) of each fishing boat, and the error size is obtained; The error direction is the horizontal angle from the Beidou terminal calculated position to the true position, calculated in the geographic coordinate system with 0° as north and clockwise increasing, and the specific formula is: Calculate the difference between longitude and latitude (unit: radian): ; ; Based on the mathematical coordinate system: x-axis east, y-axis north, counterclockwise positive, calculate the initial direction angle using the inverse tangent function: ; Among them, cos(Lat 真,rad ) is used to correct the change of longitude difference with latitude (the longitude difference of 1° corresponds to a distance of about 111 kilometers near the equator, and tends to 0 near the poles); Convert the initial direction angle from the mathematical coordinate system direction to the geographic coordinate system direction, and the formula is: ; The final error vector (E ) of each rescued fishing boat is output; For the error direction of each rescued fishing boat, decompose it into x (east) and y (north) components in the horizontal plane: , ; Take the arithmetic mean of the x and y components of all rescued fishing boats respectively: , ; Among them, n is the number of rescued fishing boats; Convert the average vector to the geographic coordinate system direction: For each rescued fishing boat, calculate the error δ i between the error direction and the average direction: δ i =min(|θ error,i -θ avg |,360°-|θ error,i -θ avg |), take the absolute value and then average to get the direction concentration ; The directional concentration is compared with a threshold. If the directional concentration is less than or equal to the threshold, the error direction is determined to be concentrated. It should be noted that the threshold is set as follows: filter historical samples of multi-path and non-multi-path scenarios, calculate the orientation concentration of each group, and find the point where the orientation concentration distributions of the two types of samples overlap the least, which is the threshold. According to the formula The wave direction-azimuth difference was calculated. , where θ 波 It is the direction of wave propagation, θ 卫 It is the satellite azimuth angle; Satellite reflection direction θ 反 The direction from which the satellite signal reaches the fishing boat after being reflected by the waves is equal to the satellite azimuth angle + 180°, i.e.: θ 反 =(θ 卫 +180°) mod 360°; For each rescued fishing vessel, calculate the error direction θ of the rescued fishing vessel. error With respect to satellite reflection direction θ 反 deviation γ i γ i =min(|θ error,i -θ 反 |,360°-|θ error,i -θ 反 |); The error direction θ of all rescued fishing boats error With respect to satellite reflection direction θ 反 The absolute value of the deviation is taken and the mean is calculated to obtain the absolute mean of the error. The absolute mean of the error is compared with the preset absolute mean. If the absolute mean of the error is less than or equal to the preset absolute mean, the error direction is determined to be consistent with the satellite reflection direction. It should be noted that the preset absolute mean is set as follows: the deviation between the error direction of the multipath effect and the satellite reflection direction is essentially the offset of the reflected wave direction caused by the tilt of the sea wave surface. When the satellite signal is reflected by the sea wave, the reflection angle equals the incident angle. If there is a slope α on the sea wave surface, the reflection direction will be offset by 2α, which leads to the upper limit of the deviation between the error direction and the ideal reflection direction being 2α, that is, the preset absolute mean is 2α. If the error directions are concentrated and the error directions are consistent with the satellite reflection direction, then spatial consistency is determined to exist; In step two, the process of analyzing the coincidence of the positioning error of the fishing vessel to be rescued in the target sea area with the wave period includes: For any fishing vessel rescued in each positioning error rescue operation: Extract the wave cycle, i.e. the wave crest interval, during the rescue period from the historical meteorological database, and record the wave crest time; Extract the continuous positioning points (longitude and latitude) of the rescued fishing vessels from the Beidou terminal log, and calculate the horizontal error with the true position , t is the timestamp, forming an error sequence ; Segment the error sequence by the wave period, calculate the Pearson correlation coefficient r of the error peak value and the wave peak time, compare r with the preset correlation coefficient, count the proportion of the number of fishing vessels with r greater than the preset correlation coefficient in all rescued fishing vessels, and compare it with the preset proportion, if greater than the preset proportion, the error rescue has periodic consistency; It should be noted that the preset correlation coefficient is the correlation threshold of the single-ship error peak value and the wave peak time, and the preset proportion is the proportion threshold of the number of fishing vessels meeting the correlation, which is set by the person skilled in the art; In step two, the process of determining whether the positioning error is caused by multipath effect includes: For any one positioning error rescue: If it has periodic consistency and spatial consistency, it is determined that the positioning error is caused by multipath effect, and the error rescue is marked as a target rescue; It can be understood that the judgment logic of whether the positioning error is caused by multipath effect is: The essence of multipath effect is the superposition and interference of satellite direct wave and sea wave reflected wave, and the sea wave has periodic motion characteristics (time dimension), and the reflection direction and intensity of the sea wave in the same sea area have spatial similarity (spatial dimension), which directly leads to the corresponding time and space rules of the positioning error: Time dimension: the sea wave moves with a fixed period (peak to trough to peak), the propagation path and phase difference of the reflected wave also change with the period, and finally the positioning error presents a fluctuation synchronized with the sea wave period; Spatial dimension: in the same sea area, the wave direction and height distribution of the sea wave are relatively uniform, and if multiple fishing vessels