Ship navigation system based on compass direction finding and satellite fusion

Abstract

The present application relates to the field of marine navigation technology of ship, specifically relates to a ship marine navigation system based on compass direction finding and satellite fusion, comprising a data acquisition module, a disturbance correlation analysis module, a pseudo course determination module, a navigation credibility evaluation module, a dynamic fusion weight adjustment module and a stable heading output module;Through collecting the compass heading data, satellite positioning data and ship body attitude data of the target ship, the correlation between the heading change and the roll disturbance is analyzed, and the pseudo heading change caused by the sea state disturbance is identified in combination with the track change state;Based on the disturbance correlation result, the credibility of the compass navigation and satellite navigation is dynamically evaluated, and the fusion weight of different navigation data is adaptively adjusted to generate stable heading information, which is beneficial to reduce the influence of ship body attitude disturbance on the heading measurement result under complex sea state conditions, and improve the continuous stability and navigation reliability of ship heading navigation.

Description

Technical Field

[0001] This invention relates to the field of marine navigation technology, specifically to a marine navigation system based on compass direction finding and satellite fusion. Background Technology

[0002] With the development of intelligent navigation and autopilot technologies for ships, the requirements for the accuracy of heading measurement and the stability of navigation in ship navigation systems are constantly increasing. Existing ship navigation systems often use compass navigation equipment to obtain ship heading information and combine it with satellite navigation equipment to obtain ship position, track and speed information in order to achieve ship heading navigation and track control. However, under complex sea conditions, the hull is easily affected by attitude disturbances such as roll, pitch and wave impact, which causes periodic fluctuations in the heading information output by the compass, thereby affecting the stability of the ship heading measurement results.

[0003] In existing technologies, most solutions to the compass heading fluctuation problem use methods such as filtering compensation, inertial correction, or fixed-weight fusion to process navigation data. However, these methods usually only smooth or fuse the heading data itself, lacking analysis of the correlation between ship attitude disturbances and heading changes. It is difficult to distinguish whether the current heading change is caused by the target ship's actual turning or by pseudo-heading changes caused by sea state disturbances. When sea state disturbances are strong, if the disturbed compass data is still directly used in navigation fusion, it will lead to problems such as heading deviation, navigation jitter, and even instability in autopilot control.

[0004] Furthermore, most existing navigation fusion schemes employ fixed fusion strategies or static weight allocation methods, which cannot dynamically adjust the participation weights of different navigation data based on the current sea state disturbance state and the reliability of navigation data. This results in poor adaptability of navigation fusion results under complex sea state conditions. Therefore, how to effectively identify pseudo-heading changes caused by sea state disturbances under complex sea state conditions and dynamically adjust the fusion strategy based on the degree of disturbance to navigation data in order to improve the continuous stability and reliability of ship heading navigation has become an urgent technical problem to be solved in the field of ship navigation. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a ship navigation system based on compass direction finding and satellite fusion, which can effectively solve the problem of poor navigation stability under complex sea conditions in the background technology.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a ship navigation system based on compass orientation finding and satellite fusion, comprising:

[0007] The data acquisition module is used to collect compass heading data, satellite positioning data, and hull attitude data of the target vessel during its navigation process, and to perform time synchronization processing to construct a correlation dataset corresponding to the target vessel's current navigation state and sea state disturbance state.

[0008] The disturbance correlation analysis module is used to extract heading change rate, roll frequency and track change information based on the correlation dataset, analyze the periodic correspondence between heading change and ship attitude disturbance, and determine whether the current heading change is related to the ship attitude disturbance by combining the actual track change status of the target ship.

[0009] The navigation reliability assessment module is used to determine whether the current course change is a pseudo-course change caused by sea state disturbance based on the correlation analysis results between course change and ship attitude disturbance and the track change status. Based on the pseudo-course determination results, it analyzes the degree of impact of sea state disturbance on compass heading data and satellite navigation data, so as to construct the corresponding navigation reliability model.

[0010] The dynamic fusion weight adjustment module is used to dynamically adjust the fusion weight of compass heading data and satellite navigation data based on the navigation reliability model, and to perform fusion correction on the compass heading data and satellite navigation data based on the adjusted fusion weight.

[0011] The stable heading output module is used to generate stable heading information corresponding to the target vessel based on the fused navigation results, and to use the stable heading information for the target vessel's heading navigation, track maintenance and autopilot control.

[0012] Preferably, the specific process of data collection and extraction is as follows:

[0013] The compass heading data includes the target vessel's current heading angle, the satellite positioning data includes the target vessel's position coordinates, track direction, and speed, and the hull attitude data includes the target vessel's roll angle, pitch angle, and rate of change of attitude.

[0014] The real-time collected compass heading data, satellite positioning data, and ship attitude data are time-synchronized under a unified time reference to determine the temporal correspondence between heading change data, satellite trajectory data, and ship attitude data at each moment, and to construct a correlation dataset between the target ship's current navigation state and sea state disturbance state.

[0015] Based on the compass heading data after time synchronization processing, the heading change rate within the corresponding time period is extracted. Based on the ship attitude data, the roll frequency within the corresponding time window is extracted. Based on the satellite positioning data, the track change information of the target ship within a continuous time period is extracted.

[0016] The heading change rate, roll frequency, and track change information are correlated to establish a basis for correlation analysis between heading changes, attitude disturbances, and actual track changes.

[0017] Preferably, the specific process for determining whether the current heading change is related to the ship's attitude disturbance is as follows:

[0018] Based on the rate of change of heading and the roll frequency, a synchronous correlation analysis is performed on the trend of heading change and the trend of roll change within a continuous time window. By comparing the change of direction of the fluctuation and the periodic correspondence between the heading change and the roll disturbance, it is determined whether the current heading change is synchronously related to the ship's attitude disturbance.

[0019] By combining track change information, the actual track change status of the target vessel within the corresponding time window is correlated and verified. By comparing the track direction changes at consecutive times, it is determined whether the actual track of the target vessel remains in a continuous and stable state.

[0020] When the change in heading and the roll disturbance are synchronous, and the actual trajectory of the target vessel remains in a continuous and stable state, it is determined that the current change in heading and the vessel attitude disturbance are related.

[0021] Preferably, the specific process for determining whether the current course change is a pseudo-course change caused by sea state disturbance is as follows:

[0022] The number of associated moments within a continuous time window that satisfy the condition that the direction of the heading change is consistent with the direction of the roll disturbance and the difference between the period of the heading change and the period of the roll disturbance is less than a preset period threshold.

