CN-122015864-A - Initial alignment method for dynamic base of submarine craft based on multi-navigation Kalman filtering

CN122015864ACN 122015864 ACN122015864 ACN 122015864ACN-122015864-A

Abstract

The invention discloses a submarine navigation dynamic base initial alignment method based on multi-navigation system Kalman filtering, which comprises the steps of constructing N navigation systems which are uniformly distributed in azimuth, sequentially increasing included angles between x axes of the N navigation systems and x axes of the N systems, splitting three-dimensional attitude misalignment angles of STEKF algorithm models on each navigation system, independently running STEKF algorithm on each navigation system to obtain initial course angle estimation of a carrier in each navigation system, and obtaining an optimal navigation system and a final alignment result corresponding to the optimal navigation system by utilizing a preset screening rule according to the initial course angle estimation in each navigation system. According to the invention, the problem of large azimuth misalignment angle is converted into the problem of small misalignment angle estimation which can be accurately processed by the linear model by constructing a plurality of navigation systems covering all azimuth angles, and the problem of estimation accuracy reduction of state transformation Kalman filtering and nonlinear filtering under the condition of large misalignment angle can be overcome based on three-dimensional attitude misalignment angle splitting to select an optimal result from parallel filters.

Inventors

SHEN QIANG
YANG WENGUO
LIU JIEYU
LI CAN
Li Xinsan
ZHOU XIAOGANG
QIN WEIWEI
ZHAO JING

Assignees

中国人民解放军火箭军工程大学

Dates

Publication Date: 20260512
Application Date: 20260228

Claims (10)

