Method for quickly calibrating installation matrix based on error cancellation

文档序号：1950524 发布日期：2021-12-10 浏览：21次中文

阅读说明：本技术 一种基于误差对消的安装矩阵快速标定方法 (Method for quickly calibrating installation matrix based on error cancellation ) 是由冯士伟林松刘燕锋周益徐李佳郅银周王月韩璐邱金娟张沛勇张心悦于 2021-09-14 设计创作，主要内容包括：本发明公开了一种基于误差对消的安装矩阵快速标定方法,所述方法包括如下步骤：将IMU的陀螺安装矩阵[G]-(CG)进行陀螺输入轴坐标系到机械坐标系的转换得到机械坐标系下的IMU的陀螺安装矩阵[G]-(mG)；将IMU的加计安装矩阵[A]-(CA)进行加计输入轴坐标系到机械坐标系的转换得到机械坐标系下的IMU的加计安装矩阵[A]-(mA)；将机械坐标系下的IMU的陀螺安装矩阵[G]-(mG)通过机械坐标系到基准镜坐标系的转换矩阵得到基准镜坐标系的IMU的陀螺安装矩阵将机械坐标系下的IMU的加计安装矩阵[A]-(mA)通过机械坐标系到基准镜坐标系的转换矩阵得到基准镜坐标系的IMU的加计安装矩阵本发明有效提升IMU导航的姿态精度、速度精度、位置精度,亦可通过自动化测试序列的编制,有效提升IMU产品的标定测试效率。(The invention discloses an installation matrix rapid calibration method based on error cancellation, which comprises the following steps: mounting the gyroscope of an IMU to a matrix [ G ]] CG Converting the gyroscope input shaft coordinate system to a mechanical coordinate system to obtain a gyroscope installation matrix G of the IMU under the mechanical coordinate system] mG (ii) a Add the meter installation matrix [ A ] of IMU] CA Converting the coordinate system of the adding input shaft into the mechanical coordinate system to obtain an adding installation matrix [ A ] of the IMU under the mechanical coordinate system] mA (ii) a Installing a gyro of the IMU under a mechanical coordinate system into a matrix G] mG Obtaining a gyro mounting matrix of an IMU of a reference mirror coordinate system through a conversion matrix from a mechanical coordinate system to the reference mirror coordinate system Adding and installing matrix [ A ] of IMU under mechanical coordinate system] mA Obtaining an addition installation matrix of the IMU of the reference mirror coordinate system through a conversion matrix from the mechanical coordinate system to the reference mirror coordinate system The invention effectively improves the attitude precision, speed precision and position precision of IMU navigation, and can also effectively improve the calibration test efficiency of IMU products through the compilation of an automatic test sequence.)

1. An installation matrix rapid calibration method based on error cancellation is characterized by comprising the following steps:

the method comprises the following steps: after being installed on a regular hexahedron tool, an IMU product is placed on a test turntable;

step two: defining 6 postures from a testing turntable TM system to a hexahedral tooling C system in the testing process of 3 gyro installation matrixes;

step three: recording the pulse output of 3 gyros under 6 postures, and obtaining a gyro installation matrix G of the IMU according to the pulse output of 3 gyros under 6 postures]_CG；

Step four: defining 24 postures from a testing turntable TM system to a hexahedral tool C system in the testing process of 3 accelerometer installation matrixes;

step five: recording pulse outputs of three accelerometers under twenty-four additional measurement postures, and obtaining an additional measurement installation matrix [ A ] of the IMU according to the pulse outputs of the three accelerometers under the twenty-four additional measurement postures]_CA；

Step six: mounting the gyroscope of an IMU to a matrix [ G ]]_CGConverting the gyroscope input shaft coordinate system to a mechanical coordinate system to obtain a gyroscope installation matrix G of the IMU under the mechanical coordinate system]_mG；

Add the meter installation matrix [ A ] of IMU]_CAConverting the coordinate system of the adding input shaft into the mechanical coordinate system to obtain an adding installation matrix [ A ] of the IMU under the mechanical coordinate system]_mA；

Step seven: installing a gyro of the IMU under a mechanical coordinate system into a matrix G]_mGObtaining a gyro mounting matrix of an IMU of a reference mirror coordinate system through a conversion matrix from a mechanical coordinate system to the reference mirror coordinate system

Adding and installing matrix [ A ] of IMU under mechanical coordinate system]_mAObtaining an addition installation matrix of the IMU of the reference mirror coordinate system through a conversion matrix from the mechanical coordinate system to the reference mirror coordinate system

2. The method for rapidly calibrating the installation matrix based on error cancellation according to claim 1, characterized in that: in the second step, 6 postures from the testing of the rotary table TM system to the hexahedral tooling C system in the testing process of the 3 gyro installation matrixes are as follows:

a first gyro attitude: the input shaft of the first gyro is vertical to the table surface of the rotary table and faces downwards, and the gyro conversion array from the rotary table TM to the hexahedral tooling C is

A second gyro attitude: the input shaft of the first gyro is vertical to the table surface of the rotary table and upwards, and the gyro conversion array from the rotary table TM to the hexahedral tooling C is

The third gyro attitude: the input shaft of the second gyro is vertical to the table surface of the rotary table and faces downwards, and the gyro conversion array from the rotary table TM system to the hexahedral tooling C system is