[0023] The proportion of the number of associated moments to the total number of sampling moments in the current time window is used as the perturbation association ratio.

[0024] The number of stable track moments within the current time window whose change in track direction is less than a preset track offset threshold is counted, and the proportion of the number of stable track moments to the total number of sampling moments in the current time window is taken as the track stability ratio.

[0025] The disturbance correlation ratio is compared with the preset correlation ratio threshold, and the track stability ratio is compared with the preset stability ratio threshold.

[0026] When the proportion of disturbance correlation is greater than or equal to the preset correlation proportion threshold, and the proportion of track stability is greater than or equal to the preset stability proportion threshold, the course change within the current time window is determined to be a pseudo course change caused by sea state disturbance; otherwise, the current course change is determined to be a real course change caused by the actual course adjustment of the target vessel.

[0027] Preferably, the specific process for analyzing the impact of the target vessel's current sea state disturbance on compass heading data and satellite navigation data is as follows:

[0028] Based on the heading change rate within the current time window, calculate the standard deviation of the heading change rate sequence within the current time window. Combining the disturbance correlation ratio and the reference benchmark value of the standard deviation, calculate the compass disturbance level index corresponding to the target vessel in the current time window.

[0029] Obtain the average value of the absolute value of the change in track direction within the current time window, and calculate the satellite disturbance level index corresponding to the target ship in the current time window by combining the track stability percentage and the reference value of the average value.

[0030] Preferably, the reliability of different navigation data under the current navigation state is quantified, and a corresponding navigation reliability model is constructed accordingly. The specific process is as follows:

[0031] Based on the compass disturbance level index corresponding to the target vessel in the current time window, and combined with the compass reliability attenuation coefficient and the compass reference reliability, the disturbance level is mapped to the reliability attenuation coefficient, and the compass navigation reliability corresponding to the target vessel in the current time window is calculated.

[0032] By combining the satellite interference level index, satellite baseline reliability, and satellite reliability attenuation coefficient corresponding to the target ship in the current time window, the satellite navigation reliability of the target ship in the current time window is calculated.

[0033] Construct a navigation reliability model for the current time window. The navigation reliability model includes a two-dimensional vector composed of compass navigation reliability and satellite navigation reliability, the reliability ratio of compass navigation reliability to satellite navigation reliability, and the navigation status level.

[0034] Combining the pre-set compass navigation confidence threshold and satellite navigation confidence threshold, the navigation status level is divided into high confidence status level, compass degraded status level, satellite degraded status level, and dual-source degraded status level.

[0035] Preferably, the specific process for analyzing the participation weights of compass heading data and satellite navigation data in the subsequent dynamic fusion process is as follows:

[0036] Based on the confidence ratio output by the navigation confidence model, the theoretical participation weight of the compass heading data is calculated, and the theoretical participation weight of the satellite navigation data is the residual weight.

[0037] The theoretical weights are based on compass heading data and satellite navigation data, and the weight boundaries are restricted according to the navigation status level: if the current status level is high confidence, then no constraints are imposed on the theoretical weights.

[0038] If the current status is either a compass downgrade or a satellite downgrade status, then the theoretical weight of the severely affected party is subject to an upper limit constraint, restricting its participation weight from exceeding the preset downgrade weight upper limit.

[0039] If the current state is a dual-source degradation state, then symmetry constraints are applied to the theoretical weights of both sources.

[0040] Obtain the compass heading data weights for the current time window after level constraint restrictions, combine them with the compass heading data weights for the adjacent previous time window and the preset smoothing coefficient, calculate the compass participation weights after weight smoothing transition processing, and then obtain the corresponding satellite participation weights.

[0041] Preferably, the specific process of weighted fusion of compass heading data and satellite navigation data within the current time window is as follows:

[0042] The satellite trend baseline heading is obtained by performing a moving average of the satellite track headings within the current time window and historical time windows.

[0043] Extract all time-synchronized compass heading angle sampling points within the current time window and calculate the average compass heading for the current time window.

[0044] Subtract the satellite trend reference heading from the average compass heading within the current time window to obtain the instantaneous deviation angle of the compass relative to the satellite trend reference.

[0045] The value of the deviation correction attenuation coefficient is defined as the value corresponding to the satellite participation weight, and the effective deviation correction amount after dynamic weight attenuation of the compass instantaneous deviation angle is calculated.

[0046] The effective deviation correction is added to the satellite trend reference heading to obtain the preliminary fused heading of the target vessel.

[0047] Preferably, the specific process of applying track continuity constraints and historical trend consistency constraints to the initial fused heading to eliminate residual fluctuations and generate stable heading information is as follows:

[0048] The length of the adaptive smooth window is determined based on the navigation confidence status within the current time window. A short time window is used when the navigation confidence is high, and a long time window is used when the navigation confidence is low.

[0049] Based on a determined adaptive smoothing window length, an equal-weighted moving average is performed on the fused navigation results within the current time window and historical time windows to obtain a preliminary stable heading.

[0050] Based on the trajectory change trend corresponding to satellite navigation data, a trajectory continuity constraint is applied to the initial stable heading, and the deviation status between the current initial stable heading and the corresponding trajectory change trend is determined.

[0051] When the deviation exceeds the preset constraint range, the current preliminary stable course is constrained and corrected.

[0052] Based on the fusion navigation results after completing the constraints on track continuity and historical trend consistency, corresponding stable heading information is generated.

[0053] Preferably, the specific process of using the generated stable heading information as the heading input for the target ship's heading navigation, track holding, and autopilot control systems is as follows:

[0054] The generated stable heading information is used as the current heading input of the target ship's heading navigation system, and the heading correction is performed based on the stable heading information to correct the deviation between the target ship's current navigation direction and the preset navigation direction.

[0055] Meanwhile, the stable heading information is used as the heading reference information of the track-keeping control system, and the track-keeping process of the target vessel is continuously adjusted in combination with the current track change status of the target vessel.

[0056] The stable heading information is then sent to the target ship's autopilot control system as a basis for real-time heading control during the target ship's autopilot process.

[0057] The technical solution provided by this invention has the following advantages compared with the known prior art:

[0058] 1. By performing correlation analysis on the heading change rate, roll frequency and track change information, it is helpful to distinguish between the actual turning of the target ship and the pseudo-heading change caused by sea state disturbance, thereby solving the problem that the compass heading is easily affected by the ship's attitude disturbance under complex sea state conditions.

