SINS-DVL integrated navigation method based on cumulative DVL speed measurement model

Abstract

The invention discloses an SINS-DVL integrated navigation method based on a cumulative DVL speed measurement model, and the method comprises the following steps: S1, introducing an SINS resolving principle and an SINS-DVL integrated navigation principle, and laying a theoretical foundation for the proposal of the SINS-DVL integrated navigation method based on the cumulative DVL speed measurement model; the method aims at mining DVL speed information so as to further improve the SINS-DVL integrated navigation performance; in a classical filtering method, the DVL speed at the current moment can only be used once, that is, filtering recursion at the current moment is realized, according to an attitude matrix recursion rule, a measurement model is reconstructed, the DVL speed within a fixed time interval is used for the current filtering recursion, an SINS-DVL integrated navigation model adopting an accumulated speed is provided, and finally, the filtering recursion at the current moment is realized. The effectiveness of the method is verified through marine shipborne experiments.

Description

Technical Field

[0001] The invention relates to the technical field of inertial base integrated navigation, and in particular to a SINS-DVL integrated navigation method based on a cumulative DVL velocity measurement model. Background Art

[0002] The strapdown inertial navigation system (SINS) is a key device in the field of underwater navigation. It can provide continuous navigation information, namely attitude, velocity, position, angular rate and acceleration information. However, due to its dead reckoning characteristics, the corresponding navigation error will accumulate over time, which greatly limits the long-term application of SINS. Combined navigation can utilize the complementary characteristics of multiple sensors to improve or eliminate the divergence of SINS navigation errors, and has become a common method in practical applications. For example, integration with the global navigation satellite system (GNSS) - SINS-GNSS combined navigation. However, the global navigation satellite system cannot be used in the underwater field. In the case of GNSS denial, the Doppler odometer (DVL) becomes an auxiliary device to replace GNSS and is currently widely used to enhance the long-duration navigation performance of SINS underwater.

[0003] SINS-DVL integrated navigation is usually carried out in an indirect combination manner, that is, the error of navigation parameters is estimated through a linear state space model and some filtering algorithms. Then the divergence of SINS is corrected by the navigation error to obtain accurate navigation information. Under this indirect integration architecture, the corresponding research of SINS-DVL mainly focuses on two aspects: constructing a mathematical model including a state model and a measurement model, and developing a filtering algorithm with better performance. Among them, the state model of SINS-DVL is the error state model of SINS. Thanks to the linearization characteristics of the indirect model, the commonly used error state model is a linear model with a small initial error assumption. Due to different coordinate systems, different SINS mechanical arrangements are generated, and then different state error models appear. In recent years, researchers have realized that the use of Lie group states can greatly improve the performance of the state model. Therefore, this is mainly because the matrix Lie group can be used to represent the global state. The advantage of this global state representation is that it is due to the logarithmic linearization of the generated error state model, which is of great significance for research in the field of initial alignment even in the case of arbitrarily large initial misalignment.

[0004] In terms of measurement models, they can be roughly divided into two categories, loose coupling and tight coupling, based on the different uses of DVL output. For loose coupling, the DVL output used is the three-dimensional velocity of the carrier system. In contrast, for tight coupling, the beam output of the DVL is directly used as the measurement without converting it into three-dimensional velocity. Compared with loose coupling, tight coupling is more robust and can work even in limited beam measurement conditions. As for filtering algorithms, the research hotspot is Kalman filtering and its improved filtering with robustness or adaptability. It can be seen that there are many improvements with superior performance for state models and filtering algorithms. Summary of the invention

[0005] The purpose of the present invention is to solve the above technical problems and to propose a SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A SINS-DVL integrated navigation method based on a cumulative DVL velocity measurement model comprises the following steps:

[0008] S1, introduces the SINS solution principle and SINS-DVL integrated navigation principle, and lays the theoretical foundation for the SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model;

[0009] S2, according to the chain rule of the posture matrix, derive τ∈[t m ,t] time DVL measurement calculation formula;

[0010] S3, according to the cumulative DVL speed formula obtained in step S2, derive the SINS-DVL cumulative speed measurement equation, and then propose a SINS-DVL integrated navigation model based on the cumulative DVL speed measurement model.