A fourth gyro attitude: the input shaft of the second gyro is vertical to the table surface of the rotary table and upwards, and the gyro conversion array from the rotary table TM to the hexahedral tooling C is

And a fifth gyro attitude: the input shaft of the third gyro is vertical to the table surface of the rotary table and faces downwards, and the gyro conversion array from the rotary table TM system to the hexahedral tooling C system is

A sixth gyro attitude: the input shaft of the third gyro is vertical to the table surface of the rotary table and upwards, and the gyro conversion array from the rotary table TM system to the hexahedral tooling C system is

3. The method for rapidly calibrating the installation matrix based on error cancellation according to claim 2, characterized in that: in the third step, a gyro installation matrix [ G ] of the IMU is obtained according to the pulse output of 3 gyros under 6 postures]_CGThe method comprises the following steps:

step (31): recording the first forward rotation pulse output of the first gyro, the first forward rotation pulse output of the second gyro, the first forward rotation pulse output of the third gyro during forward rotation of the test turntable during attitude of the first gyro, the first reverse rotation pulse output of the first gyro during reverse rotation of the test turntable, the first reverse rotation pulse output of the second gyro and the first reverse rotation pulse output of the third gyro during attitude of the first gyro to obtain the nonlinear error epsilon of the scale factor of the first gyro during attitude of the first gyro_g1 ¹And a projection theta of the second gyro installation deviation in the first gyro direction at the first gyro attitude_g21 ¹And a projection theta of a third gyro mounting deviation in the first gyro direction at the first gyro attitude_g31 ¹；

Step (32): recording second forward rotation pulse output of the first gyro, second forward rotation pulse output of the second gyro, second forward rotation pulse output of the third gyro during forward rotation of the test turntable during attitude of the second gyro, second reverse rotation pulse output of the first gyro during reverse rotation of the test turntable, second reverse rotation pulse output of the second gyro and second reverse rotation pulse output of the third gyro to obtain first gyro scale factor nonlinear error epsilon during attitude of the second gyro_g1 ²And a projection theta of the second gyro installation deviation in the first gyro direction at the second gyro attitude_g21 ²And a projection theta of the third gyro mounting deviation in the first gyro direction at the second gyro attitude_g31 ²；

Step (33): first gyro scale factor nonlinear error epsilon from first gyro attitude_g1 ¹And a projection theta of the second gyro installation deviation in the first gyro direction at the first gyro attitude_g21 ¹And a projection theta of the third gyro installation deviation in the first gyro direction at the first gyro attitude_g31 ¹First gyro scale factor is not in second gyro attitudeLinear error epsilon_g1 ²And a projection theta of the second gyro installation deviation in the first gyro direction at the second gyro attitude_g21 ²And a projection theta of a third gyro mounting deviation in the first gyro direction at the second gyro attitude_g31 ²Obtaining the average value epsilon of the nonlinear errors of the first gyro scale factor when the first gyro attitude and the second gyro attitude are obtained_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21Projection average value theta of third gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g31；

Step (34): recording third forward pulse output of the first gyro, third forward pulse output of the second gyro, third forward pulse output of the third gyro, third reverse pulse output of the first gyro when the test rotary table rotates forwards and the test rotary table rotates backwards, third reverse pulse output of the second gyro and third reverse pulse output of the third gyro when the test rotary table rotates backwards to obtain second gyro scale factor nonlinear error epsilon when the third gyro is in posture_g2 ³And a projection theta of the first gyro mounting deviation in the second gyro direction at the third gyro attitude_g12 ³And a projection theta of a third gyro mounting deviation in a second gyro direction at a third gyro attitude_g32 ³；

Step (35): recording the fourth forward pulse output of the first gyro, the fourth forward pulse output of the second gyro, the fourth forward pulse output of the third gyro during forward rotation of the test turntable during attitude of the fourth gyro, the fourth reverse pulse output of the first gyro during reverse rotation of the test turntable, the fourth reverse pulse output of the second gyro and the fourth reverse pulse output of the third gyro during reverse rotation of the test turntable, and obtaining the second gyro scale factor nonlinear error epsilon during attitude of the fourth gyro_g1 ⁴And the projection theta of the first gyro installation deviation in the second gyro direction in the fourth gyro attitude_g12 ⁴And a projection theta of the third gyro mounting deviation in the second gyro direction at the fourth gyro attitude_g32 ⁴；

Step (36): according to the third gyroscope attitudeNon-linear error epsilon of spiral scale factor_g2 ³And a projection theta of the first gyro mounting deviation in the second gyro direction at the third gyro attitude_g12 ³And a projection theta of a third gyro mounting deviation in the first gyro direction at the third gyro attitude_g31 ³First gyro scale factor nonlinear error epsilon in fourth gyro attitude_g1 ⁴And the projection theta of the first gyro installation deviation in the second gyro direction in the fourth gyro attitude_g12 ⁴And a projection theta of the third gyro mounting deviation in the first gyro direction at the fourth gyro attitude_g31 ⁴Obtaining the average value epsilon of the nonlinear errors of the second gyro scale factors when the third gyro attitude and the fourth gyro attitude are obtained_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32；