[0059] 2. Based on the exponential decay model, the disturbance level of compass and satellite navigation is quantified, a dynamic reliability model is constructed and the fusion weights are adjusted accordingly. At the same time, a level constraint and a weight smooth transition mechanism are introduced to solve the problem of poor adaptability of traditional fixed weight fusion methods under complex sea conditions, and ensure the accuracy and stability of fused headings under different sea conditions.

[0060] 3. By applying track continuity constraints and historical trend consistency constraints on the basis of weighted fusion, and adopting an adaptive smooth window length that matches the navigation status level, the ship's rapid response capability for actual steering is preserved, while effectively eliminating residual high-frequency fluctuations caused by sea state disturbances. This helps to reduce the operating frequency of the autopilot and the wear of the steering gear, and improve the stability of the ship's navigation.

[0061] 4. The output stable heading data conforms to common ship communication protocols such as NMEA0183 and NMEA2000, and can directly replace the original heading data of the compass and input it into existing ship systems such as electronic charts, autopilot and autopilot, without the need for large-scale modification of the original hardware and software architecture, which helps to reduce system upgrade costs.

[0062] 5. Providing continuous, stable, and high-precision heading information under severe sea conditions such as high winds and waves helps avoid navigation accidents such as track deviation, grounding, and collisions caused by navigation data distortion, and is conducive to ensuring the safety of ships, crew, and cargo. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0064] Figure 1 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0066] The present invention will be further described below with reference to embodiments.

[0067] Please see Figure 1 As shown, a ship navigation system based on compass orientation finding and satellite fusion includes at least:

[0068] The navigation and attitude data acquisition module is used to collect compass heading data, satellite positioning data and hull attitude data of the target vessel in real time, so as to build a data foundation for the target vessel's current navigation status and sea state disturbance status.

[0069] In one specific embodiment, the compass heading data, satellite positioning data, and hull attitude data of the target vessel are collected in real time while it is in motion. The specific process is as follows: the compass heading data includes the target vessel's current heading angle, the satellite positioning data includes the target vessel's position coordinates, track direction, and speed, and the hull attitude data includes the target vessel's roll angle, pitch angle, and attitude change rate.

[0070] The target ship's current heading angle is obtained in real time by a compass sensor installed on the target ship, and its position coordinates, track direction and speed are obtained in real time by a satellite navigation module. At the same time, the target ship's roll angle, pitch angle and attitude change rate (referring to the degree of change of the target ship's attitude parameters per unit time) are obtained in real time by an attitude sensor.

[0071] The collected compass heading data, satellite positioning data, and ship attitude data are time-synchronized under a unified time reference to determine the temporal correspondence between heading change data, satellite trajectory data, and ship attitude data at each moment. After time synchronization is completed, the compass heading data, satellite positioning data, and ship attitude data at the corresponding moment are associated and stored to construct an associated dataset of the target ship's current navigation state and sea state disturbance state.

[0072] It should be noted that a satellite navigation module refers to a navigation equipment module used to receive satellite navigation signals and output the target ship's position, speed, and track information.

[0073] The time synchronization processing refers to the time alignment processing of compass heading data, satellite positioning data and ship attitude data collected at different times according to a unified time reference, so that the various types of data form corresponding data relationships at the same time, so as to ensure that subsequent steps can analyze the correlation between heading changes, satellite trajectory changes and ship attitude disturbances based on the same time dimension.

[0074] The disturbance correlation analysis module is used to extract the heading change rate from the compass heading data, the roll frequency from the ship attitude data, and the track change information from the satellite positioning data based on the correlation dataset constructed in step one.

[0075] The correspondence between the heading fluctuation period characterized by the heading change sequence and the roll disturbance period characterized by the roll frequency is analyzed, and combined with the track change information at continuous moments, it is determined whether the current heading change is related to the ship's attitude disturbance.

[0076] When the target vessel's current course change is related to the vessel's attitude disturbance, determine whether the current course change is a pseudo-course change caused by sea state disturbance.

[0077] In one specific embodiment, the specific process of extracting the heading change rate from the compass heading data, the roll frequency from the hull attitude data, and the track change information from the satellite positioning data is as follows: the compass heading data of consecutive moments after time synchronization processing is arranged in a time series, and the heading change rate within the corresponding time period is calculated based on the change difference between the heading angles of adjacent moments to characterize the current heading change trend of the target ship.

[0078] Periodic fluctuation analysis is performed on the roll angle data at continuous time intervals to extract the repetitive fluctuation period of the roll angle within a continuous time window. This repetitive fluctuation period is then used as the roll frequency of the current target ship to characterize the periodic impact of sea state disturbances on the attitude of the target ship.

[0079] After extracting the heading change rate and roll frequency, trajectory correlation analysis is performed on satellite positioning data at continuous time intervals. By comparing the changes in trajectory direction and the degree of trajectory deviation between adjacent time position coordinates, the trajectory change information of the target ship in continuous time intervals is extracted to characterize the continuous change state of the target ship's actual navigation trajectory.

[0080] By correlating the heading change rate, roll frequency, and track change information, a basis for correlation analysis between heading changes, attitude disturbances, and actual track changes is established.

[0081] In one specific embodiment, the process of determining whether the current heading change is related to the ship's attitude disturbance is as follows: based on the constructed associated dataset and the extracted heading change rate, roll frequency and track change information, the heading change trend and the roll change trend of the target ship within a continuous time window are synchronously correlated.

[0082] The heading change rate at consecutive moments is used to construct a heading change sequence in chronological order, and the roll angle change data at corresponding moments are used to construct a roll disturbance sequence. By comparing the consistency of the fluctuation direction and the correspondence of the fluctuation period between the heading change sequence and the roll disturbance sequence within a continuous time window, it is assessed whether the current heading change changes synchronously with the roll disturbance.

[0083] If the direction of the heading change corresponds to the direction of the roll disturbance continuously within a continuous time window, and the period of the heading change is consistent with the period of the roll disturbance, then it is determined that there is a disturbance correlation between the heading change of the current target vessel and the hull attitude disturbance; otherwise, it is determined that there is no disturbance correlation between the heading change of the current target vessel and the hull attitude disturbance.

[0084] Based on the extracted track change information, the actual track change status of the target vessel within the corresponding time window is verified. The change in track direction between consecutive adjacent moments is calculated (first by subtraction, then by absolute value calculation of the result), and the change in track direction is compared with a preset track offset threshold. When the change in track direction at each adjacent moment within a consecutive time window is less than the preset track offset threshold, it is determined that the actual track of the target vessel remains in a continuous and stable state.

