Absolute relative gravimeter combined measurement method

Abstract

The application discloses a kind of absolute relative gravimeter combination measurement methods, comprising the following steps: S1, with the drift error and its rate of change of strapdown relative gravimeter, the rest error and its rate of change and the error and its rate of change of atomic interference absolute gravimeter as system state;S2, with the difference of gravity anomaly measured by two gravimeters as measurement, construct combination measurement model;S3, by the robust Kalman filtering based on maximum entropy criterion, reliable error state is predicted.The method in the application combines the characteristics of atomic interference absolute gravimeter and strapdown gravimeter, realizes taking advantage of each other, effectively improves the precision and reliability of far sea gravity measurement, faces reality, effectively solves the problems of drift error accumulation of strapdown relative gravimeter and poor dynamic performance of atomic interference absolute gravimeter during far sea measurement, and through the combination measurement of both, the relative gravimeter also avoids needing to return regularly for drift error calibration, improves the precision of gravity measurement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gravity measurement, and in particular to a combined measurement method of absolute relative gravimeter. BACKGROUND

[0002] The ocean, which occupies about 71% of the earth's area, has rich resources of minerals, biology, etc. The measurement of the ocean gravity field helps people to further understand the ocean, explore the ocean, and develop the ocean. As one of the main components of the ocean gravity field, the ocean gravity anomaly plays an important role in the fields of ocean gravity background field mapping, ocean resource detection, underwater matching navigation, etc. This is mainly because the current gravity measurement technology is difficult to support long-time and high-dynamic measurement. The commonly used gravity measurement devices include strapdown relative gravimeters, atomic interference absolute gravimeters, spring relative gravimeters, etc. Among them, the strapdown gravimeter has better dynamic performance, and the atomic gravimeter has no drift characteristics, so they have become widely concerned in the measurement field.

[0003] The strapdown gravimeter is based on Newton's law of motion, and uses the data of specific force, attitude, displacement, etc. provided by the global navigation satellite system (GNSS) or other devices that can provide position information and the strapdown inertial navigation system (SINS) to extract relative gravity. The technologies related to the strapdown gravimeter, such as integration, miniaturization, temperature control, gravity extraction, etc. are relatively mature, and do not require auxiliary stabilization platform, vibration isolation system, etc. The degree of integration is high, and the shipborne gravity measurement can be carried out under any sea conditions. However, due to the inherent properties of the device itself, there is a drift accumulated over time, which requires the strapdown gravimeter to return to the pre-calibration site for zero drift error correction after a period of navigation, which also makes it difficult for the strapdown gravimeter to measure for a long time.

[0004] The atomic interference absolute gravimeter is based on optics, and uses the magneto-optical well to cool and trap atoms, so that they can freely fall. Then the atomic interference is realized by using matter waves, and the gravity information is extracted from the interference pattern. Because there is no mechanical wear in the work, the atomic interference absolute gravimeter has the advantages of no drift and strong static stability. However, the preparation of the atomic group and the vertical falling must be maintained at all times during the measurement, so the atomic gravimeter has very high requirements for the environment, and must rely on the assistance of the temperature control system, the stabilization platform, the vibration isolation system, etc. The dynamic performance is poor, and in the far sea measurement, it is difficult to resist the interference of wind and waves, and it is easy to introduce gross errors. This is the key to restricting the atomic gravimeter to measure in the far sea.

[0005] In view of the problems and defects of the above two gravity measurement methods, we propose a combined measurement method of absolute relative gravimeter to solve the above problems. SUMMARY

[0006] The application aims at solving the problems in the prior art and provides an absolute-relative gravity meter combined measurement method.

[0007] To achieve the above-mentioned purpose, the application adopts the following technical scheme:

[0008] An absolute-relative gravity meter combined measurement method comprises the following steps:

[0009] S1, taking the drift error and its change rate of the strapdown relative gravity meter, the remaining error and its change rate, and the error and its change rate of the atomic interference absolute gravity meter as the system state;

[0010] S2, taking the difference between the gravity anomalies measured by the two gravity meters as the measurement, and constructing a combined measurement model;

[0011] S3, predicting the reliable error state through robust Kalman filtering based on the maximum entropy criterion;

[0012] S4, performing error compensation on the gravity anomaly of the strapdown relative gravity meter to obtain more accurate gravity information.