are in the same satellite signal reflection path coverage area, they will be disturbed by similar reflected waves, resulting in consistent error direction and size; For positioning error rescue, the role of determining whether the positioning error is caused by multipath effect is: Attributing the positioning error rescue to multipath effect, accurately screening out the error rescue caused by multipath effect (i.e. target rescue) from all positioning errors, and excluding the interference of other error causes (such as equipment failure); Step three: Correlation analysis of wave height and positioning error in different satellite elevation intervals in the target rescue, and construction of a multipath error compensation model; In step three, the construction process of the multipath error compensation model includes: Real-time wave height, satellite elevation angle, and wave direction-azimuth angle difference are obtained from historical target rescue data as input parameters, and positioning error size and error direction are obtained as output parameters, wherein the satellite elevation angle is the horizontal angle between the satellite and the antenna of the fishing boat, and is used to measure the position of the fishing boat; A sliding window is defined, and samples in a continuous satellite elevation angle interval are intercepted for fitting, with the satellite elevation angle as the axis, the window size being W, and the sliding step being 1; It should be noted that, based on the attenuation law of reflected wave energy with elevation angle, when the satellite elevation angle is lower than the threshold value, the signal propagates close to the sea surface, the sea surface is approximately specular reflection, the reflected wave energy is strong, and the interference after superposition is significant. When the satellite elevation angle is in a certain interval greater than the threshold value, the sea surface is mixed reflection, the reflected wave energy rapidly attenuates with the increase of the satellite elevation angle, and the interference is weakened. When the satellite elevation angle is greater than the upper limit of the interval, the signal propagation path is far away from the sea surface, the sea surface is diffuse reflection, the reflected wave energy is dispersed, and cannot form effective superposition interference. At this time, the multipath effect can be ignored. For the samples in each sliding window, the wave height and the error are linearly fitted, and the linear regression equation in each sliding window is wherein k i is the wave height coefficient in the i-th sliding window, b i is the constant term in the i-th sliding window. The regression coefficients are solved by the least square method to obtain the linear regression equation in each sliding window. It can be understood that the physical meaning of the wave height coefficient k is that, in this satellite elevation angle interval, the positioning error increases by m when the wave height increases by 1 m. The physical meaning of the constant term b is that, in this satellite elevation angle interval, the basic error when the wave height is 0 is the inherent error of the device and the interference of non-wave height factors. Each sliding window is fitted one by one, and the center satellite elevation angle (the average of the satellite elevation angles at the beginning and end of the window), the wave height coefficient k i , and the constant term b i of each window are recorded. The center satellite elevation angle of each sliding window is taken as the x-axis, and k i and b i are taken as the y-axis. The wave height coefficient-satellite elevation angle curve and the constant term-satellite elevation angle curve are drawn to directly observe the variation law of the parameters with the satellite elevation angle. It should be noted that the analysis of the curve characteristics is based on the expected physical mechanism: k-E curve: at low satellite elevation angle, the signal propagates close to the sea surface, the influence of wave height on reflected wave is strong -> k value is large and stable; with the increase of the satellite elevation angle, the signal propagation path is far away from the sea surface, the influence of wave height is weakened -> k value gradually decreases; when the satellite elevation angle exceeds a certain threshold value, k value will suddenly decrease (abrupt change), and then the downward trend becomes slow (in the medium elevation angle region, the influence of wave height is stable at a low level; b-E curve: the constant term b mainly reflects non-wave height factors (such as equipment error), and changes relatively flat with elevation angle, and the sudden change is not obvious (therefore, the sudden change of k value is preferred as the grouping basis); The sudden change point is the satellite elevation angle position where the change amplitude of k value suddenly increases, which is identified by calculating the difference of k value of adjacent windows: Calculate the difference of k value of adjacent windows: , that is, the absolute value of the difference of k value between the i+1th window and the ith window; Calculate the mean μ and standard deviation σ of all , and set the sudden change threshold as μ+2σ, and the that exceeds the sudden change threshold is regarded as significant difference, corresponding to the sudden change of k value; Traverse the k value difference of all adjacent sliding windows , and compare it with the sudden change threshold, and the upper limit of the previous sliding window corresponding to the first that exceeds the sudden change threshold is the satellite elevation angle grouping point; Divide the historical samples into two elevation angle groups according to the satellite elevation angle grouping point, and perform linear regression on the data of each group, and the linear regression equation is , the regression coefficients are solved by the least square method, the linear regression equation is obtained, and the wave height interval of each satellite elevation angle group is recorded; The error direction needs to be determined in combination with the wave direction-azimuth angle difference and the satellite reflection law, and the historical data statistics show that: The error direction is consistent with the satellite reflection direction, and the reflection direction = satellite azimuth , , take - when , take + when Integrate the error size and the relationship between wave height and satellite elevation angle, and form a double-output model of error size + error direction: ;

