Rocket seaborne alignment method and system based on federated filtering

Abstract

The invention discloses a rocket seaborne alignment method based on federated filtering and a system thereof. The rocket seaborne alignment method based on federated filtering comprises the following steps of performing equipment power-up and filter filtering parameter initialization; after equipment power-up and initialization of filtering parameters of the filter are completed, whether coarse alignment calculation can be carried out or not is judged; if the coarse alignment calculation can be carried out, carrying out coarse alignment calculation to obtain an attitude angle; and after coarse alignment calculation is completed and an attitude angle is obtained, fine alignment calculation is performed. According to the method, data of the multiple sensors are fused through the federated filtering algorithm, and the alignment precision of the rocket during offshore launching is remarkably improved. Meanwhile, the influence of the complex marine environment on rocket alignment can be effectively dealt with, and the stability of the launching process is improved. Moreover, by optimizing the structure of the federated filter, the fault tolerance performance of the system is improved, and even if a certain sensor breaks down, accurate alignment of the rocket can be ensured.

Description

Technical Field

[0001] The present application relates to the field of rockets, and in particular, to a method and system for rocket alignment at sea based on federal filtering. Background Art

[0002] Traditional rocket alignment schemes at sea mainly rely on the combination of the global satellite navigation system (GNSS) and the inertial navigation system (INS). When the hull is stationary, coarse alignment is achieved through compass effect analysis to obtain the initial attitude. Afterwards, the difference between the position deviation and velocity deviation of GNSS and INS is used as a band estimate to perform Kalman filtering, and the initialization value is corrected with the filtering result to determine the initial attitude of the rocket. Other commonly used alignment schemes at sea include transfer alignment and mooring combination alignment. Transfer alignment uses the information of the aligned main inertial navigation system to perform initial alignment on the sub-inertial navigation system that needs to be aligned, and uses the navigation parameters (such as speed, position) or output values ​​(such as acceleration, angular velocity, attitude angle) calculated by the main inertial navigation system and the sub-inertial navigation system to match. In the moored state, the mooring combination alignment uses the known dock position to assist the CNS (starlight navigation system) to forward information, obtain high-precision carrier attitude, and then directly assign coarse attitude to the inertial navigation system, and then further perform fine alignment to measure drift and compensate for component errors. Considering the climate environment and dynamic use conditions in the atmosphere on the sea surface, the weighted recursive least squares method can be used to perform precise alignment of the starlight / inertial combination. Through the weighted processing link, the weight analysis of the star sensor observation quantity is realized, and the alignment time is shortened under the condition of ensuring the alignment accuracy. However, the traditional combined navigation marine alignment scheme has the following disadvantages: 1. Highly dependent on satellite signals. At sea, especially near the coast, islands or encountering bad weather, satellite signals will be blocked or interfered, resulting in reduced alignment accuracy or even inability to complete alignment. 2. Error accumulation. Although the inertial navigation system can provide navigation information autonomously, its errors will accumulate over time. After long-term use, the alignment information needs to be corrected, otherwise it will affect the navigation accuracy. 3. Long alignment time. Traditional alignment methods may take a long time to collect and process data to ensure the accuracy of alignment. This may become a problem in the case of emergency launch or rapid deployment. 4. Affected by the environment. The marine environment is complex and changeable, and environmental factors such as waves and wind speed may affect the accuracy and stability of alignment. 5. High equipment cost. In order to achieve high-precision alignment, high-precision and high-stability navigation equipment and sensors may be required, which increases the cost of the equipment and the complexity of maintenance. Transfer alignment requires a higher-precision inertial navigation system for auxiliary alignment, which is more expensive and takes longer to match parameters, so the alignment time is longer.

[0003] Therefore, how to provide a method for achieving precise alignment of a rocket when launching at sea while meeting the alignment requirements and avoiding the impact of environmental influences has become an urgent problem to be solved in this field. Summary of the invention

[0004] The present application proposes a method for rocket alignment at sea based on federal filtering, comprising the following steps: powering on the device and initializing the filter parameters; after completing the powering on of the device and initializing the filter parameters, determining whether a coarse alignment calculation can be performed; if a coarse alignment calculation can be performed, performing a coarse alignment calculation to obtain an attitude angle; after completing the coarse alignment calculation to obtain the attitude angle, performing a fine alignment calculation.