[0013] Preferably, according to step S2, the combined measurement model is as follows:

[0014]

[0015] wherein the model state is a p and is the relative gravity meter drift error and its change amount, a o and is the relative gravity meter remaining error and its change amount, b e and are the absolute gravity meter error and its change amount respectively, x k-1 is the state value at the last moment, x k is the current state value

[0016] The state transition matrix is that is, from the state at k-1 moment to the state at k moment

[0017] The system noise is k-1 The corresponding system noise matrix is Q k-1 , Z k is the measurement, Z k = δg xd - δg jd , δg xd is the gravity anomaly measured by the relative gravity meter, δg jd is the gravity anomaly provided by the absolute gravity meter, and the measurement transition matrix is H k= [1 0 1 0 -10], measurement noise ξ k The corresponding system noise matrix is R k .

[0018] Preferably, according to step S3, the Kalman filtering algorithm involves the following steps:

[0019]

[0020] wherein, is the predicted value of state x at the previous time, is the one-step predicted value of x at the current time, P k-1 is the error covariance matrix of the state at the previous time, P k,k-1 is the error covariance matrix of the one-step prediction , is the covariance matrix of the measurement prediction relative to the state prediction , is the error covariance matrix of the measurement prediction, K k is the filtering gain, is the state prediction value at time k, P k is the corresponding error covariance matrix.

[0021] Preferably, according to step S4, after the error covariance matrix of the measurement prediction is calculated according to the Kalman filtering formula in step S3, the measurement information, i.e.

[0022]

[0023] Then construct a judgment test value:

[0024]

[0025] The detection criterion is as follows:

[0026]

[0027] wherein, χ 2 (α) is a pre-set fault chi-square threshold value, which can correspond to a certain false alarm rate, and the corresponding relationship between χ 2 (α) and the false alarm rate can be obtained from the chi-square distribution table. When normal, according to the formula the normal Kalman filtering algorithm is performed, i.e. the filtering gain, state prediction value and update covariance are calculated, and when abnormal, the maximum entropy Kalman filtering is performed, i.e.

[0028]

[0029] Ψ k = diag [k σ (E k,i )]

[0030]

[0031] where E k is the constructed maximum entropy factor, Ψ k is the maximum entropy weight matrix, k σ (·) is the maximum entropy kernel function, is the maximum entropy corrected measurement noise matrix, are the corrected measurement prediction error covariance matrix and filter gain, respectively.

[0032] The present application aims to provide an atomic interference absolute gravimeter-strapdown relative gravimeter combined measurement method, and aims to solve the problems of drift of the strapdown relative gravimeter and the atomic interference absolute gravimeter being easily affected by high mobility in the process of far sea measurement.

[0033] The method in the present application combines the characteristics of the atomic interference absolute gravimeter and the strapdown gravimeter, realizes taking advantages and making up for disadvantages, effectively improves the precision and reliability of far sea gravity measurement, effectively solves the problems of drift error accumulation of the strapdown relative gravimeter and poor dynamic performance of the atomic interference absolute gravimeter in the process of far sea measurement, and through combined measurement of the two, the need of the relative gravimeter for returning to the base station for drift error calibration is avoided, and the efficiency and precision of far sea measurement are improved. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a working flowchart of the strapdown relative gravimeter.

[0035] Figure 2 is a working flowchart of the atomic interference absolute gravimeter.

[0036] Figure 3 is a principle block diagram of the atomic interference absolute gravimeter-strapdown relative gravimeter combined measurement method.

[0037] Figure 4 is a reference gravity simulation diagram in the simulation experiment performed in the present application.

[0038] Figure 5 is a strapdown gravity measurement result simulation diagram in the simulation experiment performed in the present application.

[0039] Figure 6 is an atomic interference absolute gravity measurement result simulation diagram in the simulation experiment performed in the present application.

[0040] Figure 7 A comparison chart of gravity measurement results of three measurement methods performed in the present application.