Step (37): recording fifth forward pulse output of the first gyro, fifth forward pulse output of the second gyro, fifth forward pulse output of the third gyro during forward rotation of the test turntable during attitude of the fifth gyro, fifth reverse pulse output of the first gyro during reverse rotation of the test turntable, fifth reverse pulse output of the second gyro and fifth reverse pulse output of the third gyro during reverse rotation of the test turntable, and obtaining third gyro scale factor nonlinear error epsilon during attitude of the fifth gyro_g3 ⁵And a projection theta of the first gyro mounting deviation in the third gyro direction at the fifth gyro attitude_g13 ⁵And a projection theta of the second gyro mounting deviation in the third gyro direction at the fifth gyro attitude_g23 ⁵；

Step (38): recording sixth forward pulse output of the first gyro, sixth forward pulse output of the second gyro, sixth forward pulse output of the third gyro and sixth reverse pulse output of the first gyro, sixth reverse pulse output of the second gyro and sixth reverse pulse output of the third gyro in the test turntable forward rotation process when the sixth gyro is in posture, and obtaining third gyro scale factor nonlinear error epsilon in the sixth gyro posture_g3 ⁶And a projection theta of the first gyro mounting deviation in the third gyro direction at the sixth gyro attitude_g13 ⁶And a projection theta of the second gyro mounting deviation in the third gyro direction at the sixth gyro attitude_g23 ⁶；

Step (39): third gyro scale factor nonlinear error epsilon from fifth gyro attitude_g3 ⁵And a projection theta of the first gyro mounting deviation in the third gyro direction at the fifth gyro attitude_g13 ⁵And a projection theta of the second gyro mounting deviation in the third gyro direction at the fifth gyro attitude_g23 ⁵Third gyro scale factor nonlinear error epsilon at sixth gyro attitude_g3 ⁶And a projection theta of the first gyro mounting deviation in the third gyro direction at the sixth gyro attitude_g13 ⁶And a projection theta of the second gyro mounting deviation in the third gyro direction at the sixth gyro attitude_g23 ⁶Obtaining the average value epsilon of the nonlinear errors of the scale factors of the third gyroscope in the fifth gyroscope attitude and the sixth gyroscope attitude_g2Projected average value theta of first gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g13Projected average value theta of second gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g23；

Step (310): first gyro scale factor nonlinear error mean epsilon from first and second gyro attitudes_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21Projection average value theta of third gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g31Mean value epsilon of nonlinear errors of the second gyro scaling factor at the third gyro attitude and the fourth gyro attitude_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32Fifth gyro attitude and sixth gyro attitudeThird gyro scale factor nonlinear error mean value epsilon in gyro attitude_g2Projected average value theta of first gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g13Projected average value theta of second gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g23Obtaining a gyro mounting matrix [ G ] of the IMU]_CG。

4. The method for rapidly calibrating the installation matrix based on error cancellation according to claim 1, characterized in that: in the fourth step, 24 postures from the test turntable TM system to the hexahedral tooling C system in the test process of the 3 accelerometer installation matrixes are as follows:

1 st addition posture: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Adding the attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

And 3, adding and calculating the attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

And 4, adding and calculating the attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Adding posture at 5 th: input axis of the first accelerometer is verticalThe accelerometer conversion array from the TM system of the rotary table to the C system of the hexahedral tool is arranged upwards on the table surface of the rotary table

Adding and calculating the attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding and counting attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding the attitude: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding posture at 9 th: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tooling C system is

Adding posture at 10 th: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tooling C system is

11 th addition posture: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tooling C system is

12 th addition posture: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tooling C system is

13 th addition attitude: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

The 14 th adding posture: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding posture at 15: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding posture at 16 th: the input shaft of the second accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding posture at 17 th: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

18 th addition attitude: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

19 th addition posture: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Addition posture 20: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and faces downwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Addition posture of 21 st: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and upwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

The 22 nd adding posture: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and upwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Adding posture at 23: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and upwards, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Adding posture at 24 th: the input shaft of the third accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is connected with the hexahedral tool CSpeedometer conversion array as

5. The method for rapidly calibrating the installation matrix based on error cancellation according to claim 4, characterized in that: in the fifth step, an adding and metering installation matrix [ A ] of the IMU is obtained according to the pulse output of the three accelerometers under the twenty-four adding and metering postures]_CAThe method comprises the following steps:

step (51): according to pulse outputs of three accelerometers under the 1 st addition attitude, the 2 nd addition attitude, the 3 rd addition attitude, the 4 th addition attitude, the 5 th addition attitude, the 6 th addition attitude, the 7 th addition attitude and the 8 th addition attitude, obtaining a first accelerometer scale factor nonlinear error average value epsilon when the 1 st addition attitude is changed into the 8 th addition attitude_a1And the projection average value theta of the installation deviation of the second accelerometer in the direction of the first accelerometer from the 1 st addition posture to the 8 th addition posture_a21And the projection average value theta of the installation deviation of the third accelerometer in the direction of the first accelerometer from the 1 st addition posture to the 8 th addition posture_a31；

Step (52): according to pulse outputs of three accelerometers under the 9 th adding posture, the 10 th adding posture, the 11 th adding posture, the 12 th adding posture, the 13 th adding posture, the 14 th adding posture, the 15 th adding posture and the 16 th adding posture, the average value epsilon of the nonlinear errors of the scaling factors of the second accelerometer from the 9 th adding posture to the 16 th adding posture is obtained_a2And the projection average value theta of the installation deviation of the first accelerometer in the direction of the second accelerometer from the 9 th adding posture to the 16 th adding posture_a12And the projection average value theta of the installation deviation of the third accelerometer in the direction of the second accelerometer from the 9 th adding posture to the 16 th adding posture_a32；

