A method for accurately correcting inertial navigation errors by comprehensively utilizing underwater navigation information

Abstract

The application relates to a kind of inertial navigation error accurate correction methods of comprehensive utilization underwater multiple navigation information, with Doppler log velocity information as reference information, the longitude and latitude error of inertial navigation latitude oscillation error is obtained when two adjacent peaks, based on the longitude and latitude error of two time points, the inertial navigation is implemented single point correction, and the longitude divergence coefficient of inertial navigation is calculated simultaneously.After single point correction, the inertial navigation output compensates the longitude divergence error and is used as the guide information of carrier navigation, the carrier is guided to the area of the fluctuation of marine physical field, and the matching position is obtained by feature matching positioning.Taking the matching position as the starting point, the Doppler log velocity is pushed, and the inertial navigation output is corrected by using the position information, so that the inertial navigation position error can be autonomously and accurately calibrated in underwater scenes where accurate external position information such as satellite navigation cannot be obtained, and the oscillation term in the inertial navigation error can be effectively inhibited in subsequent time, and the information quality of inertial navigation is comprehensively improved.

Description

Technical Field

[0001] This invention belongs to the field of underwater autonomous navigation technology, specifically relating to a method for accurately correcting inertial navigation errors by comprehensively utilizing various underwater navigation information. Background Technology

[0002] Currently, there are limited means of ensuring navigation information in deep-sea and underwater environments, mainly including inertial navigation, Doppler velocity estimation, ocean physical field matching positioning, and underwater acoustic positioning. Each of these single navigation methods has limitations. For example, the position error of an inertial navigation system (INS) accumulates and diverges over time, leading to decreased navigation accuracy over extended periods; Doppler velocity estimation accuracy decreases with increasing distance; ocean physical field matching positioning accuracy is limited by the distribution characteristics of the ocean physical field and is only applicable to areas where the ocean physical field exhibits significant variations; underwater acoustic positioning requires the construction of an underwater acoustic positioning reference network, and a single method is insufficient to meet the long-term autonomous and precise navigation needs in deep-sea and underwater environments. Therefore, the theoretical methods for effectively integrating and utilizing different methods still require further in-depth research. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for accurate correction of inertial navigation errors that comprehensively utilizes multiple underwater navigation information, with a high-precision inertial navigation system as the main component and Doppler velocity estimation and ocean physical field matching positioning as auxiliary components. This method combines the advantages of several underwater navigation and positioning methods to support the realization of long-term, high-precision underwater navigation.

[0004] The technical solution adopted by this invention to solve the above-mentioned technical problems is as follows: a method for accurate correction of inertial navigation errors by comprehensively utilizing multiple underwater navigation information, the method being implemented based on underwater integrated navigation system inertial navigation, Doppler log, and physical field matching positioning equipment, and including the following steps:

[0005] Step 1: Output longitude λ using inertial navigation system ins (t), latitude Let t be time. Using Doppler log velocity estimation information as the reference information, calculate the longitude observation error Δλ when the Earth's periodic oscillation term in the inertial navigation latitude error is at two adjacent peaks t1 and t2. t1 , △λ t2 Latitude observation error based on At time t2, a single-point calibration is performed on the inertial navigation system (INS), and the INS longitude error divergence coefficient k is calculated. λ ;

[0006] Step 2: After single-point calibration of the inertial navigation system at time t2, calculate the carrier navigation reference longitude λ. s (t)=λ ins (t)-k λ (t-t2), latitude The carrier at navigation reference longitude λs (t), latitude Guided by the current, the ship sails to an area where the ocean physical field varies. Using a physical field matching and positioning device, feature matching is performed to obtain the matching longitude λ. m ,latitude

[0007] Step 3: Match longitude λ at time t3 m ,latitude Starting from the point of origin, a Doppler velocimeter velocity measurement and positioning are performed. The inertial navigation system (INS) output is then calibrated to obtain the INS output longitude λ. inso (t), latitude The above three steps are used to correct inertial navigation errors.