[0021] , k2xH+b2 is the linear regression equation of error and wave height in the satellite elevation angle interval [H1, H2], and [H0, H1], [H1, H2] are two elevation angle groups divided by the satellite elevation angle grouping point; The role of constructing a multi-path error compensation model is to establish a quantitative relationship between input parameters and positioning error based on historical data of target rescue and normal rescue, to provide a feasible mathematical model for subsequent real-time correction of positioning error, and to be the core link to solve how to compensate for multi-path error; Step four: according to the sea wave parameters of normal rescue and target rescue and the position of the rescued fishing boat, a multi-path effect risk prediction model is constructed, and the current rescue is judged to be at risk of multi-path effect by combining the current sea wave parameters and the position of the fishing boat to be rescued; In step four, the construction process of the multi-path effect risk prediction model includes: Extracting continuous wave data in the positioning error period of normal rescue and target rescue; Eliminate extreme wave height using 3σ principle: eliminate wave height data greater than three times the mean value, which may be sensor interference; Identify wave peaks by extreme value method, i.e. the highest point between two adjacent troughs, record the wave height H i of each wave, i.e. the vertical distance between the wave peak and the previous trough; Arrange the wave heights in descending order: H1≥H2≥...≥H N , where N is the number of wave heights; Extract the top one-third wave height: take k=N / 3 wave heights (if N is not a multiple of 3, take the integer part, such as N=152, then k=50); Calculate the average of the top one-third wave height to get the effective wave height: H s =(H1+H2+...+H k ) / k; Build a prediction model, input the current sea wave parameters and the position parameters of the fishing boat to be rescued, and output the probability of positioning error caused by multi-path effect, which is: Dependent variable is whether multi-path effect occurs (Y); Y=1 (target rescue): positioning error is greater than or equal to threshold, where the threshold is set according to the nominal accuracy of Beidou system and the actual demand of rescue scene; Y=0 (normal rescue): positioning error is within normal range; Extract sea wave parameters and fishing boat position parameters from historical data, combining the core causes of multi-path effect; Wherein, the sea wave parameters include effective wave height, sea wave period, wave direction-azimuth angle difference, and the fishing boat position parameters include satellite elevation angle; Standardize sea wave parameters and fishing boat position parameters using Z-score: standardized feature value=(original value-feature mean) / feature standard deviation; Construct the interaction term of sea wave parameters and fishing boat position parameters, and the interaction feature is the wave height-elevation angle coordination term: X=H×(30°-θ 卫 ), where H is the wave height and θ 卫 is the satellite elevation angle; Calculate the Pearson correlation coefficient of the independent variable and Y, and retain the features with Pearson correlation coefficient greater than or equal to the preset correlation coefficient to get the feature set [x1,x2,...,xi ], wherein x i is the ith feature; The cleaned samples are proportionally divided into a training set and a test set, the training set is used to fit the model parameters, and the test set is used to verify the model generalization ability; Taking the feature set [x1, x2,..., x i ] as the independent variable, a probability prediction formula is constructed: ; Wherein β0 is a constant term, β1~β4 is the regression coefficient (reflecting the influence strength and direction of the independent variable on ); The regression coefficients are fitted using the training set, and the risk prediction model of the multi-path effect is finally obtained; The sea wave parameters of the current rescue and the position of the fishing boat to be rescued are input into the risk prediction model of the multi-path effect, and the probability of the current fishing boat positioning appearing the multi-path effect is output; The probability is compared with the preset probability, if the probability is greater than or equal to the preset probability, the current rescue exists the risk of multi-path effect; It should be noted that the preset probability is the judgment threshold of the multi-path effect risk, which is set by the person skilled in the art; The role of constructing the multi-path effect risk prediction model is: Predict the occurrence probability of the multi-path effect before real-time rescue, avoid the passive situation of correcting after the positioning error has occurred, and belong to the pre-risk prevention and control link; Step five: if so, input the current sea wave parameters and the position of the fishing boat to be rescued into the multi-path error compensation model, dynamically correct the Beidou positioning result, and obtain the corrected positioning coordinates of the fishing boat to be rescued; In step five, the process of dynamically correcting the Beidou positioning result to obtain the corrected positioning coordinates includes: Determine the satellite elevation angle according to the current Beidou positioning position of the fishing boat to be rescued, judge the satellite elevation angle group, input the wave height into the corresponding formula, and obtain the positioning error size ; Calculate the wave direction-azimuth angle difference , substitute it into the direction formula to obtain the error direction θ 误 ; The error vector is: offset 误 meters in the direction of θ 原 ; Let the original coordinates be (Lon 原 , Lat 修 ), and the corrected coordinates be (Lon 修 , Lat 原 ), and the offset coordinates on the earth's surface are obtained through the azimuth-distance to latitude and longitude formula: ; ; wherein R is the average radius of the earth; Obtain the Beidou positioning data to be corrected from the rescue platform: original coordinates (Lon 原 ,Lat 原 ), and positioning timestamp (aligned with the real-time weather data timestamp); After converting θ 误 into radians, the positioning error size and error direction are substituted into the azimuth-distance conversion latitude and longitude formula to obtain the corrected positioning coordinates; The role of the dynamic correction of the Beidou positioning result is: when it is predicted that there is a risk of multipath effect, the original positioning data is corrected by using the multipath effect compensation model, and finally the accurate position of the fishing boat is output, which is the landing execution link of the whole process.