[0041] Figure 8 A simulation chart of relative gravimeter drift in a semi-physical simulation experiment performed in the present application.

[0042] Figure 9 A chart of internal consistency curves of relative gravimeter measurement results in a semi-physical simulation experiment performed in the present application after introducing simulated drift.

[0043] Figure 10 A chart of internal consistency curves of atomic interference absolute gravimeter measurement results in a semi-physical simulation experiment performed in the present application after introducing high-mobility interference.

[0044] Figure 11 A chart of internal consistency curves of atomic interference absolute gravimeter-joint relative gravimeter combined measurement method in a semi-physical simulation experiment performed in the present application. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments.

[0046] REFERENCE Figures 1-11

[0047] 1 Basic principles and characteristics of joint relative gravity measurement and atomic interference absolute gravity measurement

[0048] The commonly used coordinate systems of joint relative gravimeters include navigation coordinate system (n-system), carrier coordinate system (b-system), geographic coordinate system (c-system), geocentric coordinate system (e-system) and inertial coordinate system (i-system). As people are used to using longitude, latitude and height to represent the navigation information of a carrier, the geographic coordinate system with the direction of "east-north-sky" is recorded as the navigation coordinate system. The direction of the b-system corresponding to the n-system is represented as "right-front-up". The navigation coordinate system and the carrier coordinate system with calculation errors are respectively referred to as n'-system and b'-system. Gravity measurement is performed in the n-system.

[0049] In joint gravity measurement based on SINS / GNSS, the direct difference method is most commonly used. According to Newton's law of motion and Einstein's general theory of relativity, the carrier motion acceleration in specific force information is deducted to obtain gravity information. Gravity, gravity anomaly and other information are related to position. Since the position is usually represented by longitude, latitude and height, gravity measurement is usually performed in the n-system. The specific measurement principle is as follows.

[0050]

[0051] where g n is the gravity of the earth at a certain point, also called the true gravity. and are the three-axis acceleration and velocity of the carrier, respectively. is the specific force compensated by the integrated navigation of the accelerometer. is the specific force with error, measured by the accelerometer. is the angular velocity with error, measured by the gyroscope. is the rotation matrix of n-frame relative to b-frame. is the angular velocity of the earth rotation in n-frame. is the angular velocity of the earth rotation in b-frame. The symbol

[0052] The gravity disturbance is the difference between the true gravity and the normal gravity. The disturbed gravity is:

[0053] δg n = g n - γ n (2)

[0054] where γ n is the normal gravity, which can be calculated by WGS-84, CGCS2000, etc.

[0055] The gravity usually mentioned is actually the vertical component of formula (2), that is, the gravity anomaly. The gravity anomaly is:

[0056]

[0057] where L and h are the latitude and height of the carrier, respectively. R M and R N are the meridian and prime vertical radii of the earth, respectively. ω ie is the angular velocity of the earth rotation is called the Eotovs correction. is the correction of the vertical acceleration in the sky. is the vertical component of , the specific force in n-frame,

[0058] γ = 978032.53349 (1 + 0.00530244 sin 2 L - 0.00000582 sin 2 L) - 0.3086H (4)

[0059] where H is the ellipsoidal height of the calculation point. ​

[0060] If the height at the measurement point cannot be accurately obtained, the corresponding normal gravity formula is as follows.

[0061] γ = 978032.53349 (1 + 0.00530244 sin 2 L - 0.00000582 sin 2 L) (5)

[0062] According to formulas (3)-(5), the flow of the SINS / GNSS-based gravity measurement method is shown in FIG. 2. Figure 1 In the figure, are the carrier velocities measured by the SINS and the GNSS, respectively. are the positions output by the SINS and the GNSS, respectively. It can be considered that the parameters with are calculated from the navigation information provided by the SINS and are in the n'-system. After the combined navigation is inhibited, these parameters do not have errors. It is generally considered that the velocity and position provided by the GNSS are accurate and do not have errors. represents a vertical acceleration correction term, which is obtained by once differentiating the skyward velocity of Eotovs and γ represent the Eotvos correction value and the normal gravity value calculated according to the velocity and position provided by the GNSS.