[0008] The specific steps of step 1 are as follows:

[0009] 1) Based on the inertial navigation system output longitude λ ins (t), latitude Using the inertial navigation output longitude and latitude at time t0 as initial values, Doppler log velocity-time extrapolation is performed, and the extrapolated longitude λ is output. v (t), latitude Determine the latitude observation error of the inertial navigation system Extract using curve fitting or frequency domain analysis The intermediate period is the precise period T of the 24-hour signal, which is the period of the Earth's periodic oscillation term in the inertial navigation latitude error;

[0010] 2) The inertial navigation latitude observation error reaches its peak at time t1. Calculate the inertial navigation longitude observation error Δλ at time t1. t1 =

[0011] λ ins (t1)-λ v (t1), Latitude observation error Through navigation control, the vehicle is positioned at the peak time t2 of the inertial navigation latitude observation error. Returning to the vicinity of the carrier's location at time t1, calculate the inertial navigation longitude observation error Δλ at time t2. t2 =λ ins (t2)-λ v (t2), latitude observation error

[0012] 3) with λ ins (t2) represents longitude reference information. To obtain latitude reference information, a single-point calibration is performed on the inertial navigation system at time t2;

[0013] 4) Calculate the longitude error divergence coefficient The step 3 involves matching longitude λ at time t3.m ,latitude Starting from the point of origin, a Doppler velocimeter velocity measurement and positioning are performed. The inertial navigation system (INS) output is then calibrated to obtain the INS output longitude λ. inso (t), latitude The method is as follows:

[0014] 1) Match longitude λ at time t3 m ,latitude Starting from the longitude λ, a Doppler speed log is used to calculate the longitude, and the calculated longitude is output as λ. vm (t), latitude

[0015] 2) Inertial navigation output latitude Observation Changes over time were observed Timing begins when the positive or negative peak value is reached. At time t4, after T / 2 time intervals, the Doppler log's velocity-time measurement outputs the longitude λ. vm (t4), latitude Time inertial navigation output longitude λ inso (t)=λ ins (t)+λ vm (t4)-λ ins (t4), latitude This completes the inertial navigation output correction.

[0016] The advantages and positive effects of this invention are: (1) This invention utilizes the characteristic of small relative position error of Doppler velocity measurement and positioning under closed or near-closed carrier track conditions to accurately estimate the amplitude information of the Earth oscillation period term of the inertial navigation latitude error and the constant value term of the inertial navigation latitude error, which are used to solve the reference information for single-point calibration of the inertial navigation, realizing accurate compensation of the inertial navigation latitude error, and also greatly eliminating the part of the Earth oscillation period error of the inertial navigation longitude and heading caused by northward and celestial gyroscope drift. At the same time, the selection of the single-point calibration timing eliminates the part of the Earth oscillation period error of the inertial navigation longitude and heading caused by eastward gyroscope drift. Conventional single-point calibration methods require external absolute position information and cannot suppress the Earth periodic oscillation error. That is, under the condition of no accurate position reference information, this method starts from the mechanism of inertial navigation error generation and uses the error of the reference information during calibration to offset the error caused by the drift of the inertial navigation gyroscope. While realizing accurate compensation of the inertial navigation latitude error, it effectively suppresses the Earth oscillation period error term in the inertial navigation longitude and heading error, breaks through the bottleneck of single-point calibration capability, and achieves the effect that traditional single-point calibration cannot obtain.

[0017] (2) After applying this method to the inertial navigation system (INS), the fluctuation of the INS position error is significantly reduced because the Earth period error term is suppressed. At this time, the INS output position is used to assist physical field matching and positioning, supporting the acquisition of more accurate and reliable matching position information. Based on the acquired matching position information, the INS output is corrected. By combining single-point calibration and output correction, comprehensive and accurate calibration of the INS error is achieved, which can significantly improve the long-term autonomous navigation accuracy and navigation information quality of underwater INS. Attached Figure Description

[0018] Figure 1 This is a flowchart of the inertial navigation error correction method that comprehensively utilizes multiple underwater navigation information according to the present invention. Detailed Implementation

[0019] The present invention will be further described in detail below through specific embodiments. The following embodiments are merely descriptive and not limiting, and should not be used to limit the scope of protection of the present invention.

[0020] like Figure 1 As shown, a method for accurately correcting inertial navigation errors by comprehensively utilizing multiple underwater navigation information includes the following steps:

[0021] Step 1: Output longitude λ using inertial navigation system ins (t), latitude Let t be time. Using Doppler log velocity estimation information as the reference information, calculate the longitude observation error when the Earth's periodic oscillation term in the inertial navigation latitude error is at two adjacent peaks t1 and t2.

[0022] △λ t1 , △λ t2 Latitude observation error based on At time t2, a single-point calibration is performed on the inertial navigation system (INS), and the INS longitude error divergence coefficient k is calculated. λ This single-point calibration method, by controlling the error of latitude reference information and selecting the timing, can achieve accurate latitude calibration even when precise external position reference information is unavailable. Simultaneously, it achieves the effect of suppressing Earth periodic oscillation errors in inertial navigation, an effect that cannot be achieved even with precise external position information calibration. The specific steps are as follows:

[0023] (1) Inertial navigation output longitude λ ins (t), latitude t represents time, and λ represents the inertial navigation longitude at time t0 = 1h (h is an hour, the same below). ins (t0), latitude Using the initial value, perform Doppler speed log estimation for longitude estimation, and output the longitude λ. v (t), latitude Determine the latitude observation error of the inertial navigation system Extract using curve fitting or frequency domain analysis The precise period T of the signal with a medium period of approximately 24 hours is calculated to be T = 23.8h. T is the period of the Earth's periodic oscillation term in the inertial navigation latitude error.