[0005] In the above-mentioned method for rocket alignment at sea based on federal filtering, equipment power-on and parameter initialization include performing parameter initialization after the rocket is powered on on the launch platform.

[0006] The method for rocket alignment at sea based on federal filtering as described above, wherein determining whether a coarse alignment calculation can be performed includes the following sub-steps: calculating the sway period, speed and misalignment angle according to the speed and attitude increment measured by the inertial navigation system; obtaining the lever arm mean value according to the calculation results of the sway period, speed and misalignment angle; determining whether a coarse alignment calculation can be performed according to the lever arm mean value.

[0007] The above-mentioned method for rocket sea alignment based on federal filtering, in which a rough alignment calculation is performed to obtain the attitude angle includes the following sub-steps: constructing the projection of the gravity vector and the earth's rotation vector in the rocket body coordinate system and the navigation coordinate system based on the data collected by the inertial navigation system; calculating the direction cosine matrix from the carrier coordinate system to the navigation coordinate system based on the geometric relationship between the gravity vector and the earth's rotation vector; and solving the attitude angle based on the direction cosine matrix.

[0008] The method for rocket alignment at sea based on federal filtering as described above, wherein the precise alignment calculation includes the following sub-steps: preprocessing the data acquired by the starlight navigation system, the satellite navigation system and the inertial navigation system; fusing the preprocessed data and outputting the optimal state estimation.

[0009] A rocket sea alignment system based on federated filtering specifically comprises: an initialization unit, a judgment unit, a coarse alignment unit and a fine alignment unit; the initialization unit is used for powering on the device and initializing parameters; the judgment unit is used for judging whether coarse alignment calculation can be performed; if coarse alignment calculation can be performed, the coarse alignment unit performs coarse alignment calculation to obtain attitude angle; the fine alignment unit is used for performing fine alignment calculation.

[0010] In the above-mentioned rocket sea alignment system based on federal filtering, the initialization unit performs equipment power-on and parameter initialization, including parameter initialization after the rocket is powered on on the launch platform.

[0011] The rocket sea alignment system based on federal filtering as described above, wherein the judgment unit judges whether a coarse alignment calculation can be performed, including the following sub-steps: calculating the sway period, speed and misalignment angle according to the speed and attitude increment measured by the inertial navigation system; obtaining the lever arm mean value according to the calculation results of the sway period, speed and misalignment angle; judging whether a coarse alignment calculation can be performed according to the lever arm mean value.

[0012] As described above, in the rocket sea alignment system based on federal filtering, the coarse alignment unit performs coarse alignment calculations to obtain the attitude angle, which includes the following sub-steps: constructing the projections of the gravity vector and the earth's rotation vector in the rocket body coordinate system and the navigation coordinate system based on the data collected by the inertial navigation system; calculating the direction cosine matrix from the carrier coordinate system to the navigation coordinate system based on the geometric relationship between the gravity vector and the earth's rotation vector; and solving the attitude angle based on the direction cosine matrix.

[0013] In the rocket sea alignment system based on federal filtering as described above, the precise alignment calculation performed by the precise alignment unit includes the following sub-steps: preprocessing the data acquired by the starlight navigation system, the satellite navigation system and the inertial navigation system; fusing the preprocessed data and outputting the optimal state estimate.