[0063] In an atomic interferometer absolute gravimeter, in order to measure the gravity acceleration, atoms are cooled and trapped by a three-dimensional magneto-optical trap in a stable environment with vacuum and constant temperature, and the atoms are further cooled to the order of micro Kelvin (μK) by polarization gradient cooling, and then the optical field is turned off to make the atoms freely fall under the action of gravity. Then, the falling cold atoms are split, reflected and combined by using π / 2-π-π / 2 three-beam laser and the counter-propagating laser pulses formed by the mirrors, so as to realize M-Z type interference and transfer the acceleration information of the atoms relative to the mirror into the interference fringes. The phase difference of the atomic interference of the two paths is Δφ = k eff ·gT 2 , wherein, is the effective wave vector of the Raman light, k1 and k2 are the wave vectors of the two Raman lights, λ is the wavelength of the laser, and T is the time interval between the two Raman lights.

[0064] In order to compensate for the Doppler frequency shift in the free-fall time of the atoms, the frequency difference between the two Raman lights needs to be linearly chirped, at this time, the interference phase difference can be expressed as Δφ = (k eff ·g-2πα)T 2

[0065] When the chirp rate α = k eff ·g / 2π, the Raman light phase shift is completely consistent with the gravity phase shift.

[0066] In the process of Raman light, the internal state of the atom changes. At the initial moment, the atom is in the state of |g>, after the first beam of π / 2 Raman light, part of the atoms jump to the state of |e>, after the second beam of π Raman light, the atoms in the state of |g> jump to the state of |e>, and the atoms in the state of |e> jump to the state of |g>, after the third beam of Raman light, the atomic wave packet superposition interference occurs. The population P of the atom in the state of |e> after interference is measured by a photodetector, and the relationship between P and the gravity phase shift Δφ is

[0067] Where P0 is the fringe shift, and C is the contrast. Under static conditions, since the measured gravity value is constant or changes very little, by changing the chirp rate α and the Raman light interval time T, a plurality of interference fringes can be fitted, and the gravity value corresponding to the center α0 of the fringe is the local gravity value:

[0068] It can be seen that the strapdown gravity measurement is based on SINS, so it has high dynamic measurement performance, but due to the existence of mechanical devices, there is inevitably a zero drift problem. These problems may lead to a decrease in the accuracy of the measurement results, so these instruments need to be returned to the standard reference point for error calibration and calibration on a regular basis. For long-period ocean measurement tasks, it is often more difficult and time-consuming to perform accurate calibration in a marine environment, so this requirement for regular calibration has become a limiting factor. Benefiting from its non-mechanical working mode, the atomic gravimeter uses optical equipment based on atomic interference, which does not require mechanical centering and is basically fully automatic, thereby avoiding the problem of mechanical wear and tear. In addition, its internal vacuum system is sealed and does not require external force to evacuate, and there are no movable parts inside, which further enhances its stability and avoids zero drift. However, due to the high environmental requirements of optical devices, the atomic gravimeter will face more complex temperature, vibration, and electromagnetic environmental factors when performing shipborne measurement compared to the laboratory environment, which will introduce more measurement noise and errors. However, it can be seen that the characteristics of the two gravimeters are actually complementary.

[0069] 2 Establishment of combined measurement model

[0070] In order to overcome the shortcomings of two gravity meters, and realize the acquisition of accurate gravity information in the open sea measurement, based on the characteristics of the strapdown relative gravity meter and the atomic interference absolute gravity meter, the application designs an atomic interference absolute gravity meter-strapdown relative gravity meter combined measurement method for the open sea. The method takes the drift error and its rate of change of the strapdown relative gravity meter, the remaining error and its rate of change, and the error and its rate of change of the atomic interference absolute gravity meter as the system state, takes the difference between the gravity anomalies measured by the two gravity meters as the measurement, and constructs a combined measurement model. Then, the reliable error state is estimated through the robust Kalman filtering based on the maximum entropy criterion, the error of the gravity anomaly of the strapdown relative gravity meter is compensated, and more accurate gravity information is obtained.

[0071] The working block diagram of the combined measurement is shown in the accompanying Figure 3 , and the combined measurement model is as follows.