[0024] (2) The inertial navigation latitude observation error reaches its peak at time t1. Calculate the inertial navigation longitude observation error Δλ at time t1. t1 =

[0025] λ ins (t1)-λ v (t1), Latitude observation error Through navigation control, the vehicle is positioned at the peak time t2 of the inertial navigation latitude observation error. Returning to the vicinity of the carrier's location at time t1, calculate the inertial navigation longitude observation error Δλ at time t2. t2 =λ ins (t2)-λ v (t2), latitude observation error

[0026] (3) Inertial navigation latitude observation error Calculate the inertial navigation longitude observation error Δλ at time t1 = 25h, when the Earth's periodic oscillation term reaches its negative peak. t1 =λ ins (t1)-λ v (t1), Latitude observation error Through navigation control, the vehicle is made to return to its location near time t1 at a positive peak time t2 = 25 + 23.8 / 2 = 36.9h after the inertial navigation latitude observation error. Choosing to return to the vicinity of the vehicle's location at time t1 ensures that the vehicle's trajectory from t1 to t2 is closed. At this point, the relative position error calculated by the Doppler log velocity-to-positioning method between t1 and t2 is minimized, approximately a few ten-thousandths of the distance traveled. Therefore, if conditions permit, the vehicle is kept at a suitable low speed to reduce the total distance traveled and further reduce the relative positioning error. The inertial navigation longitude observation error Δλ at time t2 is calculated. t2 =λ ins (t2)-λ v (t2), latitude observation error

[0027] (4) with λ ins (t2) represents longitude reference information. Using latitude reference information, a single-point calibration is performed on the inertial navigation system (INS) at time t2. The error in the latitude reference information during this single-point calibration does not affect the constant term in the subsequent INS latitude error, but it will cause oscillations in the INS longitude, latitude, and heading errors. Using precise latitude reference information will not produce these oscillations. However, since the error of the latitude reference information used for INS calibration in this method is approximately equal to the constant term in the previous INS latitude error, the oscillations caused by this error will precisely cancel out the Earth oscillation period error terms in the INS longitude, latitude, and heading errors originally caused by the equivalent northward and azimuth gyroscope drift. Therefore, this method achieves better results than using precise latitude reference information. The calibration timing is chosen at time t2, which is the peak position of the latitude Earth periodic oscillation error in the inertial navigation system. At this time, the Earth oscillation periodic term in the heading error is at the zero crossing position, and the heading error is close to the constant value error. This will exactly offset the Earth oscillation periodic error term in the longitude, latitude, and heading errors of the inertial navigation system caused by the eastward gyroscope drift. Therefore, in addition to accurately correcting the latitude error after single-point calibration, it will also effectively suppress the Earth oscillation periodic error term in the longitude, latitude, and heading errors of the inertial navigation system, making the changes in inertial navigation error more stable and improving the quality of navigation information.

[0028] (5) Calculate the longitude error divergence coefficient

[0029] Step 2: After single-point calibration of the inertial navigation system at time t2, calculate the carrier navigation reference longitude λ. s (t)=λ ins (t)-k λ (t-t2), latitude The carrier at navigation reference longitude λ s (t), latitude Guided by the current, the ship sails to an area where the ocean physical field varies. Feature matching and positioning are then performed using a physical field matching and positioning device, and the matching longitude λ is obtained at time t3. m ,latitude The ocean physical field includes the ocean gravity field, geomagnetic field, and seafloor topography. A field exhibiting variation along the east-west direction is selected, such as the gravity field. The matching location is determined by calculating the standard deviation of the difference between the gravity measurement sequence and the carrier track's east-west offset over a period of time. The track with the smallest standard deviation corresponds to the matching location. If both the ocean gravity field and geomagnetic field exhibit variation, the corresponding contour lines are found on the gravity map based on the carrier's current gravity measurement value, and on the geomagnetic map based on the carrier's current geomagnetic field measurement value. The intersection of these two contour lines is the matching location.