[0014] This application has the following beneficial effects:

[0015] This application uses a federated filtering algorithm to fuse multiple sensor data, significantly improving the alignment accuracy of rockets during sea launches. At the same time, it can effectively deal with the impact of complex marine environments on rocket alignment and improve the stability of the launch process. In addition, this application improves the fault tolerance of the system by optimizing the structure of the federated filter, and can ensure the precise alignment of the rocket even if a sensor fails. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is a flow chart of a rocket sea alignment method based on federal filtering provided according to an embodiment of the present application;

[0018] Figure 2is a schematic diagram of the distribution of the federated filters provided according to an embodiment of the present application;

[0019] Figure 3 It is a schematic diagram of the internal structure of a rocket sea alignment system based on federal filtering provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0021] In order to solve the problems existing in traditional combined navigation without introducing higher costs, the present invention proposes a method and system for rocket sea alignment based on federal filtering. The method utilizes a federal filtering algorithm to fuse multiple sensor data, while meeting the alignment requirements and avoiding the influence of environmental influences, thereby achieving precise alignment of the rocket when it is launched at sea.

[0022] Embodiment 1

[0023] like Figure 1 As shown, the rocket sea alignment method based on federal filtering provided in the embodiment of the present application specifically includes the following steps:

[0024] Step S110: Power on the device and initialize the filter parameters.

[0025] After the rocket is powered on on the launch platform, its parameters are initialized and the angular velocity output in real time by the inertial navigation system is used to calculate the rocket body's shaking period.

[0026] The solution process includes: using a sliding window with a length of 30s, calculating the frequency domain data of the angular velocity in the window through Fourier transform, taking the frequencies corresponding to the two peaks with the largest power within a frequency of 1Hz as the shaking frequency, and the shaking period is the corresponding inverse.

[0027] Step S120: After completing the device power-on and the initialization of the filter parameters, determine whether the coarse alignment calculation can be performed.

[0028] In the rough alignment calculation, it is necessary to obtain the speed and attitude increment measured by the inertial navigation system (INS), and calculate the shaking period, speed, and misalignment angle.

[0029] The calculation process of the shaking period refers to step S110, which specifically uses a sliding window of 30 seconds in length, calculates the frequency domain data of the angular velocity measured by the inertial navigation system within the window through Fourier transform, and takes the frequencies corresponding to the two peaks with the largest power within a frequency of 1 Hz as the shaking frequency. The shaking period is the corresponding inverse.

[0030] Misalignment Angle Specifically expressed as:

[0031]

[0032] in, is the coordinate transformation matrix corresponding to the misalignment angle, + and - represent the values ​​at the current moment and the previous moment respectively. is the angular velocity output by the inertial navigation system, ω ie is the Earth's rotation angular velocity, τ i The time for calculation, It is the coordinate transformation matrix from geographic system to arrow system.

[0033] speed Specifically expressed as:

[0034]

[0035] in, Output acceleration for the inertial navigation system, is the local gravitational acceleration vector.

[0036] The displacement of the inertial navigation system can be calculated according to the calculated speed and attitude, and the distance between the inertial navigation system and the satellite receiver can be calculated according to the current positioning of the GNSS as the lever arm, and the mean value of the lever arm is calculated according to a certain time window, and compared with the installation position between the devices. If the difference between the calculated mean value of the lever arm and the installation position between the devices is less than the specified threshold, the calculated value is considered valid, the parameters are valid, and step S130 is executed. Otherwise, the calculated value is considered invalid, and the operation of step S120 is repeated, that is, the shaking period, speed, and misalignment angle are repeatedly calculated, and the mean value of the lever arm is calculated according to the calculated speed and attitude, and whether the rough alignment calculation can be performed is determined according to the mean value of the lever arm.

[0037] Step S130: perform coarse alignment calculation to obtain the attitude angle.

[0038] The rocket body acceleration and angular velocity, or velocity and attitude increment, determined by the inertial navigation system and measured by the gyroscope and accelerometer are obtained. According to the rocket body acceleration and angular velocity, or velocity and attitude increment, the projection of the gravity acceleration vector and the earth rotation angular velocity vector in the rocket body coordinate system (b system) and the navigation coordinate system (n system) are constructed.

[0039] The projection of the b system is the result of the inertial navigation system output, and the projection of the n system is related to the coordinate system and longitude and latitude. Taking the North Celestial East system as an example, the gravity acceleration vector is g n = [0; g; 0], the angular velocity of the earth's rotation is w n =[w ie *cosL;w ie *sinL;0], L is the latitude.