[0072]

[0073] Among them, the model state is a p and are the drift error and its change of the relative gravity meter, a o and are the remaining error and its change of the relative gravity meter, b e and are the error and its change of the absolute gravity meter respectively. x k-1 is the state value at the last moment, x k is the current state value.

[0074] The state transition matrix is that is, from the state at k-1 moment to the state at k moment.

[0075] The system noise η k-1 corresponding to the system noise matrix is Q k-1 .

[0076] Z k is the measurement, Z k = δg xd - δg jd , δg xd is the gravity anomaly measured by the relative gravity meter, and δg jd is the gravity anomaly provided by the absolute gravity meter.

[0077] The measurement transition matrix H k = [1 0 1 0 -1 0].

[0078] The measurement noise ξ k corresponding to the system noise matrix is R k .

[0079] This measurement model enables combined measurements using a strapdown relative gravimeter and an atomic interferometer absolute gravimeter. Leveraging the drift-free characteristic of the absolute gravimeter, the drift error of the relative gravimeter can be estimated in real time, thereby reducing its impact and achieving drift-free measurements at sea during long voyages. This eliminates the need for the relative gravimeter to return for calibration, improving the efficiency and accuracy of gravity measurements at sea.

[0080] 3 Robust Kalman Filter

[0081] However, in offshore measurements, factors such as waves and wind often affect the absolute gravimeter, causing vibrations and other disturbances that lead to anomalies in the combined measurement information and thus affect the accuracy of the combined measurement. To avoid this situation, a robust strategy based on the maximum entropy criterion is introduced into the Kalman filter to obtain a robust Kalman filter, which improves the robustness of data fusion.

[0082] 3.1 Kalman Filtering

[0083] The Kalman filter algorithm involves the following steps

[0084]

[0085] in, Let x be the predicted value of state x at the previous time step. Let P be the predicted value of x at the current time step. k-1 State of the previous moment The error covariance matrix, P k,k-1 For one-step prediction The error covariance matrix. For measurement and prediction Relative to state prediction The covariance matrix. K represents the error covariance matrix of the measurement prediction. k This represents the filter gain. P represents the predicted state value at time k. k for The corresponding error covariance matrix.

[0086] Kalman filtering excels in dealing with Gaussian noise by utilizing only the second-order moment information of the error. However, its performance degrades in non-Gaussian noise environments (such as vibration). To overcome this limitation, this invention introduces the maximum correlation entropy criterion to integrate the second-order and higher-order moment information of the error, enabling Kalman filtering to provide more accurate state estimation performance under non-Gaussian noise conditions.

[0087] 3.2 Robust Kalman Filtering Based on Maximum Entropy and Chi-square Test

[0088] Since the probability of anomalies occurring during navigation is not very high, a chi-square test is first introduced to construct a decision criterion in order to fully utilize the robust strategy. When an anomaly is detected, maximum entropy robustness is applied; otherwise, normal Kalman filtering is performed.

[0089] After calculating the error covariance of the measurement prediction according to the Kalman filter formula Then, the measurement information is calculated, that is...

[0090]

[0091] Then construct the judgment test value:

[0092]

[0093] The detection criteria are as follows:

[0094]

[0095] Where, χ 2 (α) is a pre-set fault chi-square threshold value, which can correspond to a certain false alarm rate. χ can be obtained from the chi-square distribution table. 2 The relationship between (α) and false alarm rate.

[0096] When the condition is normal, the normal Kalman filtering algorithm is performed according to formulas (14)-(16), i.e., the filter gain, state prediction value, and updated covariance are calculated. When the condition is abnormal, maximum entropy Kalman filtering is performed, i.e.

[0097]

[0098]

[0099] In the formula, E k ψ is the maximum entropy factor constructed. k Let k be the maximum entropy weight matrix. σ (·) is the maximum entropy kernel function. This is the measurement noise matrix after maximum entropy correction. These are the corrected measurement prediction error covariance matrix and the filter gain, respectively.

[0100] By using robust Kalman spectroscopy, the impact of absolute gravimeter measurement anomalies caused by disturbances such as high maneuverability and vibration during offshore measurements can be avoided, thus improving the robustness of the combined measurements.