Step (53): according to pulse outputs of three accelerometers under the 17 th adding posture, the 18 th adding posture, the 19 th adding posture, the 20 th adding posture, the 21 st adding posture, the 22 th adding posture, the 23 th adding posture and the 24 th adding posture, obtaining a third adding posture from the 17 th adding posture to the 24 th adding postureAccelerometer scale factor nonlinear error mean epsilon_a3The projection average value theta of the installation deviation of the first accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a13The projection average value theta of the installation deviation of the second accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a23；

Step (54): the mean value epsilon of the nonlinear errors of the scaling factors of the first accelerometer according to the 1 st to 8 th addition postures_a1And the projection average value theta of the installation deviation of the second accelerometer in the direction of the first accelerometer from the 1 st addition posture to the 8 th addition posture_a21And the projection average value theta of the installation deviation of the third accelerometer in the direction of the first accelerometer from the 1 st addition posture to the 8 th addition posture_a31And the mean value epsilon of the nonlinear errors of the scaling factors of the second accelerometer from the 9 th adding posture to the 16 th adding posture_a2And the projection average value theta of the installation deviation of the first accelerometer in the direction of the second accelerometer from the 9 th adding posture to the 16 th adding posture_a12And the projection average value theta of the installation deviation of the third accelerometer in the direction of the second accelerometer from the 9 th adding posture to the 16 th adding posture_a32Third accelerometer scaling factor nonlinear error mean epsilon from 17 th accelerometer attitude to 24 th accelerometer attitude_a3The projection average value theta of the installation deviation of the first accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a13The projection average value theta of the installation deviation of the second accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a23Obtaining an addition installation matrix [ A ] of the IMU]_CA。

6. The method for rapidly calibrating the installation matrix based on error cancellation according to claim 1, characterized in that: in step three, the gyro mounting matrix [ G ] of the IMU]_CGComprises the following steps:

7. the method for rapidly calibrating the installation matrix based on error cancellation according to claim 1, characterized in that: in step five, the IMU's additively installed matrix [ A ]]_CAComprises the following steps:

8. the method for rapidly calibrating the installation matrix based on error cancellation according to claim 3, characterized in that: in step (33), the mean value of the nonlinear errors ε of the first gyro scaling factors at the first gyro attitude and the second gyro attitude_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21Projection average value theta of third gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g31Comprises the following steps:

9. the method for rapidly calibrating the installation matrix based on error cancellation according to claim 3, characterized in that: in step (36), the mean value of the nonlinear errors ε of the second gyro scale factors at the third gyro attitude and the fourth gyro attitude_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32Comprises the following steps:

10. the method for rapidly calibrating the installation matrix based on error cancellation according to claim 3, characterized in that: in step (39), third gyro scale factor nonlinear error mean ε at fifth and sixth gyro attitudes_g2Projected average value theta of first gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g13Projected average value theta of second gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g23Comprises the following steps:

Technical Field

The invention belongs to the technical field of calibration test of an Inertial Measurement Unit (IMU) product, and particularly relates to a method for rapidly calibrating an installation matrix based on error cancellation.

Background

In the process of installing a whole machine, an input shaft of a gyroscope and an accelerometer of an inertial measurement unit (hereinafter abbreviated as IMU) has certain deviation relative to an ideal position, and an installation matrix of a product forms a non-orthogonal matrix due to the deviation. The IMU navigation precision in the landing process of the landing cabin is influenced by the above conditions, and the larger the installation deviation is, the higher the non-orthogonality is, and the more obvious the influence on the IMU navigation precision is. With the deep development of the deep space detection technology in China, the demand for calibrating and testing the IMU installation matrix with high precision and high efficiency is increasing day by day.

In a traditional IMU (inertial measurement Unit) installation matrix testing method, a gyroscope is subjected to rate calibration testing based on a single-axis rotary table, an accelerometer is subjected to position calibration testing based on a marble platform, reference information of the single-axis rotary table and the marble platform is different, so that relative postures of the gyroscope and the accelerometer installation matrix have certain uncertainty, and IMU inertial navigation precision is influenced. Meanwhile, the two sets of reference equipment are respectively tested, so that the pose state repetition rate is high, the testing efficiency is low, and the calibration test of the IMU installation matrix is inconvenient to develop in high efficiency and in batch.

Disclosure of Invention

The technical problem solved by the invention is as follows: the method overcomes the defects of the prior art, provides a quick calibration method of the installation matrix based on error cancellation, effectively improves the attitude precision, speed precision and position precision of IMU navigation, and can also effectively improve the calibration test efficiency of IMU products through compilation of an automatic test sequence.