[0085] Based on the above disturbance correlation analysis results and track correlation verification results, it is determined that the current course change is more likely to be caused by attitude change due to hull roll disturbance, rather than a course change caused by the actual turning of the target vessel.

[0086] It should be noted that the specific process for setting the preset track offset threshold is as follows: The threshold is set based on historical track change data of the target vessel under normal straight-line navigation conditions. Specifically, under stable sea conditions and without actual turning of the target vessel, the track direction change is continuously collected at adjacent moments within multiple time windows, and the track direction change range corresponding to each time window is statistically analyzed. The maximum value of the track direction change under normal navigation conditions is used as the base offset value. A preset safety margin (statistically determined based on satellite track fluctuation data and attitude disturbance data of the target vessel under different sea conditions) is added to the base offset value to obtain the preset track offset threshold. For example, if the maximum track direction change at adjacent moments is 1.8 degrees when the target vessel is continuously navigating under stable sea conditions, a safety margin of 0.7 degrees is added to this value, and 2.5 degrees is determined as the preset track offset threshold.

[0087] In one specific embodiment, the specific process of determining whether the current heading change is a pseudo heading change caused by sea state disturbance is as follows: count the number of associated moments within a continuous time window that satisfy the heading change direction being consistent with the roll disturbance direction and the period difference (referring to the absolute value of the period difference) between the heading change period and the roll disturbance period being less than a preset period threshold.

[0088] A correlated moment refers to a moment within a continuous time window when the heading change feature and the roll disturbance feature at a certain sampling moment simultaneously meet a preset correlation condition. For example, at a certain sampling moment, if the current heading change direction is consistent with the current roll disturbance direction, and the period difference between the current heading change period and the current roll disturbance period is less than a preset period threshold, then the sampling moment is determined as a correlated moment.

[0089] The proportion of the number of associated moments to the total number of sampling moments in the current time window is used as the perturbation association ratio.

[0090] The number of stable track moments within the current time window whose change in track direction is less than a preset track offset threshold is counted, and the proportion of the number of stable track moments to the total number of sampling moments in the current time window is taken as the track stability ratio.

[0091] The disturbance correlation ratio is compared with the preset correlation ratio threshold, and the track stability ratio is compared with the preset stability ratio threshold.

[0092] When the proportion of disturbance correlation is greater than or equal to the preset correlation proportion threshold, and the proportion of track stability is greater than or equal to the preset stability proportion threshold, the course change within the current time window is determined to be a pseudo course change caused by sea state disturbance; otherwise, the current course change is determined to be a real course change caused by the actual course adjustment of the target vessel.

[0093] It should be noted that the preset period threshold is determined based on the historical navigation data of the target vessel under different sea state conditions. The distribution of the period difference between the heading change period and the roll disturbance period under sea state disturbance scenarios is statistically analyzed, and the degree of synchronous correlation between heading change and roll disturbance under different period difference conditions is analyzed. For example, when the period difference is concentrated less than 0.3 seconds in most sea state disturbance scenarios, the preset period threshold is set to 0.3 seconds.

[0094] The preset correlation percentage threshold is used to represent the lower limit of the correlation between heading changes and roll disturbances within the current time window. Its specific value is determined statistically based on the historical navigation data of the target vessel under different sea state conditions. Specifically, multiple historical time windows in which the target vessel exhibits significant roll disturbances during stable straight navigation are selected. The percentage of correlated moments within each time window that meet the criteria of "the direction of heading change is consistent with the direction of roll disturbance" and "the period difference between the period of heading change and the period of roll disturbance is less than the preset period threshold" is statistically analyzed. For example, when the percentage of correlated moments in most sea state disturbance scenarios is concentrated between 70% and 90%, the preset correlation percentage threshold is set to 75% to ensure that the current heading change is determined to have an attitude disturbance correlation only when there is a high degree of synchronous correlation between the heading change and the roll disturbance.

[0095] The preset stability percentage threshold is used to indicate the degree to which the actual track of the target vessel remains stable within the current time window. Its specific value is determined by comparing the track change data of the target vessel in normal navigation and actual turning states. Specifically, the percentage of stable track moments that meet the condition of "track direction change less than preset track offset threshold" is statistically analyzed in both straight stable navigation and actual turning states, and the distribution differences of stable track moment percentages in different navigation states are analyzed. For example, in straight stable navigation, the percentage of stable track moments is higher than 80%, while in actual turning state, the percentage of stable track moments is lower than 60%. Therefore, the preset stability percentage threshold is set to 80%.

[0096] By performing correlation analysis on the rate of change of heading, roll frequency, and track change information, it is helpful to distinguish between the actual turning of the target vessel and the pseudo-heading changes caused by sea state disturbances, thereby solving the problem that compass heading is easily affected by hull attitude disturbances under complex sea state conditions.

[0097] The navigation reliability assessment module is used to analyze the impact of the target vessel's current sea state disturbance on compass heading data and satellite navigation data based on the pseudo-heading determination results and the heading change rate, attitude change rate, and track change information within a continuous time window.

[0098] By combining the proportion of disturbance correlation, the proportion of track stability, the rate of change of heading, and the rate of change of attitude, the corresponding compass navigation reliability and satellite navigation reliability are calculated respectively, and the reliability of different navigation data under the current navigation state is quantified.

[0099] Based on the calculated compass navigation reliability and satellite navigation reliability, a navigation reliability model for the current target ship is constructed, and the participation weights of compass heading data and satellite navigation data in the subsequent dynamic fusion process are analyzed.

[0100] In one specific embodiment, the process of analyzing the impact of the target vessel's current sea state disturbance on compass heading data and satellite navigation data is as follows: Based on the rate of change of heading within the current time window, calculate the standard deviation within the current time window. Combined with the proportion of disturbance correlation Reference baseline value for standard deviation within the current time window Calculate the compass disturbance level index corresponding to the target ship in the current time window. : .

[0101] Get the average value of the absolute value of the change in trajectory direction within the current time window. Combined with the proportion of stable flight paths and the reference base value of this average value Then, calculate the satellite interference level index corresponding to the target ship in the current time window. : .

[0102] It should be noted that, The acquisition process is as follows: taking the current time window length of 10 seconds as an example, the system sampling frequency is 10Hz, that is, the time window contains a total of 100 sampling moments. For each sampling moment, the corresponding heading change rate value is obtained from the extracted heading change rate sequence. This value is calculated based on the difference in heading angle between adjacent moments. The standard deviation of the heading change rate sequence within the current time window is calculated.