[0101] 4. Implementation steps of the present invention in shipboard experiments

[0102] The experimental verification method of the combined measurement method of atomic interferometer absolute gravimeter and strapdown relative gravimeter for offshore applications of this invention includes the following steps.

[0103] 1) Obtain the measurement data of the strapdown relative gravity meter and the atomic interference absolute gravity meter. Specifically, according to the strapdown relative gravity meter and the atomic interference absolute gravity meter, the measurement data of the two gravity meters is obtained according to the measurement steps in the accompanying drawings Figure 1 and accompanying drawings Figure 2 .

[0104] 2) Establish a combined measurement model. According to the combined measurement steps in the accompanying drawings Figure 3 , and the model and robust Kalman filtering of 2, 3, perform combined measurement.

[0105] 3) Simulation experiment. Perform a shipborne combined navigation experiment to compare the performance of the method proposed in the application and the strapdown relative gravity meter and the atomic interference absolute gravity meter. In the simulation experiment, first, simulate a gravity reference, which is used as a standard to compare and evaluate the relative gravity anomaly and the absolute gravity anomaly accuracy, that is, it acts as prior data in the external compliance accuracy evaluation, as shown in the accompanying drawings Figure 4 . Then, simulate the measurement results of the relative gravity meter and the absolute gravity meter. The relative gravity meter has high dynamic performance, but inevitably has zero drift due to device reasons. According to its performance characteristics, the corresponding simulation measurement results are shown in the accompanying drawings Figure 5 . The absolute gravity meter is easily affected by vibration, high mobility and other environments to produce sudden changes. According to its characteristics, the simulation measurement results are shown in the accompanying drawings Figure 6 . Then, perform combined measurement according to the results of the two gravity meters.

[0106] In this experiment, the simulation parameters of the combined measurement are set as follows.

[0107] System noise matrix:

[0108] Q k-1 = diag([0.1 mGal 0.004 mGal 0.2 mGal 0.005 mGal 0.1 mGal 0.002 mGal]) 2

[0109] Measurement noise matrix: R k = diag([1 mGal]) 2

[0110] The initial error covariance matrix is as follows:

[0111] P0= diag([0.1 mGal 0.003 mGal 0.04 mGal 0.001 mGal 0.1 mGal 0.002 mGal]) 2

[0112] Initial state: x0= [000000] T .

[0113] The combined measurement results are shown in the attached figure. Figure 7 As shown in Table 1, the external compliance accuracy is as follows. The results indicate that the combined measurement results after robust filtering have the highest accuracy, meaning the deviation between the combined gravity and the reference gravity is minimal, resulting in the highest accuracy.

[0114] Table 1 External compliance accuracy of each method in the simulation experiment.

[0115]

[0116] 4) Semi-physical simulation. Sea trials were conducted using an experimental strapdown gravimeter and an atomic oceanographic gravimeter, covering a total of four measurement lines. Accuracy was used as the evaluation standard. However, the sea trials were relatively short, the strapdown gravimeter showed no drift, and the atomic gravimeter's measurement environment was stable. Therefore, using this data, a simulation of approximately one month of measurements was performed. First, the above data was expanded to approximately one month's worth of data. Then, relative gravimeter drift was simulated and incorporated into the expanded relative gravimeter data. The relative gravimeter drift is shown in the attached figure. Figure 8 As shown.

[0117] The parameter settings for the combined measurement model in this experiment are as follows:

[0118] System noise matrix:

[0119] Q k-1 =diag([0.1mGal 0.003mGal 0.04mGal 0.001mGal 0.1mGal 0.002mGal]) 2

[0120] Measurement noise matrix: R k =diag([0.5mGal]) 2

[0121] The initial error covariance matrix is ​​as follows:

[0122] P0=diag([0.1mGal 0.003mGal 0.04mGal 0.001mGal 0.1mGal 0.002mGal]) 2

[0123] Initial state: x0 = [0 0 0 0 0 0] T .