The purpose of the invention is realized by the following technical scheme: an installation matrix rapid calibration method based on error cancellation comprises the following steps: the method comprises the following steps: after being installed on a regular hexahedron tool, an IMU product is placed on a test turntable; step two: defining 6 postures from a testing turntable TM system to a hexahedral tooling C system in the testing process of 3 gyro installation matrixes; step three: recording the pulse output of 3 gyros under 6 postures, and obtaining a gyro installation matrix G of the IMU according to the pulse output of 3 gyros under 6 postures]_CG(ii) a Step four: defining 24 postures from a testing turntable TM system to a hexahedral tool C system in the testing process of 3 accelerometer installation matrixes: 1 st addition posture, 2 nd addition posture, 3 rd addition posture, 4 th addition posture, 5 th addition posture, 6 th addition posture, 7 th addition posture, 8 th addition posture, 9 th addition posture, 10 th addition posture, 11 th addition posture, 12 th addition posture, 13 th addition posture, 14 th addition posture, 15 th addition posture, 16 th addition posture, 17 th addition posture, 18 th addition posture, 19 th addition posture, 20 th addition posture, 21 st addition posture, 22 th addition posture, 23 th addition posture and 24 th addition posture; step five: recording pulse outputs of three accelerometers under twenty-four additional measurement postures, and obtaining an additional measurement installation matrix [ A ] of the IMU according to the pulse outputs of the three accelerometers under the twenty-four additional measurement postures]_CA(ii) a Step six: mounting the gyroscope of an IMU to a matrix [ G ]]_CGConverting the gyroscope input shaft coordinate system to a mechanical coordinate system to obtain a gyroscope installation matrix G of the IMU under the mechanical coordinate system]_mG(ii) a Add-on mounting of IMUMatrix [ A ]]_CAConverting the coordinate system of the adding input shaft into the mechanical coordinate system to obtain an adding installation matrix [ A ] of the IMU under the mechanical coordinate system]_mA(ii) a Step seven: installing a gyro of the IMU under a mechanical coordinate system into a matrix G]_mGObtaining a gyro installation matrix C of the IMU of the reference mirror coordinate system through a conversion matrix from the mechanical coordinate system to the reference mirror coordinate system_Gm; adding and installing matrix [ A ] of IMU under mechanical coordinate system]_mAObtaining an addition installation matrix C of the IMU of the reference mirror coordinate system through a conversion matrix from the mechanical coordinate system to the reference mirror coordinate system_Am。

In the method for rapidly calibrating the installation matrix based on error cancellation, in the second step, 6 postures from the testing turntable TM system to the hexahedral tooling C system in the testing process of the 3 gyro installation matrices are as follows:

And a fifth gyro attitude: the input shaft of the third top is vertical to the table top of the rotary tableLower, the gyro conversion array from the rotary table TM system to the hexahedral tooling C system is

In the method for quickly calibrating the installation matrix based on error cancellation, in the third step, the gyro installation matrix [ G ] of the IMU is obtained according to the pulse output of 3 gyros under 6 postures]_CGThe method comprises the following steps:

step (31): recording the first forward rotation pulse output of the first gyro, the first forward rotation pulse output of the second gyro, the first forward rotation pulse output of the third gyro during forward rotation of the test turntable during attitude of the first gyro, the first reverse rotation pulse output of the first gyro during reverse rotation of the test turntable, the first reverse rotation pulse output of the second gyro and the first reverse rotation pulse output of the third gyro during attitude of the first gyro to obtain the nonlinear error epsilon of the scale factor of the first gyro during attitude of the first gyro_g11. Projection theta of second gyro installation deviation in first gyro direction in first gyro attitude_g21 ¹And a projection theta of a third gyro mounting deviation in the first gyro direction at the first gyro attitude_g31 ¹；

Step (33): first gyro scale factor nonlinear error epsilon from first gyro attitude_g1 ¹And a projection theta of the second gyro installation deviation in the first gyro direction at the first gyro attitude_g21 ¹And a projection theta of the third gyro installation deviation in the first gyro direction at the first gyro attitude_g31 ¹First gyro scale factor nonlinear error epsilon in second gyro attitude_g1 ²And a projection theta of the second gyro installation deviation in the first gyro direction at the second gyro attitude_g21 ²And a projection theta of a third gyro mounting deviation in the first gyro direction at the second gyro attitude_g31 ²Obtaining the average value epsilon of the nonlinear errors of the first gyro scale factor when the first gyro attitude and the second gyro attitude are obtained_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21Projection average value theta of third gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g31；

Step (35): recording the fourth forward pulse output of the first gyro, the fourth forward pulse output of the second gyro, the fourth forward pulse output of the third gyro and the fourth reverse pulse output of the first gyro, the fourth reverse pulse output of the second gyro and the third gyro during the forward rotation of the test turntable and the reverse rotation of the test turntableOutputting the fourth inversion pulse of the spiral to obtain the nonlinear error epsilon of the second gyro scale factor when the fourth gyro is in the attitude_g1 ⁴And the projection theta of the first gyro installation deviation in the second gyro direction in the fourth gyro attitude_g12 ⁴And a projection theta of the third gyro mounting deviation in the second gyro direction at the fourth gyro attitude_g32 ⁴；

Step (36): first gyro scale factor nonlinear error epsilon from third gyro attitude_g2 ³And a projection theta of the first gyro mounting deviation in the second gyro direction at the third gyro attitude_g12 ³And a projection theta of a third gyro mounting deviation in the first gyro direction at the third gyro attitude_g31 ³First gyro scale factor nonlinear error epsilon in fourth gyro attitude_g1 ⁴And the projection theta of the first gyro installation deviation in the second gyro direction in the fourth gyro attitude_g12 ⁴And a projection theta of the third gyro mounting deviation in the first gyro direction at the fourth gyro attitude_g31 ⁴Obtaining the average value epsilon of the nonlinear errors of the second gyro scale factors when the third gyro attitude and the fourth gyro attitude are obtained_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32；