[0103] The acquisition process is as follows: Historical navigation data segments of the target vessel in sea state 3 or below, maintaining a straight and constant speed, are pre-selected. Using the same time window length and sampling frequency as the current time window, the standard deviation of the rate of change of course within each time window is calculated. Then, the statistical median or mean of the standard deviations of all stationary windows is taken as the... The value of ; for example, in multiple straight-line navigation tests in calm sea conditions, the median of the standard deviation of each steady window was statistically obtained to be 0.2 degrees / second, that is, the set value is 0.2 degrees / second. The value is 0.2 degrees per second, which serves as an objective benchmark for distinguishing between "normal fluctuations" and "intensified disturbances".

[0104] The acquisition process is as follows: Taking 100 sampling moments within the current 10-second time window as an example, for each pair of adjacent sampling moments within the window, the absolute value of the corresponding change in track direction is obtained from the extracted track change information, resulting in 100-1 valid values. The arithmetic mean of these values ​​is then calculated to obtain the average value of the absolute value of the change in track direction within the current time window.

[0105] The acquisition process is as follows: Historical navigation data segments of the target vessel in sea state 3 or below, maintaining a straight and constant speed without actual turning operations, are pre-selected. Using the same time window length and sampling frequency as the current time window, the absolute value of the change in track direction within each time window is extracted, and its window average is calculated. Then, the statistical median or mean of the average values ​​corresponding to all stable windows is taken as the data. The value of .

[0106] In one specific embodiment, the reliability of different navigation data under the current navigation state is quantified, and a corresponding navigation reliability model is constructed accordingly. The specific process is as follows: based on the exponential decay model, the compass disturbance level index corresponding to the target ship in the current time window is used... Combined with the compass reliability attenuation coefficient and the reliability of compass reference The degree of disturbance is mapped to a credibility attenuation coefficient: The compass navigation reliability of the target ship in the current time window is calculated. .

[0107] Combined with the satellite interference level index corresponding to the target ship in the current time window Satellite benchmark credibility Satellite credibility attenuation coefficient and satellite interference level indicators Calculate the satellite navigation reliability of the target ship within the current time window. : .

[0108] After obtaining the compass navigation reliability and satellite navigation reliability respectively, a navigation reliability model for the current time window is constructed. The navigation reliability model includes a two-dimensional vector composed of compass navigation reliability and satellite navigation reliability, the reliability ratio of compass navigation reliability to satellite navigation reliability, and the navigation status level.

[0109] The navigation status levels include high reliability status level, compass degraded status level, satellite degraded status level, and dual-source degraded status level, which are evaluated in combination with preset compass navigation reliability thresholds and satellite navigation reliability thresholds.

[0110] Specifically, when the compass navigation reliability is greater than or equal to the compass navigation reliability threshold and the satellite navigation reliability is greater than or equal to the satellite navigation reliability threshold, the target vessel is determined to be in a high reliability status. When the compass navigation reliability is less than the compass navigation reliability threshold and the satellite navigation reliability is greater than or equal to the satellite navigation reliability threshold, the target vessel is determined to be in a compass downgraded status. When the compass navigation reliability is greater than or equal to the compass navigation reliability threshold and the satellite navigation reliability is less than the satellite navigation reliability threshold, the target vessel is determined to be in a satellite downgraded status. When the compass navigation reliability is less than the compass navigation reliability threshold and the satellite navigation reliability is less than the satellite navigation reliability threshold, the target vessel is determined to be in a dual-source downgraded status.

[0111] It should be noted that the exponential decay model is based on the exponential decay mathematical model in the fields of reliability engineering and signal processing. This model is often used to describe the nonlinear decay law of system performance or signal strength as external stress or disturbance accumulates. In this scheme, the model is used to map the calculated compass disturbance level index and satellite disturbance level index to the corresponding compass navigation reliability and satellite navigation reliability, respectively, so that the reliability decreases as the sea state disturbance intensifies, and the decay accelerates when the disturbance is severe.

[0112] The process of determining the reliability of the compass reference is as follows: Under low sea state and stable straight-line navigation conditions, compass heading data is collected and compared with high-precision reference heading (such as fiber optic compass or differential GNSS heading). The root mean square error of the heading deviation is calculated, and the confidence level corresponding to the root mean square error is taken as the reference reliability. For example, when the root mean square error of the heading deviation is 0.5°, the corresponding reference reliability is taken as 0.95.

[0113] The process of determining the reliability of satellite references is as follows: Under conditions of open water, no obstruction, and stable ship navigation, satellite positioning data is collected, and the stability of its trajectory and positioning accuracy (such as HDOP value and positioning standard deviation) are statistically analyzed. The reliability is comprehensively evaluated based on the nominal accuracy and measured accuracy of the receiver. For example, when the positioning standard deviation is less than 1 meter, the corresponding reference reliability is taken as 0.98.

[0114] The process of determining the compass reliability decay coefficient is as follows: Select historical navigation data of the target vessel under various sea state levels, calculate the compass disturbance level index and the corresponding compass heading deviation in each time window, take the compass disturbance level index as the independent variable and the normalized value of the heading deviation as the dependent variable, and determine the compass reliability decay coefficient that best matches the reliability decay curve with the measured deviation trend through exponential function fitting. The value of the compass reliability decay coefficient is usually between 0.5 and 1.5. The larger the value, the more sensitive the compass reliability is to sea state disturbances.

[0115] The process of determining the satellite reliability attenuation coefficient is as follows: Select historical navigation data of the target ship under various sea state levels, calculate the satellite disturbance level index and the corresponding satellite track deviation (the deviation between the satellite track direction and the ship's actual track direction) in each time window; with the satellite disturbance level index as the independent variable and the normalized value of the track deviation as the dependent variable, determine the value of the satellite reliability attenuation coefficient by fitting an exponential function. The value of the satellite reliability attenuation coefficient is usually between 0.3 and 1.

[0116] The compass navigation reliability threshold and satellite navigation reliability threshold are determined based on the statistical correlation between the reliability calculation results of the target vessel under different sea state conditions in historical navigation data and the corresponding actual navigation performance. Specifically, each historical time window of the target vessel in sea state disturbance scenario and normal navigation scenario is selected, and the numerical distribution range of compass navigation reliability and satellite navigation reliability under sea state disturbance scenario and normal navigation scenario is statistically analyzed. The correspondence between the reliability value and the conclusion of "whether the navigation data meets the normal acceptance requirements" is statistically analyzed, and the reliability value that can distinguish between high reliability state and degraded state is determined as the corresponding threshold.

