Initial alignment method and device for inhibiting starting error of inertial device
By employing a multi-segment initial alignment method, combined with Kalman filtering and inertial device data processing, the problem of inertial device startup error affecting initial alignment accuracy was solved, achieving high-precision initial alignment results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-30
- Publication Date
- 2026-03-31
AI Technical Summary
In existing strapdown inertial navigation systems, the rapid temperature rise of inertial components after power-on causes measurement drift errors that severely affect the Kalman filter's correction of initial alignment accuracy.
A multi-segment initial alignment method is adopted, which is divided into coarse alignment and fine alignment stages by establishing the time sequence nodes of the azimuth angle. The basic azimuth angle is determined by using dual-vector attitude determination, and several alignment time periods are divided in the fine alignment stage. Azimuth angle state estimation and data calculation are combined with Kalman filtering. Finally, the initial alignment azimuth angle is formed by fusion calculation.
It effectively suppresses the startup error of inertial devices, improves the initial alignment accuracy of strapdown inertial navigation systems, overcomes startup error levels, and adapts to the inherent structural characteristics of inertial devices.
Smart Images
Figure CN121761932A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of inertial navigation technology, and more specifically to an initial alignment method and apparatus for suppressing startup errors of inertial devices. Background Technology
[0002] Strapdown inertial navigation systems (SINS) can autonomously provide navigation information such as the vehicle's attitude, velocity, and position. Since SINS perform integral calculations based on Newton's laws of motion, initial alignment is required before navigation calculations to autonomously determine its initial orientation. The system's internal inertial devices include three gyroscopes and three accelerometers, and its north-finding accuracy is primarily determined by the equivalent eastward gyroscope drift error. Startup errors in both gyroscopes and accelerometers can lead to equivalent eastward gyroscope drift errors. Existing technologies typically use Kalman filtering to correct for inertial sensor errors and attitude misalignment angles through state estimation. However, when the SINS immediately seeks north after power-on, the rapid temperature rise of the inertial devices after power-on causes significant drift errors in the measurement results, severely impacting the Kalman filter's correction of initial alignment accuracy. Summary of the Invention
[0003] In view of the above problems, embodiments of the present invention provide an initial alignment method and apparatus for suppressing startup errors of inertial devices, thereby solving the technical problem that startup errors of inertial devices have a significant impact after power-on startup.
[0004] The initial alignment method for suppressing startup error of inertial devices according to embodiments of the present invention includes:
[0005] Establish timing nodes for azimuth angles, and form coarse alignment and fine alignment stages after power-on startup based on the timing nodes;
[0006] During the coarse alignment stage, the basic azimuth angle is determined using two-vector attitude determination.
[0007] In the fine alignment stage, several alignment time periods are divided according to the timing nodes. During the alignment time period, the azimuth state is estimated by Kalman filtering and the azimuth state is calculated by combining the inertial device data.
[0008] The initial alignment azimuth is formed by fusion calculation based on the azimuth angles of each alignment time period.
[0009] In one embodiment of the present invention, the time-series nodes are spaced equally.
[0010] In one embodiment of the present invention, the dual vectors are angular velocity vector and linear acceleration vector, and the basic azimuth angle H0 of time node T0 is formed using inertial data between time nodes T0 and T1.
[0011] In one embodiment of the present invention, the Kalman filter uses 10 states as the system state vector: eastward misalignment angle, northward misalignment angle, celestial misalignment angle, eastward velocity error, northward velocity error, eastward gyroscope drift, northward gyroscope drift, celestial gyroscope drift, eastward accelerometer drift, and northward accelerometer drift, and uses eastward velocity error and northward velocity error as the measurement vector.
[0012] In one embodiment of the present invention, the step of dividing several alignment time periods according to time sequence nodes includes:
[0013] The fine alignment phase is divided into three alignment periods, including:
[0014] The I-th alignment time period is constructed using sequential time nodes T0-T1-T2-T3;
[0015] The second alignment period is constructed using sequential time nodes T1-T2-T3;
[0016] The third alignment period is constructed using sequential time nodes T2-T3.
[0017] In one embodiment of the present invention, the step of estimating the azimuth state through Kalman filtering during the alignment period and calculating the azimuth state by combining inertial device data includes:
[0018] During the first alignment period:
[0019] Using the base azimuth angle H0 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T0. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H11 of time node T1.
[0020] Using the estimated azimuth angle H11 as the initial azimuth angle, the strapdown inertial navigation solution is performed using the inertial device data from time node T1 to time node T3 to obtain the solution azimuth angle H13 at time T3.