[0040] Using the geometric relationship between vectors, the direction cosine matrix from the rocket body coordinate system (b system) to the navigation coordinate system (n system) is calculated. Specifically expressed as:

[0041]

[0042] in represents the angular velocity of the Earth's rotation in the n system, is the angular velocity output of the inertial navigation system, Represents the output acceleration of the inertial navigation system.

[0043] According to the direction cosine matrix, the attitude angles such as pitch angle, roll angle and azimuth angle are calculated, which are the results of rough alignment.

[0044] Still taking the northeast sky as an example, the pitch angle θ, roll angle γ, and heading angle ψ are specifically expressed as:

[0045]

[0046] Among them C 32 Represents the 3rd row and 2nd column element in the direction cosine matrix, C 31 Represents the element in row 3 and column 1 of the direction cosine matrix, C 33 Represents the 3rd row and 3rd column element in the direction cosine matrix, C 12 Represents the first row and second column element in the direction cosine matrix, C 22 Represents the element in row 2 and column 2 of the direction cosine matrix.

[0047] Step S140: After completing the rough alignment calculation to obtain the attitude angle, perform the fine alignment calculation.

[0048] The precise alignment calculation specifically includes the following sub-steps:

[0049] Step S1401: pre-processing the data acquired by the constellation navigation system (CNS), the satellite navigation system (GNSS) and the inertial navigation system (INS).

[0050] The data from the constellation navigation system (CNS), satellite navigation system (GNSS) and inertial navigation system (INS) are preprocessed to remove noise and outliers, and the speed information input by the satellite navigation system (GNSS) is low-pass filtered to reduce the error caused by sea sloshing.

[0051] Step S1402: Use a federated filter to fuse the pre-processed data obtained by the star navigation system (CNS), the satellite navigation system (GNSS) and the inertial navigation system (INS), and output the optimal state estimation.

[0052] like Figure 2 As shown in FIG. 1 , a specific structural diagram of a federated filter is shown, wherein the federated filter includes multiple sub-filters (sub-filter 1 and sub-filter 2) and a main filter. The federated filter adopts a decentralized filtering structure to optimally fuse the state estimation errors of a starlight navigation system (CNS), a satellite navigation system (GNSS), and an inertial navigation system (INS) to obtain a globally optimal state estimation.

[0053] Among them, the satellite navigation system, inertial navigation system and starlight navigation system first pass through different sub-filters to obtain a one-step filtering result. Taking the combination of the three as an example, the satellite navigation system and inertial navigation system enter sub-filter 1, the starlight navigation system and inertial navigation system enter sub-filter 2, and the results of sub-filters 1 and 2 enter the main filter.

[0054] Specifically, the satellite navigation system will determine the precise position, velocity and time information of the rocket body and send it to the main filter. The inertial navigation system will send the acceleration and angular velocity of the rocket body, or the velocity and attitude increments measured by the gyroscope and accelerometer to sub-filter 1. The starlight navigation system uses the position of the stars to determine the attitude and position of the rocket body and send it to sub-filter 2. Both sub-filters will output error estimates of position, velocity and attitude. The main filter performs secondary filtering on the results of the two sub-filters, and the output result is the global optimal state estimate.

[0055] In the filtering process, appropriate weights can be allocated according to the characteristics and accuracy of each system (that is, different weights are assigned in the filtering process for different short-term and long-term stabilities of different sensors) to improve the alignment accuracy.

[0056] Step S1403: Calculate the precise posture and position of the carrier based on the output optimal state estimation.

[0057] The precise posture and position of the carrier are calculated based on the output of the federated filter.

[0058] The output of the main filter is the error of attitude and position. The initial values ​​of attitude and position are corrected according to the error. Specifically, the position information is subtracted and the attitude is multiplied by quaternion.

[0059] Correct the errors of the inertial navigation system to improve its alignment accuracy. Compare the fine alignment results with the known position or reference position to verify the alignment accuracy.

[0060] If the alignment accuracy does not meet the requirements, the parameters of the federated filter can be adjusted or data acquisition and filtering can be performed again.