[0011] Preferably, the SINS-DVL integrated navigation principle: using the SINS mechanical arrangement of the Earth-centered Earth-fixed coordinate system (ECEF), the corresponding SINS update equation is given by the following formula:

[0012]

[0013] In the formula, is the attitude rotation matrix from the carrier system to the ECEF coordinate system; is the ground speed in ECEF, is the position vector, and are the carrier angular velocity and specific force measured by the gyroscope and accelerometer, respectively, is the Earth's rotation angular velocity vector, is the gravity vector represented by ECEF, and (·×) represents the antisymmetric matrix of ·.

[0014] Preferably, the SINS-DVL integrated navigation process is as follows:

[0015] A1, using the measured and Execute the following formula to solve SINS;

[0016]

[0017] A2, based on the linear error state space model, uses auxiliary information, i.e., DVL velocity, to estimate the error of navigation parameters;

[0018] A3, compensating the navigation parameters through error state estimation.

[0019] Preferably, the cumulative DVL speed formula is derived as follows: The cumulative DVL speed formula is based on τ∈[t m ,t] time is obtained by the recursive formula of the attitude rotation matrix.

[0020] Preferably, the navigation results of the proposed method and the traditional method are compared through a shipboard measurement experiment to prove the effectiveness and applicability of the proposed method, including the following steps:

[0021] B1, obtain the original data of SINS and DVL;

[0022] B2, perform SINS solution, according to the formula Get the cumulative speed calculated by SINS;

[0023] B3, according to the formula Get the cumulative speed of DVL;

[0024] B4, according to the formula and Construct a SINS-DVL integrated navigation model based on cumulative speed;

[0025] B5, conduct SINS-DVL integrated navigation;

[0026] B6. Experiments,The proposed method is evaluated using shipboard test data.

[0027] Preferably, the onboard test data is collected on board a ship.

[0028] Preferably, the platform used in the experiment in step B6 is equipped with high-precision SINS, GPS and DVL, and the output of SINS-GPS integrated navigation is used as reference data for comparison.

[0029] Preferably, the chain rule posture matrix of the posture matrix can be decomposed as follows:

[0030]

[0031] In the formula, Represents the posture matrix at time t, t∈[0,t]. represents the ECEF frame attitude from time 0 to t, represents the posture at time 0, Represents the posture of the carrier system from time 0 to t.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention focuses on the use process of DVL measurement. Different from the method of replacing loose coupling with tight coupling, the present invention is committed to mining DVL speed information to further improve the SINS-DVL integrated navigation performance. In the classical filtering method, the DVL speed at the current moment can only be used once, that is, to realize the filtering recursion at the current moment. The present invention reconstructs the measurement model according to the attitude matrix recursion rule, uses the DVL speed within a fixed time interval for the current filtering recursion, and proposes a SINS-DVL integrated navigation model using cumulative speed. Finally, the effectiveness of the present invention is verified through marine shipborne experiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The shipboard trajectory of the first sea navigation experiment was mapped.

[0035] Figure 2 The shipboard trajectory of the second sea trial was plotted.

[0036] Figure 3 The latitude errors of the various methods for the first sea trial are plotted.

[0037] Figure 4 The longitude error of each method for the first sea navigation experiment is plotted.

[0038] Figure 5 The latitude errors of the various methods for the second sea trial are plotted.

[0039] Figure 6 The longitude error diagrams of various methods for the second sea navigation experiment are plotted.

[0040] Figure 7 The position error of each method for the first sea trial is plotted.

[0041] Figure 8 The position error graphs of each method for the second sea trial are plotted. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0043] Reference Figure 1-Figure 8 , a SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model, comprising the following steps:

[0044] S1, introduces the SINS solution principle and SINS-DVL integrated navigation principle, and lays the theoretical foundation for the SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model;

[0045] S2, according to the chain rule of the posture matrix, derive τ∈[t m ,t] time DVL measurement calculation formula;

[0046] S3, according to the cumulative DVL speed formula obtained in step S2, derive the SINS-DVL cumulative speed measurement equation, and then propose a SINS-DVL integrated navigation model based on the cumulative DVL speed measurement model.