[0021] During the second alignment period:
[0022] Using the estimated azimuth angle H11 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T1. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H22 of time node T2.
[0023] Using the estimated azimuth angle H22 as the initial azimuth angle, the strapdown inertial navigation solution is performed using the inertial device data from time node T2 to time node T3 to obtain the solution azimuth angle H23 at time T3.
[0024] During the third alignment period:
[0025] Using the estimated azimuth angle H22 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T2. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H33 of time node T3.
[0026] In one embodiment of the present invention, forming the initial alignment azimuth angle includes:
[0027] By fusing and solving the azimuth angles of the same time-series nodes formed in each alignment period, and adjusting the weights using the differences in the azimuth angles of the same time-series nodes, an initial alignment azimuth angle is formed.
[0028] In one embodiment of the present invention, the initial alignment azimuth angle H is:
[0029] H=(H13+H23+H33) / 3+α(H23-H13)+β(H33-H23) (1)
[0030] In the formula, H13 is the calculated azimuth angle at time T3, H23 is the calculated azimuth angle at time T3, H33 is the estimated azimuth angle at time node T3, and the values of α and β are in the range of [0,1], which are set according to the inertial device startup error level.
[0031] An initial alignment device for suppressing startup errors of inertial devices according to an embodiment of the present invention includes:
[0032] A memory for storing program code during the processing of the initial alignment method for suppressing inertial device startup error as described in any one of claims 1 to 8;
[0033] A processor for executing the program code.
[0034] An initial alignment device for suppressing startup errors of inertial devices according to an embodiment of the present invention includes:
[0035] The timing calibration module is used to establish the timing nodes of the azimuth angle, and to form the coarse alignment and fine alignment stages after power-on startup based on the timing nodes;
[0036] The coarse alignment module is used to determine the basic azimuth angle using dual-vector attitude determination during the coarse alignment stage.
[0037] The fine alignment module is used to divide the alignment period into several alignment time periods according to the timing nodes during the fine alignment stage. During the alignment time period, the azimuth state is estimated by Kalman filtering and the azimuth state is calculated by combining the inertial device data.
[0038] The fusion calculation module is used to perform fusion calculations based on the azimuth angles of each alignment time period to form the initial alignment azimuth angle.
[0039] The initial alignment method and apparatus for suppressing inertial device startup errors in this invention presents a multi-segment high-precision initial alignment method that can effectively suppress inertial device startup errors and improve the initial alignment accuracy of strapdown inertial navigation systems. By combining Kalman filtering and inertial device data calculation to form different azimuth angles for the same timing nodes, and utilizing the differences in azimuth angles to construct adjustment weights, this method fully adapts to the error factors caused by the inherent structural characteristics of the inertial devices, overcoming startup error levels. Attached Figure Description
[0040] Figure 1 The diagram shown is a flowchart of an initial alignment method for suppressing startup error of inertial devices according to an embodiment of the present invention.
[0041] Figure 2 The diagram shown is a schematic diagram of the timing azimuth angle of segmented alignment in an initial alignment method for suppressing startup error of inertial devices according to an embodiment of the present invention.
[0042] Figure 3 The diagram shows the initial alignment process of an initial alignment method for suppressing startup error of inertial devices according to an embodiment of the present invention during practical application.
[0043] Figure 4 The diagram shown is a schematic representation of the architecture of an initial alignment device for suppressing startup errors of inertial devices according to an embodiment of the present invention. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0045] An embodiment of the present invention provides an initial alignment method for suppressing startup errors of inertial devices, as follows: Figure 1 As shown. In Figure 1 In this embodiment, the following are included:
[0046] Step 100: Establish the timing nodes of the azimuth angle, and form the coarse alignment and fine alignment stages after power-on startup based on the timing nodes.
[0047] Those skilled in the art will understand that after a strapdown inertial navigation system is powered on, it generates a timing azimuth angle based on inertial data generated by inertial devices, including gyroscopes and accelerometers, through timing planning. A sequence of timing nodes for azimuth angles can be formed, creating a calibrated azimuth angle timing sequence with continuously spaced timing nodes, resulting in a timing azimuth angle sequence. This azimuth angle timing sequence forms continuous coarse alignment and fine alignment stages. The coarse alignment stage is used to quickly determine the approximate azimuth angle of the carrier, while the fine alignment stage uses the approximate azimuth angle to provide reliable initial conditions for Kalman filtering, thereby enabling precise azimuth angle adjustment.