[0061] Embodiment 2

[0062] like Figure 3 As shown, this embodiment provides a rocket sea alignment system based on federal filtering, which specifically includes: an initialization unit 310, a judgment unit 320, a coarse alignment unit 330 and a fine alignment unit 340.

[0063] The initialization unit 310 is used to power on the device and initialize the filter parameters.

[0064] After the rocket is powered on on the launch platform, its parameters are initialized and the angular velocity output in real time by the inertial navigation system is used to calculate the rocket body's shaking period.

[0065] The solution process includes: using a sliding window with a length of 30s, calculating the frequency domain data of the angular velocity in the window through Fourier transform, taking the frequencies corresponding to the two peaks with the largest power within a frequency of 1Hz as the shaking frequency, and the shaking period is the corresponding inverse.

[0066] The judgment unit 320 is used to judge whether the rough alignment calculation can be performed after the device is powered on and the filter parameters are initialized.

[0067] In the rough alignment calculation, it is necessary to obtain the speed and attitude increment measured by the inertial navigation system (INS), and calculate the shaking period, speed, and misalignment angle.

[0068] The calculation process of the shaking period refers to step S110, which specifically uses a sliding window of 30 seconds in length, calculates the frequency domain data of the angular velocity measured by the inertial navigation system within the window through Fourier transform, and takes the frequencies corresponding to the two peaks with the largest power within a frequency of 1 Hz as the shaking frequency. The shaking period is the corresponding inverse.

[0069] Misalignment Angle Specifically expressed as:

[0070]

[0071] in, is the coordinate transformation matrix corresponding to the misalignment angle, + and - represent the values ​​at the current moment and the previous moment respectively. is the angular velocity output by the inertial navigation system, ω ie is the Earth's rotation angular velocity, τ i The time for calculation, It is the coordinate transformation matrix from geographic system to arrow system.

[0072] speed Specifically expressed as:

[0073]

[0074] in, Output acceleration for the inertial navigation system, is the local gravitational acceleration vector.

[0075] The displacement of the inertial navigation system can be calculated based on the calculated speed and attitude, and the distance between the inertial navigation system and the satellite receiver can be calculated based on the current positioning of the GNSS as the lever arm. Based on a certain time window, the average value of the lever arm is calculated and compared with the pre-measured values ​​between the devices to determine the validity of the calculated value.

[0076] If the calculated arm mean is within the specified threshold of the pre-measured value, the calculated value is considered valid, the parameters are valid, and the rough alignment unit 330 is executed. Otherwise, the calculated value is considered invalid, and the operation in the step judgment unit 320 is repeated.

[0077] The coarse alignment unit 330 is used to perform coarse alignment calculations to obtain attitude angles.

[0078] The rocket body acceleration and angular velocity, or velocity and attitude increment, determined by the inertial navigation system and measured by the gyroscope and accelerometer are obtained. According to the rocket body acceleration and angular velocity, or velocity and attitude increment, the projection of the gravity acceleration vector and the earth rotation angular velocity vector in the rocket body coordinate system (b system) and the navigation coordinate system (n system) are constructed.

[0079] The projection of the b system is the result of the inertial navigation system output, and the projection of the n system is related to the coordinate system and longitude and latitude. Taking the North Celestial East system as an example, the gravity acceleration vector is g n = [0; g; 0], the angular velocity of the earth's rotation is w n =[w ie *cosL;w ie *sinL;0], L is the latitude.

[0080] Using the geometric relationship between vectors, the direction cosine matrix from the rocket body coordinate system (b system) to the navigation coordinate system (n system) is calculated. Specifically expressed as:

[0081]

[0082] in represents the angular velocity of the Earth's rotation in the n system, is the angular velocity output of the inertial navigation system, Represents the output acceleration of the inertial navigation system.

[0083] According to the direction cosine matrix, the attitude angles such as pitch angle, roll angle and azimuth angle are calculated, which are the results of rough alignment.