[0030] Step 3: Match longitude λ at time t3 m ,latitude Starting from the point of origin, a Doppler velocimeter velocity measurement and positioning are performed. The inertial navigation system (INS) output is then calibrated to obtain the INS output longitude λ. inso (t), latitude The specific steps are as follows:

[0031] (1) Matching longitude λ with time t3 m ,latitude Starting from the longitude λ, a Doppler speed log is used to calculate the longitude, and the calculated longitude is output as λ. vm (t), latitude By using a Doppler odometer to measure speed and position, the accuracy of the matching position can be maintained.

[0032] (2) Inertial navigation output latitude Observation Changes over time were observed Timing begins when a positive or negative peak is reached, and continues until... The Doppler log outputs longitude λ at time t4 after the time interval. vm (t4), latitude Time inertial navigation output longitude λ inso (t)=λ ins (t)+λ vm (t4)-

[0033] λ ins (t4), latitude This completes the inertial navigation output correction.

[0034] If we directly use the matching longitude λ m ,latitude Correcting the inertial navigation output will correct the cumulative error of the inertial navigation longitude, but the inertial navigation latitude error may not be corrected, or may even be greater. Therefore, using a Doppler log to maintain the accuracy of the matching position and correcting the inertial navigation output again when the inertial navigation latitude oscillation error crosses zero will reliably correct the inertial navigation latitude error while correcting the inertial navigation longitude error, thus improving the navigation accuracy of the inertial navigation in subsequent time.

[0035] Through the above three steps, in underwater and other application scenarios where external precise position reference information cannot be obtained, the inertial navigation error, including longitude error, latitude error, and heading error, can be accurately corrected autonomously. Furthermore, it achieves the suppression effect of oscillation error terms in longitude, latitude, and heading, which cannot be achieved by using external precise position information calibration.

Claims

1. A method for accurately correcting inertial navigation errors by comprehensively utilizing multiple underwater navigation information, the method being implemented based on an underwater integrated navigation system's inertial navigation, Doppler log, and physical field matching positioning device, characterized in that: Includes the following steps: Step 1: Output longitude from inertial navigation system ,latitude , Using Doppler log velocity estimation information as the reference, the longitude observation error in the inertial navigation latitude error is calculated when the Earth's periodic oscillation term is at two adjacent peaks t1 and t2. , Latitude observation error , ,based on , At time t2, a single-point calibration is performed on the inertial navigation system (INS), and the longitude error divergence coefficient of the INS is calculated. ; Step 2: After single-point calibration of the inertial navigation system at time t2, calculate the carrier navigation reference longitude. ,latitude The carrier uses navigation reference longitude ,latitude Guided by the ocean's physical field, the vessel navigates to an area with fluctuating ocean physical fields. It then uses a physical field matching and positioning device to perform feature matching and positioning to obtain the matching longitude. ,latitude ; Step 3: Match longitude with time t3 ,latitude Starting from this point, a Doppler velocimeter velocity measurement and positioning are performed. The inertial navigation system (INS) output is then calibrated to obtain the INS output longitude. ,latitude ; The above three steps are used to correct inertial navigation errors. 1) Based on the longitude output by the inertial navigation system ,latitude Using the inertial navigation output longitude and latitude at time t0 as initial values, Doppler log velocity-time extrapolation is performed, and the extrapolated longitude is output. ,latitude Calculate the latitude observation error of the inertial navigation system. Extract using curve fitting or frequency domain analysis methods The intermediate period is the precise period T of the 24-hour signal, which is the period of the Earth's periodic oscillation term in the inertial navigation latitude error; 2) The inertial navigation latitude observation error reaches its peak at time t1. Calculate the inertial navigation longitude observation error at time t1. Latitude observation error Through navigation control, the vehicle was able to reach the peak value at time t2 of the inertial navigation latitude observation error. Return to the vicinity of the carrier's location at time t1, and calculate the inertial navigation longitude observation error at time t2. Latitude observation error ; 3) with For longitude reference information, For latitude reference information, in The inertial navigation system is constantly being calibrated at a single point. 4) Calculate the longitude error divergence coefficient Step 3, which involves matching longitude at time t3, ,latitude Starting from the point of origin, use a Doppler velocimeter to measure velocity and determine position. Then, perform output calibration on the inertial navigation system (INS) as appropriate to obtain the INS output longitude. ,latitude The method is as follows: 1) with Matching longitude at all times ,latitude Starting from the longitude, a Doppler speed log is used to extrapolate the coordinates, and the extrapolated longitude is output. ,latitude , ; 2) Inertial navigation output latitude , observation Changes over time were observed Start timing, timer expires of Time Doppler log speed measurement and positioning output longitude ,latitude , Time inertial navigation output longitude ,latitude This means that the inertial navigation output correction is completed.

Images