[0047] In the present invention, the SINS-DVL integrated navigation principle is: using the SINS mechanical arrangement of the Earth-centered Earth-fixed coordinate system (ECEF), the corresponding SINS update equation is given by the following formula:

[0048]

[0049] In the formula, is the attitude rotation matrix from the carrier system to the ECEF coordinate system; is the ground speed in ECEF, is the position vector, and are the carrier angular velocity and specific force measured by the gyroscope and accelerometer, respectively, is the Earth's rotation angular velocity vector, is the gravity vector represented by ECEF, and (·×) represents the antisymmetric matrix of ·.

[0050] In the present invention, the SINS-DVL combined navigation process is as follows:

[0051] A1, using the measured and Execute the following formula to solve SINS;

[0052]

[0053] A2, based on the linear error state space model, uses auxiliary information, i.e., DVL velocity, to estimate the error of navigation parameters;

[0054] A3, compensating the navigation parameters through error state estimation.

[0055] In the present invention, the cumulative DVL speed formula is derived as follows: The cumulative DVL speed formula is based on τ∈[t m ,t] time is obtained by the recursive formula of the attitude rotation matrix.

[0056] In the present invention, the navigation results of the proposed method and the traditional method are compared through shipboard measurement experiments to prove the effectiveness and applicability of the proposed method, which includes the following steps:

[0057] B1, obtain the original data of SINS and DVL;

[0058] B2, perform SINS solution, according to the formula Get the cumulative speed calculated by SINS;

[0059] B3, according to the formula Get the cumulative speed of DVL;

[0060] B4, according to the formula and Construct a SINS-DVL integrated navigation model based on cumulative speed;

[0061] B5, conduct SINS-DVL integrated navigation;

[0062] B6. Experiments,The proposed method is evaluated using shipboard test data.

[0063] In the present invention, the shipboard test data is collected on board the ship.

[0064] In the present invention, the platform used in the experiment in step B6 is equipped with high-precision SINS, GPS and DVL, and the output of the SINS-GPS integrated navigation is used as reference data for comparison.

[0065] In the present invention, the chain rule posture matrix of the posture matrix can be decomposed as follows:

[0066]

[0067] In the formula, Represents the posture matrix at time t, t∈[0,t]. represents the ECEF frame attitude from time 0 to t, represents the posture at time 0, Represents the posture of the carrier system from time 0 to t.

[0068] Further detailed explanation:

[0069] To achieve the above purpose, the present invention first introduces the update equation of SINS, and then gives the SINS-DVL integrated navigation principle. Then the cumulative velocity measurement equation is derived. And a SINS-DVL integrated navigation model based on the cumulative velocity is given. Finally, through the marine SINS-DVL experiment, the effectiveness of the method proposed by the present invention is evaluated. The specific steps are as follows:

[0070] 1) The SINS solution principle and the SINS-DVL integrated navigation principle are introduced to lay a theoretical foundation for the SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model of the present invention.

[0071] 2) According to the chain rule of the posture matrix, derive τ∈[t m ,t] time.

[0072] 3) According to the cumulative DVL speed formula obtained in 2), the SINS-DVL cumulative speed measurement equation is derived, and then the SINS-DVL integrated navigation model based on the cumulative DVL speed measurement model of the present invention is proposed.

[0073] 4) Through shipborne field experiments, the navigation results of the proposed method are compared with those of the traditional method to prove the effectiveness and applicability of the proposed method.

[0074] The present invention focuses on the use of DVL measurement. In the classical filtering method, the DVL velocity at the current moment can only be used once, that is, to implement the filtering recursion at the current moment. The present invention derives τ∈[t m ,t], the DVL velocity formula within the fixed time interval is used for the current filtering recursion, and the SINS-DVL integrated navigation model using the cumulative velocity is proposed. Finally, the effectiveness of the present invention is verified through marine shipboard experiments.

[0075] The present invention is oriented to reality and improves the performance of SINS-DVL combined navigation on the basis of real-time performance.