[0048] In one embodiment of the present invention, the time interval of the azimuth sequence can be increasing, decreasing, or equal. The selection of the time interval is related to the inherent characteristics of the electromechanical parameters and inertial device parameters during the initialization of the strapdown inertial navigation system. When the parameters cause the temperature rise to be stable, an equal time interval is selected; when the parameters cause the temperature rise to be rapid, a decreasing time interval is selected; and when the parameters cause the temperature rise to be slow, an increasing time interval is selected.
[0049] Step 200: In the coarse alignment stage, the basic azimuth angle is determined using dual-vector attitude determination.
[0050] In one embodiment of the present invention, inertial data near the initial timing node is used, and the carrier attitude is determined using the two-vector attitude determination principle through two non-collinear vector data, thus determining the basic azimuth angle. For example, an angular velocity vector generated from gyroscope data and a linear acceleration vector generated from accelerometer data are used. Figure 2 As shown, the azimuth angle calculation in the coarse alignment stage at time node T0 uses the inertial data between time nodes T0 and T1 to form the basic azimuth angle H0.
[0051] Step 300: In the fine alignment stage, several alignment time periods are divided according to the timing nodes. In the alignment time period, the azimuth state is estimated by Kalman filtering and the azimuth state is calculated by combining the inertial device data.
[0052] Those skilled in the art will understand that Kalman filtering can achieve accurate correction of inertial sensor errors and attitude misalignment angles through state estimation.
[0053] In one embodiment of the present invention, the basic azimuth angle is used as the initial value. Ten states, namely, eastward misalignment angle, northward misalignment angle, celestial misalignment angle, eastward velocity error, northward velocity error, eastward gyroscope drift, northward gyroscope drift, celestial gyroscope drift, eastward accelerometer drift, and northward accelerometer drift, are used as the system state vector of the Kalman filter, and the eastward velocity error and northward velocity error are used as the measurement vectors.
[0054] In one embodiment of the present invention, azimuth state estimation and calculation are performed using Kalman filtering or a combination of Kalman filtering and inertial device data calculation during each alignment period to form the initial azimuth angle for different alignment periods.
[0055] like Figure 2 As shown, in one embodiment of the present invention, three alignment time periods are formed in the fine alignment stage, including:
[0056] The I-th alignment time period is constructed using sequential time nodes T0-T1-T2-T3;
[0057] The second alignment period is constructed using sequential time nodes T1-T2-T3;
[0058] The third alignment period is constructed using sequential time nodes T2-T3.
[0059] like Figure 2 As shown, in one embodiment of the present invention, the process of performing state estimation and data calculation using Kalman filtering and inertial device data during each alignment period includes:
[0060] During the first alignment period:
[0061] Using the base azimuth angle H0 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T0. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H11 of time node T1.
[0062] Using the estimated azimuth angle H11 as the initial azimuth angle, the strapdown inertial navigation solution is obtained by using the inertial device data from time node T1 to time node T3, and the solution azimuth angle H13 at time T3 is obtained.
[0063] During the second alignment period:
[0064] Using the estimated azimuth angle H11 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T1. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H22 of time node T2.
[0065] Using the estimated azimuth angle H22 as the initial azimuth angle, the strapdown inertial navigation solution is obtained by using the inertial device data from time node T2 to time node T3, and the solution azimuth angle H23 at time T3 is obtained.
[0066] During the third alignment period:
[0067] Using the estimated azimuth angle H22 as the initial azimuth angle, the Kalman filter parameters are initialized starting from time node T2. The state is estimated using the Kalman filter to obtain the estimated azimuth angle H33 of time node T3.
[0068] Step 400: Perform fusion calculation based on the azimuth angles of each alignment time period to form the initial alignment azimuth angle.
[0069] By fusing and solving the azimuth angles of the same time-series nodes formed in each alignment period, and adjusting the weights using the differences in the azimuth angles of the same time-series nodes, an initial alignment azimuth angle is formed.
[0070] In one embodiment of the present invention, the initial alignment azimuth angle H is formed as follows:
[0071] H=(H13+H23+H33) / 3+α(H23-H13)+β(H33-H23) (1)
[0072] In the formula, the values of α and β are in the range of [0,1], and are set according to the startup error level of the inertial device.