[0022] The technical scheme and advantages of the embodiments of the present application are: the Beidou terminal calculated position and the real position of the rescued fishing boat in historical rescues are compared and analyzed, the historical rescues are divided into positioning error rescue and normal rescue, for the positioning error rescue, the positioning error of the rescued fishing boat in the target sea area and the wave period are analyzed for fluctuation period coincidence and error consistency, it is judged whether the positioning error is caused by the multipath effect, if so, the positioning error rescue is marked as the target rescue, the wave height in different satellite elevation angle intervals and the positioning error in the target rescue are analyzed, a multipath error compensation model is constructed, according to the wave parameters and the position of the rescued fishing boat in the normal rescue and the target rescue, a multipath effect risk prediction model is constructed, the current wave parameters and the position of the fishing boat to be rescued are combined to judge whether there is a risk of multipath effect in the current rescue, if there is, the current wave parameters and the position of the fishing boat to be rescued are input into the multipath error compensation model to dynamically correct the Beidou positioning result and obtain the corrected positioning coordinates. The Beidou terminal calculated position and the real position of the rescued fishing boat in historical rescues are compared and analyzed, the historical rescues are divided into positioning error rescue and normal rescue, the positioning error rescue is analyzed for spatial consistency and periodic consistency of the positioning error vector, the target rescue caused by the multipath effect is identified, and based on the target rescue data, the wave height-positioning error linear relationship is fitted according to the satellite elevation angle grouping, the wave direction-azimuth angle difference is constructed to form a multipath error compensation model, and the wave parameters and the fishing boat position parameters of the normal / target rescue are used to construct a multipath effect risk prediction model, if it is predicted that there is a risk, the Beidou positioning result is corrected by using the compensation model, the positioning error caused by the wave multipath effect is solved, the positioning accuracy is improved to the range of the nominal accuracy of Beidou, and it is suitable for the emergency rescue of fishing boats at sea.