[0076] SINS-DVL integrated navigation principle

[0077] The present invention adopts the SINS mechanical arrangement of the Earth-centered Earth-fixed coordinate system (ECEF). The corresponding SINS update equation is given by the following formula

[0078]

[0079] In the formula, is the attitude rotation matrix from the carrier system to the ECEF coordinate system. is the ground speed in ECEF system. is the position vector. and are the carrier angular velocity and specific force measured by the gyroscope and accelerometer, respectively. is the Earth's rotation angular rate vector. is the gravity vector expressed as ECEF. (·×) represents the antisymmetric matrix of ·.

[0080] In the SINS-DVL integrated navigation of the present invention, the DVL output is the vehicle's driving speed, that is, It is related to ground speed The relationship is given by the following formula.

[0081]

[0082] Therefore, the SINS-DVL integrated navigation process of the present invention is as follows. First, using the measured and The SINS solution of formulas (1) to (3) is performed. Then, based on the linear error state space model, the auxiliary information, namely the DVL velocity, is used to estimate the error of the navigation parameters. Finally, the navigation parameters are compensated by the error state estimation. It can be seen that in the above process, the DVL velocity is only used once in a single filtering recursion. Although some smoothing methods can be used to make better use of the measurement results, such as reducing the DVL velocity noise, these methods are not real-time.

[0083] Derivation of the cumulative DVL velocity formula

[0084] The cumulative DVL velocity formula is based on τ∈[t m ,t] time. First, the recursive formula of the attitude matrix is ​​explained.

[0085] According to the chain rule of the attitude matrix, the attitude matrix can be decomposed as follows

[0086]

[0087] In the formula, Represents the posture matrix at time t, t∈[0,t]. Represents the ECEF system attitude from time 0 to time t. Indicates the posture at time 0. Represents the posture of the carrier system from time 0 to t.

[0088] This pose decomposition process has been widely used in coarse alignment based on SINS optimization. Similarly, τ∈[t m,t] can be decomposed into the posture matrix

[0089]

[0090] In the formula, Denote τ∈[t m ,t] time period, Represents the ECEF frame attitude from time τ to time t. represents the posture at time t. represents the posture of the carrier system from time τ to time t. Represents the posture matrix at time t.

[0091] Assume that the duration of DVL measurement is τ, that is

[0092]

[0093] In the formula, is the ground speed at time τ, represents the DVL measurement speed at time τ.

[0094] Substituting (6) into (7) yields

[0095]

[0096] Multiply both sides of (8) by Available

[0097]

[0098] In the time interval [t m ,t], integrating both sides of (9), we can get

[0099]

[0100] Pose Matrix can be placed outside the integral because it can be considered as a time interval [t m ,t]. Now, a new The measurement model is shown in (10).

[0101] Next, the new velocity vector is given by:

[0102]

[0103] In the formula, is the differential of the ground velocity at time t after a measurement period of τ.

[0104] Correspondingly, the new DVL measurement vector is as follows.

[0105]

[0106] In the formula, is the DVL measurement velocity differential at time t after a measurement time of τ.

[0107] SINS-DVL integrated navigation model based on cumulative DVL velocity measurement model

[0108] Next, the velocity vectors in (11) and (12) are calculated.

[0109] Assume that the current discrete time is t M =M·Δt, where Δt is the update period interval [t k-1 ,t k ] duration. The accumulated velocity vector is as follows.

[0110]

[0111] In the formula, t M The differential of the ground speed over the period. t k Ground speed at .

[0112] The differential equation for is given by the following.

[0113]

[0114] The recursive operation of the above formula is

[0115]

[0116] In the formula, Represents time 0 to t k ECEF system posture. Represents time 0 to t k-1 ECEF system posture. Indicates t k-1 Time to t k ECEF system posture.

[0117] as follows.

[0118]

[0119] The accumulated DVL velocity vector can be calculated as

[0120]

[0121] In the formula, t MDVL measures the velocity differential over a period of time. t k DVL measurement speed at .

[0122] The differential equation for is given by the following.