1. The initial alignment method of the submarine motion base based on the multi-navigation Kalman filtering is characterized by comprising the following steps of: S1, constructing N navigation systems which are uniformly distributed in azimuth, sequentially increasing the included angles between the x-axis of the N navigation systems and the x-axis of the N system, and splitting the three-dimensional attitude misalignment angle of a STEKF algorithm model on each navigation system, wherein the N system is a North east coordinate system; S2, independently running STEKF algorithm on each navigation system to obtain initial course angle estimation value of the current time carrier in each navigation system; And S3, obtaining an optimal navigation system and obtaining a final alignment result corresponding to the optimal navigation system by utilizing a preset screening rule according to the initial course angle estimation value in each navigation system.
2. The method for initial alignment of a submersible motion base based on multi-navigation kalman filtering according to claim 1, wherein S1 comprises: the included angle between the initial moment direction and the north direction of the submarine is set as Constructing N navigation systems uniformly distributed in azimuth, namely 1 st navigation system And (3) with Overlapping; Obtaining Posture transformation relationship between the series and n series: , Wherein, the Which represents the ith navigation system and, Representation of A coordinate transformation matrix tied to n-system, Representing the included angle between the x axes of two adjacent navigation systems; Splitting the three-dimensional attitude misalignment angle of each navigation system STEKF algorithm model: , Wherein, the Is shown in The three-dimensional attitude misalignment angle at the current moment in the system, Is shown in The three-dimensional attitude misalignment angle at the initial time in the system, Is shown in And the change amount of the three-dimensional attitude misalignment angle from the initial moment to the current moment is obtained.
3. The method for initial alignment of a submersible motion base based on multi-navigation kalman filtering according to claim 2, wherein S2 comprises: s2.1, obtaining a system equation of a filter in each navigation system according to the three-dimensional attitude misalignment angle of each navigation system; S2.2, uniformly integrating inertial sensor data output by the strapdown inertial navigation system in adjacent filtering periods, and updating navigation parameters of all filters in parallel by utilizing an integration result after reaching a measurement updating moment; S2.3, taking the velocity measurement value of the Doppler velometer as the observed quantity, filtering to obtain the optimal estimation of the error state, correcting the carrier posture at the initial moment of each navigation system, and converting the corrected value to obtain the initial orientation and the initial orientation of the underwater vehicle in each navigation system Is the included angle of the x-axis.
4. The method for initial alignment of a submersible motor according to claim 3, wherein the method comprises the steps of, The system equation for the in-line filter is: , Wherein, the Representation of Is used for the differentiation of the (c) and (d), Representation of The state of the STEKF algorithm in the system, In the form of a state transition matrix, For the noise to drive the matrix, Is process noise.
5. The method for initial alignment of a submersible vehicle motion base based on multi-navigation Kalman filtering of claim 4, wherein, State of STEKF algorithm in the system The expression of (2) is: , Wherein, the Representing rotation of the calculated value of the carrier speed in the navigation computer to true The error involved in the tying-up is, Representing the error involved in rotating the calculated value of the carrier position in the navigation computer to the true n i series, Indicating the zero offset of the gyroscope, Indicating that the accelerometer is zero-offset, Representing the transpose of the matrix.
6. The method for initial alignment of a submersible vehicle motion base based on multi-navigation kalman filtering of claim 4, wherein the state transition matrix And the noise driving matrix The expressions of (2) are respectively: , , Wherein, the Representation of Is used for the matrix of all zeros, Representation of Is used for the matrix of all zeros, Representing the coordinate transformation matrix from the b-system to the n i -system, Representing the calculated value of the carrier speed in the navigation computer under the n i series, Representing the calculated value of the carrier position in the navigation computer under the n i series; 、、、、、、、 And All are intermediate variables, and b represents the carrier coordinate system.
7. The method for initial alignment of a submersible vehicle motion base based on multi-navigation kalman filtering of claim 6, wherein the intermediate variables 、、、、、、、 And The expressions of (2) are respectively: , , , , , , , , , Wherein, the Indicating that the rotation angular velocity of n series relative to i series is The projection of the system is performed, Representation of Is used for the matrix of the anti-symmetry of (a), Represents n is tied to A coordinate transformation matrix of the system is provided, Representation of A coordinate transformation matrix tied to n-system, Representation of Is used for the matrix of the anti-symmetry of (a), Representing the calculated value of the position of the carrier in the navigation computer, Representation of Is used for the matrix of the anti-symmetry of (a), Representing gravitational acceleration Is used for the matrix of the anti-symmetry of (a), Indicating that the rotation angular velocity of the earth is at Projection of the system Is used for the matrix of the anti-symmetry of (a), Indicating that the rotation angular velocity of the n-series relative to the e-series is Projection of the system The i system is the geocentric inertial coordinate system; 、、 And All are intermediate variables, and the expressions are respectively: , , , , Wherein, the Representing the latitude of the carrier, h representing the height of the carrier, And Respectively represent a radius of curvature of a meridian and a radius of curvature of a unitary mortise, The rotational angular velocity of the earth is represented, 、 The north and east speeds of the carrier, respectively.
8. The method of initial alignment of a submersible motion base based on multi-navigation kalman filtering of claim 7, wherein S2.2 comprises: Let the measurement update period of Kalman filtering be T, in To the carrier system in the filtering period of (a) Coordinate transformation matrix of system Decomposing to obtain And The relation of (2) is: , Wherein, the Representing within a filtering period From time b to A coordinate transformation matrix of the system is provided, Representation of From time b to A coordinate transformation matrix of the system is provided, Representing slave Time of day Fastened inertial chain to Time of day A coordinate transformation matrix of the system is provided, Representing slave Inertial system to moment of time coincident with b system Time of day A coordinate transformation matrix of a tied inertial system, Representing slave Time b is tied to Coordinate transformation matrix of inertial system overlapping with b system at moment; is represented in the filtering period Rotation change of the system; Setting time , Representing slave Time b is tied to A coordinate transformation matrix of an inertial system which coincides with the b system at the moment, The differential equation of (2) is: , Wherein, the Representation of Is used for the differentiation of the (c) and (d), The projection of the rotation angular velocity of the b-system relative to the i-system at the i-system is shown, Representation of Is an antisymmetric matrix of (a); the initial value of (1) is ; Based on Is shown in the differential equation of (2) Intra-internally-performed integral computation to obtain a slave Time b is tied to Coordinate transformation matrix of inertial system with moment overlapped with b system Is a value of (2); the decoupling process of velocity integral estimation is converted into the following form: Wherein, the Representation of The speed of the carrier relative to the e-line at the moment The projection of the system is performed, Representation of The speed of the carrier relative to e at time The projection of the system is performed, The specific force of the carrier is indicated, Representation of The rotation angular velocity of the earth at moment The projection of the system is performed, Representation of The rotation angular velocity of the time n system relative to the e system is Projection of the system, e representing the geocentric earth coordinate system; complete in the filtering period And Is obtained from the integral calculation of each navigation system Updating navigation parameters at moment to obtain Speed of time; The decoupling process of the position integral estimation is converted into the following form: , Wherein, the Representation of The position of the moment, i.e. the projection of the position variation of the carrier with respect to the starting point in the n i series, Representation of The position of the moment in time, Representing slave Time b is tied to Coordinate transformation matrix of inertial system with moment overlapped with b system and moment ; In each navigation system, get And updating navigation parameters of all filters in parallel by using the integration results of the position, the speed and the gesture of the time carrier.
9. The method of initial alignment of a submersible motion base based on multi-navigation kalman filtering of claim 6, wherein S2.3 comprises: Taking the velocity measurement value of the Doppler velocimeter as the observed quantity, estimating the error state of the navigation system through a filtering algorithm, and obtaining the optimal estimation of the error state And Performing correction to obtain correction values And The correction process comprises the following steps: , , Wherein, the Representing from b series to The calculated values of the coordinate transformation matrix of the system in the navigation computer, Representation of A value at an initial time; in the filter of each navigation system, the pair Converting to obtain initial course angle estimation value of carrier in each navigation system I.e. initial orientation of the submarine Is the included angle of the x-axis.
10. The method of initial alignment of a submersible motion base based on multi-navigation system kalman filtering according to any one of claims 1 to 9, wherein S3 comprises: for the current navigation system , The two adjacent navigation systems are respectively And The absolute values of the initial course angle estimation values of the underwater vehicle calculated in the three navigation systems are respectively 、 And If the following conditions are satisfied: , Then determine the current navigation system Is the optimal navigation system; Current navigation system The navigation parameters calculated in the step (a) are converted into n series and are output as a final alignment result.