Step (310): first gyro scale factor nonlinear error mean epsilon from first and second gyro attitudes_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21A third gyro mounting deviation at the first gyro attitude during the first gyro attitude and the second gyro attitudeProjected mean value theta of gyro direction_g31Mean value epsilon of nonlinear errors of the second gyro scaling factor at the third gyro attitude and the fourth gyro attitude_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32Third gyro scale factor nonlinear error mean epsilon at fifth gyro attitude and sixth gyro attitude_g2Projected average value theta of first gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g13Projected average value theta of second gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g23Obtaining a gyro mounting matrix [ G ] of the IMU]_CG。

In the method for quickly calibrating the installation matrix based on error cancellation, in the fourth step, 24 postures from the test turntable TM system to the hexahedral tooling C system in the test process of the 3 accelerometer installation matrices are as follows:

And 4, adding and calculating the attitude: the input axis of the first accelerometer is perpendicular to the turntableThe table surface is downward, and the accelerometer conversion array from the rotary table TM system to the hexahedral tool C system is

Adding posture at 5 th: the input shaft of the first accelerometer is vertical to the table surface of the rotary table and upwards, and the rotary table TM is an accelerometer conversion array of the hexahedral tool C system

Adding posture at 24 th: third stepThe input shaft of the accelerometer is vertical to the table surface of the rotary table and upwards, and the accelerometer conversion array from the TM system of the rotary table to the C system of the hexahedral tool is

In the method for quickly calibrating the installation matrix based on error cancellation, in the fifth step, the addition installation matrix [ A ] of the IMU is obtained according to the pulse output of the three accelerometers under twenty-four addition postures]_CAThe method comprises the following steps:

Step (53): according to pulse outputs of three accelerometers under the 17 th adding posture, the 18 th adding posture, the 19 th adding posture, the 20 th adding posture, the 21 st adding posture, the 22 th adding posture, the 23 th adding posture and the 24 th adding posture, obtaining a third adding posture from the 17 th adding posture to the 24 th adding postureSpeedometer scale factor nonlinear error mean value epsilon_a3The projection average value theta of the installation deviation of the first accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a13The projection average value theta of the installation deviation of the second accelerometer in the direction of the third accelerometer from the 17 th adding posture to the 24 th adding posture_a23；

In the method for quickly calibrating the installation matrix based on error cancellation, in the third step, the gyro installation matrix G of the IMU]_CGComprises the following steps:

in the error cancellation-based quick calibration method for the installation matrix, in the fifth step, the added security of the IMUMounting matrix [ A ]]_CAComprises the following steps:

in the method for quickly calibrating the installation matrix based on error cancellation, in step (33), the average value epsilon of the nonlinear errors of the scale factors of the first gyroscope in the first gyroscope attitude and the second gyroscope attitude_g1Projected average value theta of second gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g21Projection average value theta of third gyro installation deviation in first gyro direction in first gyro attitude and second gyro attitude_g31Comprises the following steps:

in the method for quickly calibrating the installation matrix based on error cancellation, in step (36), the average value epsilon of the nonlinear errors of the second gyro scaling factor is obtained when the third gyro is in the attitude and the fourth gyro is in the attitude_g2Projected average value theta of installation deviation of first gyro in second gyro direction in third gyro posture and fourth gyro posture_g12Third gyro mounting deviation in the second gyro direction at the third gyro attitude and the fourth gyro attitude_g32Comprises the following steps:

in the method for quickly calibrating the installation matrix based on error cancellation, in step (39), the average value epsilon of the nonlinear errors of the scale factors of the third gyroscope in the fifth gyroscope attitude and the sixth gyroscope attitude_g2Projected average value theta of first gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g13Projected average value theta of second gyro installation deviation in third gyro direction at fifth gyro attitude and sixth gyro attitude_g23Comprises the following steps:

compared with the prior art, the invention has the following beneficial effects:

(1) according to the invention, through the combined design of a test sequence, error cancellation designs such as a gyroscope 6 position positive and negative rotation test, 8 position opposite tests of a third accelerometer group and the like are adopted, so that the rotation error of a rotary table, the inclination angle error of a table top and the machining error of a tool in the IMU installation matrix calibration test process are eliminated;

(2) the design of the test rate point and the position point are mutually independent, the calibration test sequence is free from constraint conditions, the multiplexing design of the speed and the position of the turntable can be effectively carried out through the compilation of an automatic test sequence, and the calibration test efficiency of an IMU product is improved;

(3) according to the invention, the IMU inertial navigation performance is greatly improved after the installation matrix is subjected to orthogonalization processing according to the row vector. The attitude precision can be improved by 15.8% -54.7%, the position precision can be improved by 45.2% -85.7%, and the high-precision success of model tasks is effectively guaranteed;

(4) aiming at the problem of error projection of input shaft errors concerned by users on a pitch axis and a yaw axis, the invention designs a conversion formula, and can perform rapid conversion between the installation matrix deviation of a gyroscope and an additional input shaft and the navigation azimuth deviation;