[0023] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A fishing boat rescue method based on a Beidou intelligent navigation positioning system, characterized in that: Comprise: Step one: compare the position calculated by the Beidou terminal of the rescued fishing boat in historical rescues with the real position, and divide the historical rescues into positioning error rescue and normal rescue; Step two: for positioning error rescue, analyze the fluctuation period coincidence of the positioning error of the target sea area and the wave period, and the error consistency, to determine whether the positioning error is caused by multipath effect, if so, mark the positioning error rescue as target rescue; Step three: correlate the wave height and positioning error in different satellite elevation intervals in target rescue to build a multipath error compensation model; Step four: according to the wave parameters and the position of the rescued fishing boat in normal rescue and target rescue, build a multipath effect risk prediction model, combine the current wave parameters and the position of the fishing boat to be rescued to determine whether there is a risk of multipath effect in the current rescue; Step five: if so, input the current wave parameters and the position of the fishing boat to be rescued into the multipath error compensation model to dynamically correct the Beidou positioning result and get the corrected positioning coordinates of the fishing boat to be rescued.

2. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 1, characterized in that: The way to divide the historical rescue into positioning error rescue and normal rescue is: For each rescued fishing boat in any historical rescue; Calculate the horizontal error of the Beidou terminal calculated position using the geographic coordinate distance formula; Compare the horizontal error with the error threshold, if there is a rescued fishing boat with horizontal error greater than the error threshold, mark the historical rescue as positioning error rescue; Otherwise, if there is no rescued fishing boat with horizontal error greater than the error threshold in the historical rescue, it is normal rescue.

3. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 1, characterized in that: The way to analyze the fluctuation period coincidence of the positioning error of the target sea area and the wave period is: Extract the wave period of the rescue period from the historical meteorological database, that is, the wave peak interval time, and record the wave peak time; Extract the continuous positioning points of the rescued fishing boat from the Beidou terminal log, calculate the horizontal error with the real position, and integrate them into an error sequence; For any one rescued fishing boat in each positioning error rescue: Divide the error sequence by the wave period, calculate the Pearson correlation coefficient of the error peak value and the wave peak time; Statistical the proportion of the number of rescued fishing boats with Pearson correlation coefficient greater than the preset correlation coefficient in all rescued fishing boats, and compare it with the preset proportion, if greater than the preset proportion, the error rescue has period consistency.

4. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 3, characterized in that: The process of error consistency analysis of the positioning error of the target sea area and the wave period is: For each rescued fishing boat in the target sea area; Take the Beidou positioning point of the rescued fishing boat as the starting point and the real position as the ending point to calculate the positioning error vector, which includes error direction and error size, the error direction is the horizontal angle from the Beidou terminal calculated position to the real position. Decompose the error direction of the rescued fishing boat into x and y components on the horizontal plane, and take the arithmetic mean of the x and y components of all rescued fishing boats to obtain the average direction; Calculate the error of the error direction and the average direction, and then take the absolute value to obtain the direction concentration; If the direction concentration is less than or equal to the threshold, the error direction is concentrated; Calculate the deviation of the error direction of the rescued fishing boat from the satellite reflection direction; Take the absolute value of the deviation of the error direction of all rescued fishing boats from the satellite reflection direction, and calculate the mean to obtain the error absolute mean; If the error absolute mean is less than or equal to the preset absolute mean, the error direction is consistent with the satellite reflection direction; If the error direction is concentrated and the error direction is consistent with the satellite reflection direction, there is spatial consistency.

5. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 4, characterized in that: The way to determine whether the positioning error is caused by multipath effect is: For any once positioning error rescue: If there is periodic consistency and spatial consistency, it is determined that the positioning error is caused by multipath effect, and the error rescue is marked as target rescue.

6. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 1, characterized in that: The construction method of the multipath error compensation model is: From the historical target rescue data, real-time wave height, wave direction-azimuth angle difference, and satellite elevation angle are obtained as input parameters, and positioning error size and error direction are obtained as output parameters, wherein the satellite elevation angle is the horizontal angle between the satellite and the fishing boat antenna; Based on the sliding window method, the wave height and the positioning error under different satellite elevation angles are linearly fitted, and the fitting equation slopes of the wave height and the positioning error in each sliding window are compared and analyzed to identify the satellite elevation angle grouping point; With the satellite elevation angle grouping point as the boundary, the satellite elevation angle is divided into two elevation angle groups according to the satellite elevation angle grouping point, linear regression is performed on each group of data, and the wave height interval of each satellite elevation angle group is recorded; Error direction is the same as satellite reflection direction, reflection direction = satellite azimuth , negative when, positive when, is the wave direction-azimuth angle difference; Integrate the relationship between the positioning error and the wave height and the relationship between the error direction and the satellite elevation angle to form a multipath error compensation model of error size + error direction.

7. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 6, characterized in that: The identification method of the satellite elevation angle grouping is: Fitting is performed on the wave height and the error in the continuous satellite elevation angle interval through the sliding window; The regression coefficients are solved through the least squares method to obtain the linear regression equation in each sliding window; The center satellite elevation angle, wave height coefficient, and constant term of each sliding window are recorded, wherein the wave height coefficient is the slope of the linear regression equation; Calculate the difference value of adjacent sliding windows k, and calculate the mean μ and standard deviation σ of all difference values, and set the sudden change threshold to μ+2σ; Traverse all the difference values of adjacent sliding windows k, and compare them with the sudden change threshold, and the upper limit of the previous sliding window corresponding to the first difference value exceeding the sudden change threshold is the satellite elevation angle grouping point.

8. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 1, characterized in that: The construction process of the multipath effect risk prediction model is: Extract the continuous wave data of the period of positioning error of normal rescue and target rescue; Identify the wave crest by extreme value method, that is, the highest point between two adjacent wave troughs, and record the wave height of each wave, that is, the vertical distance between the wave crest and the previous wave trough; Arrange the wave height in descending order, calculate the average value of the top third wave height, and obtain the significant wave height; Build a multi-path effect risk prediction model, input the current sea wave parameters and the position parameters of the fishing boat to be rescued, and output the probability of positioning error caused by multi-path effect; Dependent variable Y is whether multi-path effect occurs, Y = 1 represents target rescue, and Y = 0 represents normal rescue; Extract sea wave parameters and fishing boat position parameters from historical data, wherein the sea wave parameters include significant wave height, sea wave period, and wave direction-azimuth angle difference, and the fishing boat position parameters include satellite elevation angle; Build an interaction term of sea wave parameters and fishing boat position parameters, and the interaction term is the wave height-elevation angle coordination term: X = H x (30° - θ 卫 ), where H is the wave height, θ 卫 is the satellite elevation angle; Take the sea wave parameters, fishing boat position parameters, and wave height-elevation angle coordination term as input features Calculate the Pearson correlation coefficient between the input features and the dependent variable Y, retain features whose Pearson correlation coefficient is greater than or equal to a preset correlation coefficient, and obtain the feature set [x1, x2, ..., x]. i ], where x i For the i-th feature; With the feature set [x1, x2,..., x i ] as the independent variable, the probability prediction formula is constructed: ； Wherein, β0 is a constant term, and β1~β4 are regression coefficients; Use the training set to fit the regression coefficients, and finally obtain the multi-path effect risk prediction model.

9. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 8, characterized in that: The judgment method of whether the current rescue has multi-path effect risk is: Input the sea wave parameters of the current rescue and the position of the fishing boat to be rescued into the multi-path effect risk prediction model, and output the probability of multi-path effect in the positioning of the fishing boat to be rescued; If the probability is greater than or equal to the preset probability, the current rescue has multi-path effect risk.

10. The fishing boat rescue method based on the Beidou intelligent navigation positioning system according to claim 1, characterized in that: The dynamic correction method of the Beidou positioning result is: Determine the satellite elevation angle according to the position of the fishing boat to be rescued in the current Beidou positioning, judge the satellite elevation angle group, input the wave height into the multi-path error compensation model, and obtain the positioning error size; Calculate the wave direction-azimuth angle, input it into the multi-path error compensation model, and obtain the error direction; After converting the error direction into radian, input the positioning error size and error direction into the azimuth-angle-distance conversion latitude and longitude formula, and obtain the corrected position coordinates of the fishing boat to be rescued.

Images