[0117] Based on the exponential decay model, the disturbance levels of compass and satellite navigation are quantified. A dynamic reliability model is constructed and the fusion weights are adjusted accordingly. At the same time, a level constraint and a weight smooth transition mechanism are introduced to solve the problem of poor adaptability of traditional fixed weight fusion methods under complex sea conditions, and ensure the accuracy and stability of fusion headings under different sea conditions.

[0118] In one specific embodiment, the process of analyzing the participation weights of compass heading data and satellite navigation data in the subsequent dynamic fusion process is as follows: based on the confidence ratio output by the navigation confidence model... Through the formula: The theoretical participation weights of the compass heading data were calculated. The theoretical participation weight of satellite navigation data is then... : .

[0119] Based on the theoretical participation weights of compass heading data and satellite navigation data, the weight boundaries are restricted according to the confidence level. If the current confidence level is high, no constraints are imposed on the theoretical weights. If the current confidence level is both compass and satellite navigation data, an upper limit constraint is set on the theoretical weight of the more severely disturbed party, limiting its participation weight to no more than the preset upper limit of the degraded weight. If the current confidence level is dual-source degraded, symmetry constraints are applied to the theoretical weights of both parties.

[0120] Obtain the theoretical compass heading data weights for the current time window after applying level constraints. Combined with the compass heading data weights corresponding to the previous time window adjacent to the current time window after the level constraint restrictions. , The index number corresponds to each time window. The value of is a positive integer. Combined with a preset smoothing coefficient, the weight of the compass participation after weight smoothing transition is calculated. : Then, the satellite participation weights after weight smoothing transition processing are calculated. ( ).

[0121] It should be noted that the specific process of setting an upper limit constraint on the theoretical weight of the severely disturbed party is as follows: For example, if the reliability model determines that the current time window is in a single-source degradation state, and the severely disturbed party is the compass, the theoretical weight of the compass may still be 0.45 according to the theoretical weight formula. However, the system determines that the compass is already in the unreliable range. If it still participates in the fusion with a weight of 0.45, the pseudo-heading fluctuation of its output will cause interference. Therefore, an upper limit constraint is imposed on the weight of the compass, limiting its participation weight to within the preset degradation weight upper limit (e.g., 0.30). Even if its theoretical weight is higher than this value, the final constrained weight will only be 0.30, and the satellite weight will be adjusted to 0.70 accordingly.

[0122] The symmetry constraint refers to the following: For example, if the current time window is determined to be in a dual-source degradation state, with a compass confidence of 0.35 and a satellite confidence of 0.4, both below the low confidence threshold, the theoretical weight of the compass is approximately 0.467 and the theoretical weight of the satellite is approximately 0.533, with a weight difference of only 0.066. However, in another dual-source degradation scenario, the compass confidence drops to 0.20 while the satellite confidence remains at 0.48. In this case, the theoretical weights are 0.294 and 0.706 respectively, with a weight difference of 0.412. If this set of weights is directly adopted, the fusion result will overly rely on satellite data, while the satellite is already in a degradation state. The symmetry constraint is applied by forcibly limiting the participation weights of both parties to a level not lower than a preset lower limit (such as 0.45).

[0123] The process of setting the upper limit of the preset degradation weight is as follows: it is determined based on the historical test data of the target ship in the single-source degradation scenario; in the flight segment where the compass is known to be severely disturbed but the satellite is normal, fusion simulation is performed with different compass weights, the curve of the fusion heading error changing with the compass weight is statistically analyzed, and the compass weight value corresponding to the inflection point where the error begins to deteriorate significantly is determined as the upper limit of the degradation weight.

[0124] The process of setting the smoothing coefficient is as follows: before system deployment, it is determined through offline calibration. Historical navigation data or simulation data of the target ship under typical changing sea state scenarios are selected. Weight smoothing tests are carried out with different smoothing coefficients (such as 0.1 in the range of 0.1 to 0.9 with a step size of 0.1). A curve between the weight transition smoothness index and the credibility change delay index is constructed. The smoothing coefficient that makes the two optimal is selected as the set value.

[0125] By applying track continuity constraints and historical trend consistency constraints on the basis of weighted fusion, and adopting an adaptive smooth window length that matches the navigation status level, the ship's rapid response capability for actual steering is preserved, while effectively eliminating residual high-frequency fluctuations caused by sea state disturbances. This helps to reduce the operating frequency of the autopilot and the wear of the steering gear, and improve the stability of the ship's navigation.

[0126] The dynamic fusion weight adjustment module, based on the dynamic credibility weight, performs heading fusion correction and stable heading generation. It calls the compass participation weight and satellite participation weight output in step three to perform weighted fusion of the compass heading data and satellite navigation data within the current time window to obtain the preliminary fused heading.

[0127] By applying track continuity constraints and historical trend consistency constraints to the initial fused heading, stable heading information is generated after eliminating residual fluctuations.

[0128] In one specific embodiment, the process of weighted fusion of compass heading data and satellite navigation data within the current time window is as follows: obtain the satellite track headings within the current time window and each historical time window, perform a moving average, and obtain the satellite trend reference heading.

[0129] Specifically, although satellite heading may experience short-term jumps due to ship turbulence, its long-term trend remains stable. Therefore, the arithmetic mean of the satellite heading averages of the current time window and each historical time window is calculated as the satellite trend benchmark heading for the current window; typically, 3 to 10 historical time windows are used.

[0130] Based on the satellite trend reference heading and the average compass heading within the current time window, subtract the satellite trend reference heading from the average compass heading within the current time window to obtain the instantaneous deviation angle of the compass relative to the satellite trend reference. .

[0131] Based on the output compass participation weight and satellite participation weight, define the deviation correction attenuation coefficient. The value is taken as the value corresponding to the satellite's participation weight, and then the effective deviation correction amount of the compass instantaneous deviation angle after dynamic weight attenuation is calculated. : .

[0132] By combining the effective deviation correction and the satellite trend reference heading, the sum of the two is calculated to obtain the preliminary fused heading of the target vessel.

[0133] It should be noted that the process of obtaining the average compass heading within the current time window is as follows: extract all time-synchronized compass heading angle sampling points within the current time window from the constructed associated dataset, and take the arithmetic mean of these sampling points to obtain the average compass heading for the current time window.

[0134] In one specific embodiment, the process of applying track continuity constraints and historical trend consistency constraints to the initial fused heading, and generating stable heading information after eliminating residual fluctuations, is as follows:

[0135] Define the adaptive smooth window length for the current time window. When in a high trust level, Values When the compass or satellite is in a downgraded status level, Values When in a dual-source downgrade state, Values .