(5) the invention realizes the accurate conversion from the IMU component system to the IMU reference mirror coordinate system, establishes the accurate and orthogonalized conversion matrix from the gyroscope, the accelerometer input shaft to the reference mirror coordinate system, and facilitates the light path calibration from the user external reference to the IMU reference mirror coordinate system.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 is a schematic view of an IMU product regular hexahedron tooling installation;

FIG. 2 is a schematic view (partially) of an installation of an IMU product hexahedral fixture;

FIG. 3 is a schematic view of a mounting of a regular hexahedral tooling turntable;

FIG. 4 is a schematic diagram of a gyro mounting matrix test sequence;

FIG. 5 is a schematic view of an additive mounting matrix test sequence;

FIG. 6 is an optimal flow chart of IMU calibration test.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.

The embodiment provides an installation matrix rapid calibration method based on error cancellation, which comprises the following steps:

the method comprises the following steps: the coordinate system used in the calibration test is specified as follows:

(1) mechanical coordinate system O-x_my_mz_m(also called IMU component body coordinate system, m system for short):

1) the normal direction perpendicular to the mounting plane and pointing from the mounting plane to the optical reference mirror is + x_mDirection;

2) parallel to the mounting plane and perpendicular to the mechanical reference plane, pointing in the direction of the optical reference mirror + z_mDirection;

3)x_m、y_m、z_mforming a right-handed system.

(2) Optical reference mirror coordinate system O-x_by_bz_b：

The origin of the coordinate system is defined as the center of the upper surface of the reference mirror, x_bThe axis being normal to the top surface, z_bThe axis being in the opposite direction to the surface normal of the IMU coordinate system, y_bAxis and x_bAxis, x_bThe axis meets the right hand rule.

(3) Hexahedron tooling reference coordinate system O-x_Cy_Cz_C(abbreviation C series)

The mutual parallel and vertical relation of 6 surfaces of the hexahedral tool is better than 3 arcseconds, the hexahedral tool can be considered as a true hexahedron, and when the IMU assembly is installed on the hexahedron, under the condition that the installation errors of the gyroscope and the accelerometer are not considered, the direction of the input axis of the first gyroscope and the first accelerometer is defined as X_CAnd the input axis direction of the second gyroscope and the second accelerometer is defined as Y_CThe input axis direction of the third gyro and the third accelerometer is defined as Z_C。

(4) Rotary table switching reference coordinate system O-x_TMy_TMz_TM(TM series for short)

As shown in FIG. 3 (virtual coordinate system, the virtual frame is a hexahedral fixture), the intersection point of the rotation axis and the table top is taken as the origin, and the normal drawn from the point is taken as x_TMAn axis in the rotary table, the direction perpendicular to one edge surface of the hexahedron and pointing to the outside of the hexahedron being taken as z_TMAxis, y_TMThe axes are in orthogonal directions. Obviously, when the hexahedral tool is placed on the table top, the hexahedral tool reference coordinate system O-x_Cy_Cz_CWill coincide with the table coordinate system.

Recording the conversion array from TM system of turntable table surface system to C system of hexahedral tooling standardThen there are:

wherein x is_c、y_c、z_cRespectively, the coordinate, x, of a reference coordinate system of the hexahedral tooling_TM、y_TM、z_TMRespectively are the coordinates of the turntable transfer reference coordinate system.

Step two: conversion array for defining conversion from TM system of rotary table to C system of hexahedral tooling in gyro installation matrix test processThe specific postures are as follows:

attitude (1): the input shaft of the first top is vertical to the table top of the turntable and faces downwards,

attitude (2): the input shaft of the first top is vertical to the table top of the turntable and faces upwards,

attitude (3): the input shaft of the second top is vertical to the table top of the turntable and faces downwards,

attitude (4): the input shaft of the second top is vertical to the table top of the turntable and faces upwards,

attitude (5): the input shaft of the third top is vertical to the table top of the turntable and faces downwards,

attitude (6): the input shaft of the third top is vertical to the table top of the turntable and faces upwards,

wherein the content of the first and second substances,the gyroscope conversion array is from a rotary table TM system to a hexahedral tooling C system.

Step three: after the IMU product is installed on the regular hexahedron tool, the hexahedron tool is placed according to the posture (1) in the step two, the test is carried out after the rotary table is adjusted to be horizontal, and the pulse output F of the first gyro, the second gyro and the third gyro during the forward rotation of the rotary table is recorded_gx ¹⁺(k)、F_gy ¹⁺(k)、F_gz ¹⁺(k) Recording pulse output F of the first, second and third gyros when the turntable rotates reversely_gx ^1-(k)、F_gy ^1-(k)、F_gz ^1-(k) In that respect Wherein k is the number of each gyro pulse output group, and k is 1, 2.

Step four: adjusting the rotating speed or pulse acquisition period of the rotary table to enable F in step three_gx ¹⁺(k)、F_gy ¹⁺(k)、F_gz ¹⁺(k) And F_gx ^1-(k)、F_gy ^1-(k)、F_gz ^1-(k) The n groups of data just correspond to one rotation of the rotary table, and the pulse output data is summed (or integrated), so that:

wherein epsilon_g1 ¹First gyro scale factor non-linear error in attitude (1), θ_g21 ¹Projection of second gyro installation deviation in first gyro direction in attitude (1), theta_g31 ¹Projection of the third gyro installation deviation in the first gyro direction in attitude (1), K_gxFor the first gyro scale factor least squares fit value, K_gyFor a second gyro scale factor least squares fit value, K_gzIs the third gyro scale factor least squares fit value.