[0084] Still taking the northeast sky as an example, the pitch angle θ, roll angle γ, and heading angle ψ are specifically expressed as:

[0085]

[0086] Among them C 32 Represents the 3rd row and 2nd column element in the direction cosine matrix, C 31 Represents the element in row 3 and column 1 of the direction cosine matrix, C 33 Represents the 3rd row and 3rd column element in the direction cosine matrix, C 12 Represents the first row and second column element in the direction cosine matrix, C 22 Represents the element in row 2 and column 2 of the direction cosine matrix.

[0087] The fine alignment unit 340 is used to perform fine alignment calculation after completing the rough alignment calculation to obtain the attitude angle.

[0088] The precise alignment unit 340 includes the following submodules: a preprocessing module, a fusion module, and a precise attitude and position calculation module.

[0089] The preprocessing module is used to preprocess the data obtained by the star navigation system (CNS), satellite navigation system (GNSS) and inertial navigation system (INS).

[0090] The data from the constellation navigation system (CNS), satellite navigation system (GNSS) and inertial navigation system (INS) are preprocessed to remove noise and outliers, and the speed information input by the satellite navigation system (GNSS) is low-pass filtered to reduce the error caused by sea sloshing.

[0091] The fusion module is used to fuse the pre-processed data obtained by the star navigation system (CNS), the satellite navigation system (GNSS) and the inertial navigation system (INS) using a federated filter to output the optimal state estimation.

[0092] like Figure 2As shown in FIG. 1 , a specific structural diagram of a federated filter is shown, wherein the federated filter includes multiple sub-filters (sub-filter 1 and sub-filter 2) and a main filter. The federated filter adopts a decentralized filtering structure to optimally fuse the state estimation errors of a starlight navigation system (CNS), a satellite navigation system (GNSS), and an inertial navigation system (INS) to obtain a globally optimal state estimation.

[0093] Among them, the satellite navigation system, inertial navigation system and starlight navigation system first pass through different sub-filters to obtain a one-step filtering result. Taking the combination of the three as an example, the satellite navigation system and inertial navigation system enter sub-filter 1, the starlight navigation system and inertial navigation system enter sub-filter 2, and the results of sub-filters 1 and 2 enter the main filter.

[0094] Specifically, the satellite navigation system will determine the precise position, velocity and time information of the rocket body and send it to the main filter. The inertial navigation system will send the acceleration and angular velocity of the rocket body, or the velocity and attitude increments measured by the gyroscope and accelerometer to sub-filter 1. The starlight navigation system uses the position of the stars to determine the attitude and position of the rocket body and send it to sub-filter 2. Both sub-filters will output error estimates of position, velocity and attitude. The main filter performs secondary filtering on the results of the two sub-filters, and the output result is the global optimal state estimate.

[0095] In the filtering process, appropriate weights can be allocated according to the characteristics and accuracy of each system (that is, different weights are assigned in the filtering process for different short-term and long-term stabilities of different sensors) to improve the alignment accuracy.

[0096] The precise attitude and position calculation module is used to calculate the precise attitude and position of the carrier according to the output optimal state estimation.

[0097] The precise attitude and position of the carrier are calculated based on the output of the federated filter. The output of the main filter is the error of attitude and position. The initial values ​​of attitude and position are corrected based on the error, specifically by subtracting the position information and performing quaternion multiplication on the attitude.

[0098] Correct the errors of the inertial navigation system to improve its alignment accuracy. Compare the fine alignment results with the known position or reference position to verify the alignment accuracy.

[0099] If the alignment accuracy does not meet the requirements, the parameters of the federated filter can be adjusted or data acquisition and filtering can be performed again.

[0100] This application has the following beneficial effects:

[0101] This application uses a federated filtering algorithm to fuse multiple sensor data, significantly improving the alignment accuracy of rockets during sea launches. At the same time, it can effectively deal with the impact of complex marine environments on rocket alignment and improve the stability of the launch process. In addition, this application improves the fault tolerance of the system by optimizing the structure of the federated filter, and can ensure the precise alignment of the rocket even if a sensor fails.