Description

Initial alignment method for dynamic base of submarine craft based on multi-navigation Kalman filtering Technical Field The invention belongs to the technical field of inertial navigation, and particularly relates to a submarine motion base initial alignment method based on multi-navigation Kalman filtering. Background The underwater vehicle is an important carrier for exploring underwater mystery and developing ocean resources, and the accuracy and reliability of a navigation system of the underwater vehicle are directly related to the efficiency and success and failure of task execution. Because satellite navigation signals are difficult to spread in long distance under water, the underwater vehicle generally adopts an autonomous navigation technology with inertial navigation as a basis and various acoustic sensors as assistance. The combined navigation system consisting of Strapdown Inertial Navigation System (SINS) and Doppler Velocimeter (DVL) is a mainstream and efficient solution in the field of navigation of the submarine. Prior to operation of the integrated navigation system, an initial value of the attitude of the inertial navigation system must be determined, a process known as initial alignment. For a submarine that is docked on shore for navigational preparation, the static base alignment technique is well established. However, the alignment-then-deployment approach limits the fast response capability of the submersible. In order to improve the maneuvering performance, a moving base alignment scheme is adopted in engineering. The core challenge of moving base alignment is that the linear vibration and angular oscillation of the carrier itself submerges useful information such as the rotational angular velocity of the earth, the gravitational acceleration, etc. required for alignment in motion noise, and accurate estimation of inertial navigation attitude must be performed with the aid of external sensors. In particular, when GNSS (global navigation satellite system) is used for assisting alignment, the submarine is required to sail on the water surface, but the concealment cannot be maintained, and in practical application, underwater motion alignment is more prone to be realized only under the assistance of DVL, but the heading angle observability is generally weak. In order to obtain a high-precision alignment result, a combined strategy of inertial system coarse alignment and Kalman filtering fine alignment is generally adopted in engineering, and through years of practice and perfection, the scheme becomes a gold combination for realizing movable base alignment. The main defect of the traditional coarse alignment and fine alignment scheme is that the initial alignment process is split into two serial stages, the alignment time is long, the waste of alignment data exists, and the quick switching of the submarine from an alignment mode to a navigation mode is prevented. In order to solve the problem, researchers design a data storage and backtracking method, and the same segment of alignment data is used for coarse alignment and fine alignment at the same time, so that the data utilization rate is improved, and the alignment time is shortened. However, the filtering calculation in the backtracking scheme must be performed in a centralized manner at the end of alignment after all the data are acquired, which causes a problem of increasing the calculation amount at the end of alignment. If the computing power of the navigation computer is limited, the output of the navigation parameters may be delayed. Another research hotspot of moving base alignment is initial alignment with a large misalignment angle, and the method can reduce the dependence of a Kalman filter on the accuracy of an initial value of an attitude, so that the rough alignment time is greatly shortened. Common nonlinear filtering methods such as unscented kalman filter (kf), bulk kalman filter (CKF), and Particle Filter (PF) are used in attempts to align motion bases. The calculation amount of the nonlinear filtering is several times or even tens times that of the traditional EKF (extended Kalman filtering) algorithm, which is a main obstacle for limiting the popularization and the use of the nonlinear filtering algorithm in engineering. The state transition Kalman filtering (STEKF) algorithm attenuates the time-variant of the filtering system equation by defining the velocity error as a nonlinear form, the inherent logic of which is in communication with the right invariant Kalman filtering. Although the invariant error equation cannot be completely satisfied, the STEKF algorithm can also realize rapid convergence of the attitude error under a large misalignment angle. In addition, the navigation parameters of STEKF algorithm are defined in the local geographic coordinate system and completely agree with the standard navigation parameter representation method. However, in the case of an extreme case where the initial heading er