[0123]

[0124] Its recursive calculation formula is:

[0125]

[0126] In the formula, Represents time 0 to t k The loading system posture. Represents time 0 to t k-1 The loading system posture. Indicates t k-1 Time to t k The loading system posture.

[0127]

[0128] After obtaining the new velocity update formula (17), the error model of SINS-DVL can be constructed. The error state of SINS-DVL is as follows.

[0129]

[0130] Where δx is the state vector. Represents the attitude angle error vector. represents the velocity error vector. represents the position error vector. b g and b a represent the gyroscope drift and accelerometer bias vector respectively.

[0131]

[0132] in, represents the parameter with error, which is the value given by SINS calculation. (·×) represents the antisymmetric matrix of ·.

[0133]

[0134] The linear state model is given by the following equation.

[0135]

[0136] Wherein, the state transfer matrix F and the noise control matrix G are as follows.

[0137]

[0138] Noise Matrix and denote the corresponding noise of the gyroscope and accelerometer, respectively.

[0139] The measurement model is given by the following formula.

[0140]

[0141] Where y represents the measurement vector.

[0142] in is the accumulated carrier velocity vector calculated based on the DVL output. is the accumulated ground velocity vector calculated based on SINS.

[0143]

[0144] According to (22) and (29), (28) can be expanded to

[0145]

[0146] for Because of its input and the initial value I 3×3 are all known error-free values, so they are also considered to be error-free values ​​(ignoring calculation errors). Only by Considering this and according to (13), we can get

[0147]

[0148] Among them, the measurement transfer matrix H M as follows.

[0149]

[0150] According to (30) and (32), the measurement model can be obtained as follows:

[0151]

[0152] Where η is the measurement noise.

[0153] Implementation steps of the present invention in shipboard experiments

[0154] The experimental verification method of the SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model of the present invention specifically comprises the following steps:

[0155] 1) Obtain the original data of SINS and DVL.

[0156] 2) Perform SINS solution and then obtain the cumulative speed calculated by SINS according to formula (29).

[0157] 3) According to formula (17), the cumulative speed of DVL is obtained.

[0158] 4) According to formulas (25) and (33), a SINS-DVL integrated navigation model based on cumulative speed is constructed.

[0159] 5) Carry out SINS-DVL integrated navigation.

[0160] 6) Experiments. The proposed method is evaluated using shipboard test data. The test data is collected on board a ship. The experimental platform is equipped with high-precision SINS, GPS, and DVL. The output of the SINS-GPS integrated navigation is used as reference data for comparison. The raw outputs of the gyroscope and accelerometer as well as the DVL velocity are processed using different integration algorithms. Table 1 summarizes the technical specifications of the equipped system.

[0161] Table 1 Parameters of main sensors

[0162]

[0163] In the experiment, the following five different SINS-DVL integrated navigation algorithms are evaluated. The conventional algorithm adopts the measurement model (4), denoted as "SINS-DVL". The SINS-DVL with the constructed cumulative velocity measurement model is denoted as "SINS / DVL-intev=time length". Four different time lengths are selected, namely 05s, 1s, 10s and 20s. The cumulative velocities in (11) and (12) are calculated according to the SINS data update rate. In the experiment, the SINS data update rate is 200Hz, while the DVL update rate is 1Hz. When calculating the cumulative velocity (12), it is assumed that the DVL velocity within 1 second is the value obtained at the last moment.

[0164] Two 5-hour test data segments were selected to evaluate the algorithms. For the first segment, the survey vessel started from a static state, while for the second segment, the survey vessel started from a dynamic state. The reference trajectory and the trajectories of the five algorithms are shown in Figure 2. Figure 1 and 2 The latitude estimation errors of the five algorithms are shown in Figure 3 and 4 The longitude estimation errors of the five integration algorithms are shown in Figure 5 and 6 The position estimation errors of the five integration algorithms are shown in Figure 7 and 8 shown.

[0165] The results show that the proposed cumulative velocity construction method can improve the performance of integrated navigation. Among the five algorithms, the integration performance gets better and better as the time interval increases. However, when the time interval increases to a certain extent, the performance improvement is no longer obvious. For the first test segment, one second is enough to greatly improve the navigation performance. In contrast, for the second segment of data, the performance improvement of one second accumulation is small.