[0073] The initial alignment method for suppressing inertial device startup errors in this invention presents a multi-segment high-precision initial alignment method that can effectively suppress inertial device startup errors and improve the initial alignment accuracy of strapdown inertial navigation systems. By combining Kalman filtering and inertial device data calculation to form different azimuth angles for the same timing nodes, and utilizing the differences in azimuth angles to construct adjustment weights, this method fully adapts to the error factors caused by the inherent structural characteristics of the inertial devices, overcoming startup error levels.
[0074] The initial alignment process formed using the initial alignment method of the above embodiments is as follows: Figure 3 As shown. In Figure 3 Including:
[0075] (1) Power on the strapdown inertial navigation system;
[0076] (2) Using the inertial data between T0 and T1, the angular velocity vector formed by the gyroscope data and the linear acceleration vector formed by the accelerometer data are obtained. The azimuth angle H0 is obtained by using the dual-vector attitude determination principle.
[0077] (3) Segmented fine alignment I: Using the azimuth angle H0 obtained from coarse alignment as the initial value, starting from time node T0, the state estimation is performed using the Kalman filter algorithm to obtain the azimuth angle H11 at time node T1. Using H11 as the initial azimuth angle, the strapdown inertial navigation solution is performed using the inertial device data from time node T1 to T3 to obtain the azimuth angle H13 at time node T3.
[0078] Segmented fine alignment II: Using the azimuth angle H11 obtained from segmented fine alignment I as the initial value, starting from time T1, the Kalman filter parameters are initialized, and the state estimation is performed using the Kalman filter algorithm to obtain the azimuth angle H22 at time T2; using H22 as the initial azimuth angle, the strapdown inertial navigation solution is performed using the inertial device data from T2 to T3 to obtain the azimuth angle H23 at time T3.
[0079] Segmented fine alignment III: Using the azimuth angle H22 obtained from segmented fine alignment II as the initial value, starting from time T2, the Kalman filter parameters are initialized, and the state estimation is performed using the Kalman filter algorithm to obtain the azimuth angle H33 at time T3.
[0080] (4) Perform azimuth data fusion and settlement. The initial alignment azimuth angle H = (H13 + H23 + H33) / 3 + α(H23 - H13) + β(H33 - H23), where α and β are in the range of [0,1] and are set according to the inertial device startup error.
[0081] An embodiment of the present invention provides an initial alignment device for suppressing startup errors of inertial devices, comprising:
[0082] The memory is used to store the program code of the initial alignment method for suppressing the startup error of the inertial device in the above embodiments during the processing.
[0083] A processor for executing program code during the processing of the initial alignment method for suppressing inertial device startup errors in the above embodiments.
[0084] The processor can be a DSP (Digital Signal Processor), an FPGA (Field-Programmable Gate Array), an MCU (Microcontroller Unit) system board, a SoC (System on a Chip) system board, or a PLC (Programmable Logic Controller) minimum system including I / O.
[0085] An embodiment of the present invention provides an initial alignment device for suppressing startup errors of inertial devices, such as... Figure 4 As shown. In Figure 4 In this embodiment, the following are included:
[0086] The timing calibration module 10 is used to establish the timing nodes of the azimuth angle, and to form the coarse alignment and fine alignment stages after power-on startup based on the timing nodes;
[0087] The coarse alignment module 20 is used to determine the basic azimuth angle using dual-vector attitude determination during the coarse alignment stage.
[0088] The fine alignment module 30 is used to divide the alignment period into several alignment periods according to the timing nodes during the fine alignment stage. During the alignment period, the azimuth state is estimated by Kalman filtering and the azimuth state is calculated by combining the inertial device data.
[0089] The fusion calculation module 40 is used to perform fusion calculation based on the azimuth angles of each alignment time period to form the initial alignment azimuth angle.
[0090] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method of initial alignment to suppress start-up errors of inertial devices, characterized by, The application relates to a method for establishing an initial alignment azimuth angle. The method comprises the following steps: establishing time nodes for azimuth angle, forming a coarse alignment and a fine alignment phase after power-on starting according to the time nodes; in the coarse alignment phase, a basic azimuth angle is determined by using double-vector orientation determination; in the fine alignment phase, a plurality of alignment time periods are divided according to the time nodes, azimuth angle state estimation is performed in the alignment time periods by using Kalman filtering, and azimuth angle state calculation is performed in combination with inertial device data; 2. The method of initial alignment to suppress inertial device start-up errors as recited in claim 1, wherein, the initial alignment azimuth angle is formed by fusing and calculating the azimuth angles in the alignment time periods.