[0102] Although the present application is described with reference to examples, this is for illustrative purposes only and is not intended to limit the present application, and changes, additions and / or deletions to the embodiments may be made without departing from the scope of the present application.

[0103] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for rocket alignment at sea based on federated filtering, characterized in that: The following steps are involved: Power on the device and initialize the filter parameters; After completing the equipment power-on and the initialization of the filter parameters, determine whether the rough alignment calculation can be performed; If a rough alignment calculation can be performed, a rough alignment calculation is performed to obtain the attitude angle; After completing the rough alignment calculation to obtain the attitude angle, perform the fine alignment calculation.

2. The method for rocket alignment at sea based on federated filtering as claimed in claim 1, characterized in that: Equipment power-on and parameter initialization include parameter initialization after the rocket is powered on on the launch platform.

3. The method for rocket alignment at sea based on federated filtering as claimed in claim 1, characterized in that: Determining whether a rough alignment calculation can be performed includes the following sub-steps: Calculate the sway period, speed and misalignment angle based on the speed and attitude increment measured by the inertial navigation system; According to the calculation results of the shaking period, speed and misalignment angle, the average value of the lever arm is obtained; Whether the rough alignment calculation can be performed is determined based on the mean value of the lever arm.

4. The method for rocket alignment at sea based on federated filtering as claimed in claim 1, characterized in that: Performing rough alignment calculations to obtain attitude angles includes the following sub-steps: Based on the data collected by the inertial navigation system, the projections of the gravity vector and the earth's rotation vector in the rocket body coordinate system and the navigation coordinate system are constructed; According to the geometric relationship between the gravity vector and the earth's rotation vector, the direction cosine matrix from the carrier coordinate system to the navigation coordinate system is calculated; According to the direction cosine matrix, the attitude angle is calculated.

5. The method for rocket alignment at sea based on federated filtering as claimed in claim 4, characterized in that: Performing precise alignment calculations includes the following sub-steps: Preprocess the data obtained by the star navigation system, satellite navigation system and inertial navigation system; The preprocessed data is fused and the optimal state estimation is output.

6. A rocket sea alignment system based on federated filtering, characterized in that: Specifically include: Initialization unit, judgment unit, coarse alignment unit, fine alignment unit; Initialization unit, used for powering on the device and initializing parameters; A judging unit, used to judge whether a rough alignment calculation can be performed; If the rough alignment calculation can be performed, the rough alignment unit performs the rough alignment calculation to obtain the attitude angle; The fine alignment unit is used to perform fine alignment calculations.

7. The rocket sea alignment system based on federated filtering as claimed in claim 6, characterized in that: The initialization unit performs equipment power-on and parameter initialization, including parameter initialization after the rocket is powered on on the launch platform.

8. The rocket sea alignment system based on federated filtering as claimed in claim 6, characterized in that: The judging unit judges whether the rough alignment calculation can be performed, including the following sub-steps: Calculate the sway period, speed and misalignment angle based on the speed and attitude increment measured by the inertial navigation system; According to the calculation results of the shaking period, speed and misalignment angle, the average value of the lever arm is obtained; Whether the rough alignment calculation can be performed is determined based on the mean value of the lever arm.

9. The rocket sea alignment system based on federated filtering as claimed in claim 6, characterized in that: The coarse alignment unit performs coarse alignment calculations to obtain the attitude angle, including the following sub-steps: Based on the data collected by the inertial navigation system, the projections of the gravity vector and the earth's rotation vector in the rocket body coordinate system and the navigation coordinate system are constructed; According to the geometric relationship between the gravity vector and the earth's rotation vector, the direction cosine matrix from the carrier coordinate system to the navigation coordinate system is calculated; According to the direction cosine matrix, the attitude angle is calculated.

10. The rocket sea alignment system based on federated filtering as claimed in claim 6, characterized in that: The fine alignment calculation by the fine alignment unit includes the following sub-steps: Preprocess the data obtained by the star navigation system, satellite navigation system and inertial navigation system; The preprocessed data is fused and the optimal state estimation is output.

Images