[0166] It should be noted that when the time interval increases, the computational burden increases simultaneously. In addition, in these algorithms, the measurement noise is still assumed to be Gaussian and is independent of the process noise. However, when the time interval increases, the measurement noise will no longer be Gaussian. Considering the performance improvement, computational burden, and filtering stability, a time interval of one second to twenty seconds is recommended.

[0167] Summary. This paper focuses on the use mechanism of DVL velocity in a loosely coupled framework based on Kalman filtering. Specifically, based on the principle of attitude matrix decomposition, the DVL velocity measurement model can be reorganized, and the DVL velocity within a certain time interval can be incorporated into the current filtered observation, so that the DVL velocity can be fully utilized. Two 5-hour data segments were tested to evaluate the performance of the proposed method. The results show that in the proposed method, using more DVL accumulated velocities (longer integration time) can indeed improve the integration performance. However, when the DVL velocity integration time interval increases, the computational burden also increases. At the same time, when more DVL velocities are incorporated into the current filtered observation, the measurement noise may not be Gaussian, nor will it be independent of the process noise. In practical applications, it is recommended that the DVL velocity integration time interval be one to twenty seconds

[0168] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model, characterized in that: The following steps are involved: S1, introduces the SINS solution principle and SINS-DVL integrated navigation principle, and lays the theoretical foundation for the SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model; S2, according to the chain rule of the posture matrix, derive τ∈[t m ,t] time DVL measurement calculation formula; S3, according to the cumulative DVL speed formula obtained in step S2, derive the SINS-DVL cumulative speed measurement equation, and then propose a SINS-DVL integrated navigation model based on the cumulative DVL speed measurement model.

2. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 1 is characterized in that: The SINS-DVL integrated navigation principle: using the SINS mechanical arrangement of the Earth-centered Earth-fixed coordinate system (ECEF), the corresponding SINS update equation is given by the following formula: In the formula, is the attitude rotation matrix from the carrier system to the ECEF coordinate system; is the ground speed in ECEF, is the position vector, and are the carrier angular velocity and specific force measured by the gyroscope and accelerometer, respectively, is the Earth's rotation angular velocity vector, is the gravity vector represented by ECEF, and (·×) represents the antisymmetric matrix of ·.

3. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 2 is characterized in that: The SINS-DVL integrated navigation process is as follows: A1, using the measured and Execute the following formula to solve SINS; A2, based on the linear error state space model, uses auxiliary information, i.e., DVL velocity, to estimate the error of navigation parameters; A3, compensating the navigation parameters through error state estimation.

4. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 2 is characterized in that: Derivation of the cumulative DVL speed formula: The cumulative DVL speed formula is based on τ∈[t m, It is obtained by the recursive formula of the attitude rotation matrix at the moment t].

5. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 1, characterized in that: The navigation results of the proposed method are compared with those of the traditional method through shipboard field experiments, which proves the effectiveness and applicability of the proposed method, including the following steps: B1, obtain the original data of SINS and DVL; B2, perform SINS solution, according to the formula Get the cumulative speed calculated by SINS; B3, according to the formula Get the cumulative speed of DVL; B4, according to the formula and Construct a SINS-DVL integrated navigation model based on cumulative speed; B5, conduct SINS-DVL integrated navigation; B6. Experiments,The proposed method is evaluated using shipboard test data.

6. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 5, characterized in that: The onboard test data is collected on board the ship.

7. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 5, characterized in that: The platform used in the experiment in step B6 is equipped with high-precision SINS, GPS and DVL, and the output of the SINS-GPS integrated navigation is used as reference data for comparison.

8. The SINS-DVL integrated navigation method based on the cumulative DVL velocity measurement model according to claim 1, characterized in that: The chain rule posture matrix of the posture matrix can be decomposed as follows: In the formula, Represents the posture matrix at time t, t∈[0,t]. represents the ECEF frame attitude from time 0 to t, represents the posture at time 0, Represents the posture of the carrier system from time 0 to t.

Images