3. The method of initial alignment to suppress inertial device start-up errors as recited in claim 1, wherein, The time nodes are equal in interval.
4. The method of initial alignment to suppress inertial device start-up errors as recited in claim 1, wherein, The double vector adopts an angular velocity vector and a linear acceleration vector, and the basic azimuth angle H0 at the time node T0 is formed by using inertial data between the time nodes T0 and T1.
5. The method of initial alignment to suppress inertial device start-up errors as recited in claim 1, wherein, The Kalman filtering takes 10 states of an east misalignment angle, a north misalignment angle, a sky misalignment angle, an east velocity error, a north velocity error, an east gyroscope drift, a north gyroscope drift, a sky gyroscope drift, an east accelerometer drift and a north accelerometer drift as a system state vector, and takes the east velocity error and the north velocity error as a measurement vector. The step of dividing a plurality of alignment time periods according to the time nodes comprises the following steps: three alignment time periods are formed in the fine alignment phase, comprising the following steps: an Ith alignment time period is constructed by using sequential time nodes T0-T1-T2-T3; a IIth alignment time period is constructed by using sequential time nodes T1-T2-T3; 6. The method of initial alignment to suppress inertial device start-up errors according to claim 5, wherein, a IIIth alignment time period is constructed by using sequential time nodes T2-T3. The step of performing azimuth angle state estimation in the alignment time periods by using Kalman filtering and performing azimuth angle state calculation in combination with inertial device data comprises the following steps: in the Ith alignment time period: the basic azimuth angle H0 is used as an initial azimuth angle, Kalman filtering parameters are initialized from the time node T0, state estimation is performed by using Kalman filtering, and an estimated azimuth angle H11 of the time node T1 is obtained; the estimated azimuth angle H11 is used as an initial azimuth angle, strapdown inertial navigation calculation is performed by using inertial device data from the time node T1 to the time node T3, and a calculated azimuth angle H13 of the time node T3 is obtained; in the IIth alignment time period: the estimated azimuth angle H11 is used as an initial azimuth angle, Kalman filtering parameters are initialized from the time node T1, state estimation is performed by using Kalman filtering, and an estimated azimuth angle H22 of the time node T2 is obtained; the estimated azimuth angle H22 is used as an initial azimuth angle, strapdown inertial navigation calculation is performed by using inertial device data from the time node T2 to the time node T3, and a calculated azimuth angle H23 of the time node T3 is obtained; in the IIIth alignment time period:
7. The method of initial alignment to suppress inertial device start-up errors according to claim 1, wherein, the estimated azimuth angle H22 is used as an initial azimuth angle, Kalman filtering parameters are initialized from the time node T2, state estimation is performed by using Kalman filtering, and an estimated azimuth angle H33 of the time node T3 is obtained. The step of forming the initial alignment azimuth angle comprises the following steps:
8. The method of initial alignment to suppress inertial device start-up errors according to claim 7, wherein, the initial alignment azimuth angle is formed by fusing and calculating the azimuth angles of the same time nodes in the alignment time periods, and the weight is adjusted by using the azimuth angle difference of the same time nodes. The initial alignment azimuth angle H is: H=(H13+H23+H33) / 3+alpha(H23-H13)+beta(H33-H23) (1) In the formula, H13 is the calculated azimuth angle at T3, H23 is the calculated azimuth angle at T3, H33 is the estimated azimuth angle at the time sequence node T3, the value range of α and β is [0, 1], and the value is set according to the inertial device start error level.
9. An initial alignment device for suppressing start-up errors of inertial devices, characterized in that The method comprises the steps that: a memory is used for storing the program code of the initial alignment method for suppressing the start error of the inertial device in the processing process as claimed in any one of claims 1 to 8; a processor is used for executing the program code.
10. An initial alignment device for suppressing start-up errors of inertial devices, characterized in that The method comprises the steps that: a time marking module is used for establishing a time sequence node of the azimuth angle, and forming the coarse alignment and fine alignment stages after the power-on start according to the time sequence node; a coarse alignment module is used for determining the basic azimuth angle by using the double vector attitude determination in the coarse alignment stage; a fine alignment module is used for dividing a plurality of alignment time periods according to the time sequence node in the fine alignment stage, performing the azimuth angle state estimation by using the Kalman filtering in the alignment time period, and performing the azimuth angle state calculation in combination with the inertial device data; a fusion calculation module is used for performing the fusion calculation according to the azimuth angle of each alignment time period to form the initial alignment azimuth angle.