[0136] (First Length) (Second length) and (Third length) represents the smoothing window length corresponding to the three levels, with the dimension being the number of time windows, satisfying... The three values ​​are determined based on the following: Take half of the time window number corresponding to the minimum turning time of the ship to ensure that there is no significant lag during actual turning; The number of time windows corresponding to integer multiples of the average roll period is taken to ensure that multiple complete roll periods can be covered during dual-source degradation, thereby effectively smoothing pseudo-heading fluctuations. Take the middle value between the two.

[0137] Based on a determined adaptive window length For the present and history An equal-weighted moving average is applied to the initial fused headings within each time window to obtain an initial stable heading. (Dimension: degree).

[0138] Using the satellite trend reference heading as the external anchoring reference, a bounded constraint correction is applied to the initially stable heading. Specifically, the track constraint deviation is defined. To initially stabilize the difference between the heading and the satellite trend reference heading, a pre-set constraint relaxation threshold is used. Correction is only applied when the deviation exceeds this threshold, and the constraint correction rules are as follows: ,in It is represented as the final stable heading (in degrees) after correction for track continuity constraints. This is a sign function that indicates the sign of the track constraint deviation. It returns +1 when the deviation is positive, -1 when it is negative, and 0 when it is zero.

[0139] It should be noted that the process of setting the constraint relaxation threshold is as follows: under the conditions of calm sea conditions, the target ship maintaining a straight and stable course and both the compass and the satellite being in normal working condition, historical deviation data between the initial stable course and the satellite trend reference course are continuously collected within multiple time windows, the distribution range of the deviation is statistically analyzed, and the 95th percentile value of the distribution range is taken as the constraint relaxation threshold.

[0140] The stable heading output module uses the generated stable heading information as the heading input for the target ship's heading navigation, track holding and autopilot control systems, so as to achieve continuous stability and navigation reliability of the target ship's heading navigation under complex sea conditions.

[0141] In one specific embodiment, the process of using the generated stable heading information as the heading input for the target ship's heading navigation, track holding, and autopilot control systems is as follows: the final stable heading after track continuity constraint correction is used as the target ship's current heading angle. It is then encapsulated into a heading data message according to protocol formats such as NMEA0183 or NMEA2000 through the ship's data bus or serial communication interface, and directly output to the heading indication interface of the target ship's bridge electronic chart display and information system, radar overlay display, and integrated bridge system, in order to replace the direct output of the original compass heading on the display terminal.

[0142] The final stable heading, corrected by track continuity constraints, is used as the benchmark for calculating the deviation between the set heading and the actual heading. It replaces the original heading angle of the compass without pseudo-heading suppression processing and is input to the heading autopilot control system of the target ship as the heading feedback signal of the autopilot PID controller or model predictive controller.

[0143] The final stable heading, after correction for track continuity constraints, is used as the current heading state input for the target ship's autopilot system. Specifically, it is used to: compare the final stable heading with the azimuth of the current segment of the planned route to calculate the lateral deviation of the track; compare the final stable heading with the target heading of the next turning point to trigger the turning timing judgment; and combine the final stable heading with the current speed to estimate the time to reach the next turning point.

[0144] The output stable heading data conforms to common ship communication protocols such as NMEA0183 and NMEA2000, and can directly replace the original heading data of the compass and input it into existing ship systems such as electronic charts, autopilot and autopilot, without the need for large-scale modification of the original hardware and software architecture, which helps to reduce system upgrade costs.

[0145] Providing continuous, stable, and high-precision heading information under severe sea conditions such as high winds and waves helps avoid navigation accidents such as track deviation, grounding, and collisions caused by navigation data distortion, and is conducive to ensuring the safety of ships, crew, and cargo.

[0146] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A ship navigation system based on compass orientation finding and satellite fusion, characterized in that, include: The data acquisition module is used to collect compass heading data, satellite positioning data and hull attitude data of the target vessel during navigation, and perform time synchronization processing to construct a correlation dataset corresponding to the target vessel's current navigation state and sea state disturbance state. The disturbance correlation analysis module is used to extract heading change rate, roll frequency and track change information based on the correlation dataset, analyze the periodic correspondence between heading change and ship attitude disturbance, and determine whether the current heading change is related to the ship attitude disturbance in combination with the actual track change state of the target ship. The navigation reliability assessment module is used to determine whether the current course change is a pseudo-course change caused by sea state disturbance based on the correlation analysis results between course change and ship attitude disturbance and the track change status. Based on the pseudo-course determination results, it analyzes the degree of impact of sea state disturbance on compass heading data and satellite navigation data in order to construct the corresponding navigation reliability model. The dynamic fusion weight adjustment module is used to dynamically adjust the fusion weight of compass heading data and satellite navigation data based on the navigation credibility model, and to perform fusion correction on the compass heading data and satellite navigation data based on the adjusted fusion weight. The stable heading output module is used to generate stable heading information corresponding to the target vessel based on the fused navigation results, and to use the stable heading information for the target vessel's heading navigation, track maintenance and autopilot control.

2. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 1, characterized in that, The specific process of data collection and extraction is as follows: The compass heading data includes the target vessel's current heading angle; the satellite positioning data includes the target vessel's position coordinates, track direction, and speed; and the hull attitude data includes the target vessel's roll angle, pitch angle, and rate of change of attitude. The real-time collected compass heading data, satellite positioning data, and ship attitude data are time-synchronized under a unified time reference to determine the temporal correspondence between heading change data, satellite trajectory data, and ship attitude data at each moment, and to construct a correlation dataset between the target ship's current navigation state and sea state disturbance state. Based on the compass heading data after time synchronization processing, the heading change rate within the corresponding time period is extracted; based on the ship attitude data, the roll frequency within the corresponding time window is extracted; and based on the satellite positioning data, the track change information of the target ship within a continuous time period is extracted. The heading change rate, roll frequency, and track change information are correlated to establish a basis for correlation analysis between heading changes, attitude disturbances, and actual track changes.

3. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 2, characterized in that, The specific process for determining whether the current course change is related to the ship's attitude disturbance is as follows: Based on the rate of change of heading and the roll frequency, a synchronous correlation analysis is performed on the trend of heading change and the trend of roll change within a continuous time window. By comparing the change of the direction of fluctuation and the periodic correspondence between the heading change and the roll disturbance, it is determined whether the current heading change is synchronously related to the ship's attitude disturbance. By combining track change information, the actual track change status of the target vessel within the corresponding time window is correlated and verified. By comparing the track direction changes at consecutive times, it is determined whether the actual track of the target vessel remains in a continuous and stable state. When the change in heading and the roll disturbance are synchronous, and the actual trajectory of the target vessel remains in a continuous and stable state, it is determined that the current change in heading and the vessel attitude disturbance are related.

4. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 3, characterized in that, The specific process for determining whether the current course change is a pseudo-course change caused by sea state disturbance is as follows: The number of associated moments within a continuous time window that satisfy the condition that the direction of the heading change is consistent with the direction of the roll disturbance and the difference between the period of the heading change and the period of the roll disturbance is less than a preset period threshold. The proportion of the number of associated moments to the total number of sampling moments in the current time window is used as the perturbation association ratio. The number of stable track moments within the current time window whose track direction change is less than a preset track offset threshold is counted, and the proportion of the number of stable track moments to the total number of sampling moments in the current time window is taken as the track stability ratio. The disturbance correlation ratio is compared with the preset correlation ratio threshold, and the track stability ratio is compared with the preset stability ratio threshold. When the proportion of disturbance correlation is greater than or equal to the preset correlation proportion threshold, and the proportion of track stability is greater than or equal to the preset stability proportion threshold, the course change within the current time window is determined to be a pseudo course change caused by sea state disturbance; otherwise, the current course change is determined to be a real course change caused by the actual course adjustment of the target vessel.

5. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 4, characterized in that, The specific process for analyzing the impact of current sea state disturbances on compass heading data and satellite navigation data of the target vessel is as follows: Based on the rate of change of course within the current time window, calculate the standard deviation of the rate of change of course sequence within the current time window. Combine the disturbance correlation ratio and the reference benchmark value of the standard deviation to calculate the compass disturbance level index corresponding to the target vessel in the current time window. Obtain the average value of the absolute value of the change in track direction within the current time window, and calculate the satellite disturbance level index corresponding to the target ship in the current time window by combining the track stability percentage and the reference value of the average value.

6. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 5, characterized in that, The reliability of different navigation data under the current navigation state is quantified, and a corresponding navigation reliability model is constructed accordingly. The specific process is as follows: Based on the compass disturbance level index corresponding to the target vessel in the current time window, combined with the compass reliability attenuation coefficient and the compass reference reliability, the disturbance level is mapped to the reliability attenuation coefficient, and the compass navigation reliability corresponding to the target vessel in the current time window is calculated. By combining the satellite interference level index, satellite baseline reliability, and satellite reliability attenuation coefficient corresponding to the target ship in the current time window, the satellite navigation reliability of the target ship in the current time window is calculated. Construct a navigation reliability model for the current time window. The navigation reliability model includes a two-dimensional vector composed of compass navigation reliability and satellite navigation reliability, the reliability ratio of compass navigation reliability to satellite navigation reliability, and the navigation status level. Combining the pre-set compass navigation confidence threshold and satellite navigation confidence threshold, the navigation status level is divided into high confidence status level, compass degraded status level, satellite degraded status level, and dual-source degraded status level.

7. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 6, characterized in that, The specific process for analyzing the weighting of compass heading data and satellite navigation data in the subsequent dynamic fusion process is as follows: Based on the confidence ratio output by the navigation confidence model, the theoretical participation weight of the compass heading data is calculated, and the theoretical participation weight of the satellite navigation data is the residual weight. The theoretical weights are based on compass heading data and satellite navigation data, and the weight boundaries are restricted according to the navigation status level: if the current status level is high confidence, then no constraints are imposed on the theoretical weights. If the current status is either a compass downgrade or a satellite downgrade, then the theoretical weight of the severely affected party is subject to an upper limit constraint, restricting its participation weight from exceeding the preset downgrade weight upper limit. If the current state is a dual-source degradation state, then a symmetry constraint is applied to the theoretical weights of both sources. Obtain the compass heading data weights for the current time window after level constraint restrictions, combine them with the compass heading data weights for the adjacent previous time window and the preset smoothing coefficient, calculate the compass participation weights after weight smoothing transition processing, and then obtain the corresponding satellite participation weights.

8. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 7, characterized in that, The specific process of weighted fusion of compass heading data and satellite navigation data within the current time window is as follows: The satellite trend baseline heading is obtained by performing a moving average of the satellite track headings within the current time window and historical time windows. Extract all time-synchronized compass heading angle sampling points within the current time window and calculate the average compass heading for the current time window; Subtract the satellite trend reference heading from the average compass heading within the current time window to obtain the instantaneous deviation angle of the compass relative to the satellite trend reference. The value of the deviation correction attenuation coefficient is defined as the value corresponding to the satellite participation weight, and the effective deviation correction amount after dynamic weight attenuation of the compass instantaneous deviation angle is calculated. The effective deviation correction is added to the satellite trend reference heading to obtain the preliminary fused heading of the target vessel.

9. The ship navigation system based on compass orientation finding and satellite fusion as described in claim 8, characterized in that, The specific process of applying track continuity constraints and historical trend consistency constraints to the initial fused heading, and generating stable heading information after eliminating residual fluctuations, is as follows: The length of the adaptive smooth window is determined based on the navigation confidence status within the current time window. A short time window is used when the navigation confidence is high, and a long time window is used when the navigation confidence is low. Based on a determined adaptive smoothing window length, an equal-weighted moving average is performed on the fused navigation results within the current time window and historical time windows to obtain a preliminary stable heading; Based on the trajectory change trend corresponding to satellite navigation data, a trajectory continuity constraint is applied to the initial stable heading, and the deviation status between the current initial stable heading and the corresponding trajectory change trend is determined. When the deviation exceeds the preset constraint range, the current preliminary stable course is constrained and corrected. Based on the fusion navigation results after completing the constraints on track continuity and historical trend consistency, corresponding stable heading information is generated.

10. The ship navigation system based on compass orientation finding and satellite fusion according to claim 9, characterized in that, The specific process of using the generated stable heading information as the heading input for the target ship's heading navigation, track holding, and autopilot control systems is as follows: The generated stable heading information is used as the current heading input of the target ship's heading navigation system, and the heading correction is performed on the deviation between the target ship's current sailing direction and the preset sailing direction based on the stable heading information; Meanwhile, the stable heading information is used as the heading reference information of the track-keeping control system, and the track-keeping process of the target vessel is continuously adjusted in combination with the current track change status of the target vessel. The stable heading information is then sent to the target ship's autopilot control system as a basis for real-time heading control during the target ship's autopilot process.

Images