[0124] Finally, two data abrupt changes caused by vibration and high-speed maneuvering were introduced into the expanded absolute gravimeter measurement data. The resulting internal coincidence curves for the relative gravimeter with introduced drift and the absolute gravimeter with introduced abrupt changes are shown in the attached figures. Figures 9-11The inner coincidence accuracy is shown in Table 2. The experimental results show that the discrete degree of the combined gravity measurement data is the smallest, the consistency and repeatability are the best, and the reliability is the highest.

[0125] Table 2 Inner coincidence accuracy of each method in semi-physical simulation experiment

[0126]

[0127] 5) Overall analysis, the application effectively solves the problems of drift error accumulation of the strapdown relative gravity meter and poor dynamic performance of the atomic interference absolute gravity meter during the far sea measurement, and the combination measurement of the two also avoids the relative gravity meter needing to return for drift error calibration at regular time intervals, and improves the efficiency and precision of the far sea measurement.

[0128] The above describes only the preferred specific implementation of the application, but the protection scope of the application is not limited to this, any skilled person in the art, according to the technical solution and the inventive concept of the application, makes equivalent replacement or change within the technical range disclosed by the application, should be covered within the protection scope of the application.

Claims

1. An absolute relative gravimeter combination measurement method, characterized by, The method comprises the following steps: S1, taking the drift error and its rate of change of the strapdown relative gravimeter, the remaining error and its rate of change, and the error and its rate of change of the atomic interference absolute gravimeter as the system state; S2, taking the difference between the gravity anomalies measured by the two gravimeters as the measurement, and constructing a combined measurement model; S3, predicting reliable error states by robust Kalman filtering based on the maximum entropy criterion; S4, compensating the error of the gravity anomaly of the strapdown relative gravimeter to obtain more accurate gravity information.

2. The method of claim 1, wherein the absolute relative gravimeter is configured to measure the absolute relative gravity in a range of 0.1 mgal to 1000 mgal. According to step S2, the combined measurement model is as follows: Wherein, the model state is a p And is the relative gravimeter drift error and its change amount, a o And is the relative gravimeter remaining error and its change amount, b e And is the absolute gravimeter error and its change amount, x k-1 is the state value of the last time, x k is the current state value State transition matrix i.e. from state at time k-1 to state at time k System noise η k-1 The corresponding system noise matrix is Q k-1 Z k is the measurement, Z k = δg xd - δg jd , δg xd is the gravity anomaly measured by the relative gravimeter, δg jd is the gravity anomaly provided by the absolute gravimeter, the measurement transfer matrix H k = [1 0 1 0 -1 0], the measurement noise ξ k The corresponding system noise matrix is R k .

3. The method of claim 2, wherein the absolute relative gravimeter combination measurement method is characterized by, According to step S3, the Kalman filtering algorithm involves the following steps: in, Let x be the predicted value of state x at the previous time step. Let P be the predicted value of x at the current time step. k-1 State of the previous moment The error covariance matrix, P k,k-1 For one-step prediction The error covariance matrix, For measurement and prediction Relative to state prediction The covariance matrix, Let K be the error covariance matrix of the measurement prediction. k For filter gain, Let P be the predicted state value at time k. k for The corresponding error covariance matrix.

4. The method of claim 3, wherein the absolute relative gravimeter combination measurement method is characterized by, According to step S4, after the measurement prediction error covariance is calculated in step S3 by the Kalman filter formula Then, the measurement information is calculated, i.e. Then a judgment test value is constructed: The detection criterion is as follows: wherein χ 2 (α) is a pre-set fault chi-square threshold value, which can correspond to a certain false alarm rate, and χ 2 (α) and the corresponding relationship of false alarm rate, when normal, according to the formula Normal Kalman filtering algorithm is carried out, that is, the filtering gain, state prediction value and update covariance are calculated, and when abnormal, the maximum entropy Kalman filtering is carried out, that is Ψ k = diag [k σ (E k,i )] where E k is the maximum entropy factor of the structure, ψ k is the maximum entropy weight matrix, k σ (·) is the maximum entropy kernel function, is the maximum entropy modified measurement noise matrix, are the modified measurement prediction error covariance matrix and the filter gain, respectively.

Images