Step five: after the IMU product is installed on the regular hexahedron tool, the hexahedron tool is placed according to the posture of the step two (2), the test is carried out after the rotary table is adjusted to be horizontal, data recording is carried out according to the step three, and data calculation is carried out according to the step four, so that the IMU product can be obtained: recording three-channel gyro pulse output during forward rotation of rotary table

Wherein epsilon_g1 ²First gyro scale factor non-linear error in attitude (2), θ_g21 ²Projection of second gyro installation deviation in first gyro direction in attitude (2), theta_g31 ²And (3) the projection of the third gyro installation deviation in the first gyro direction in the attitude (2).

Step six: averaging the left parameters of the equation (2) in the fourth step and the left parameters of the equation (3) in the fifth step to obtain the scale factor error of the gyroscope in the x direction and the input axis error coefficients of the gyroscope in the y direction and the z direction:

wherein epsilon_g1Mean value of the nonlinear errors of the first gyro scaling factors in attitude (1) and attitude (2), theta_g21The projected average value of the second gyro installation deviation in the first gyro direction in the attitude (1) and the attitude (2), theta_g31The projection average value of the third gyro installation deviation in the first gyro direction in the attitude (1) and the attitude (2).

Step seven: repeating the second step to the sixth step to obtain the scale factor error of the y-direction gyroscope and the input axis error coefficients of the x-direction gyroscope and the z-direction gyroscope:

wherein epsilon_g2 ³Non-linear error of second gyro scale factor at attitude (3), θ_g12 ³Projection of the first gyro mounting deviation in the second gyro direction in attitude (3), θ_g32 ³The projection of the third gyro installation deviation in the second gyro direction in the attitude (3); epsilon_g2 ⁴Second gyro scale factor non-linear error in attitude (4), θ_g12 ⁴Projection of the first gyro mounting deviation in the second gyro direction in attitude (4), θ_g32 ⁴The projection of the third gyro installation deviation in the second gyro direction in the attitude (4); epsilon_g2Mean value of the nonlinear errors of the second gyro scaling factors in attitude (3) and attitude (4), theta_g12The projection average value of the installation deviation of the first gyro in the second gyro direction in the attitude (3) and the attitude (4), theta_g32The projection average value of the third gyro installation deviation in the second gyro direction in the attitude (3) and the attitude (4).

Step eight: repeating the second step to the sixth step to obtain the scale factor error of the gyroscope in the z direction and the input axis error coefficients of the gyroscope in the x direction and the y direction:

wherein epsilon_g3 ⁵Third gyro scale factor non-linear error in attitude (5), θ_g13 ⁵Projection of the first gyro mounting deviation in the third gyro direction in attitude (5), θ_g23 ⁵The projection of the second gyro installation deviation in the third gyro direction in the attitude (5); epsilon_g3 ⁶Third gyro scale factor non-linear error in attitude (6), θ_g13 ⁶Projection of the first gyro mounting deviation in the third gyro direction in attitude (6), θ_g23 ⁶The projection of the second gyro installation deviation in the third gyro direction when the second gyro installation deviation is in the attitude (6); epsilon_g3Mean value of the nonlinear errors of the second gyro scaling factors in attitude (5) and attitude (6), theta_g13The projection average value of the first gyro mounting deviation in the third gyro direction in the attitude (5) and the attitude (6), theta_g23The projection average value of the second gyro installation deviation in the third gyro direction in the attitude (5) and the attitude (6).

Step nine: writing a gyro installation matrix [ G ] of the IMU according to the installation parameters calculated in the sixth step, the seventh step and the eighth step]_CGOThe following were used:

step ten: orthogonalizing the formula (7) in the ninth step according to row vectors to obtain a gyro mounting matrix [ G ] of the IMU]_CGThe following were used:

wherein the constructed row vector orthogonal basis is

Step eleven: conversion array for defining conversion from TM system of rotary table to C system of hexahedral tool in process of adding meter and installing matrix testThe specific postures are as follows:

attitude (1): the input axis of the first accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (2): the input axis of the first accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (3): the input axis of the first accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (4): the input axis of the first accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (5): the input axis of the first accelerometer is perpendicular to the table top of the turntable and faces upwards,

attitude (6): the input axis of the first accelerometer is perpendicular to the table top of the turntable and faces upwards,

attitude (7): the input axis of the first accelerometer is perpendicular to the table top of the turntable and faces upwards,

attitude (8): the input axis of the first accelerometer is perpendicular to the table top of the turntable and faces upwards,

attitude (9): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (10): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (11): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (12): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (13): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable,

attitude (14): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable,

attitude (15): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable,

attitude (16): the input axis of the second accelerometer is oriented perpendicular to the table top of the turntable,

attitude (17): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (18): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (19): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (20): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable downwards,

attitude (21): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable,

attitude (22): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable,

attitude (23): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable,

attitude (24): the input axis of the third accelerometer is oriented perpendicular to the table top of the turntable,

step twelve: after the IMU product is installed on a regular hexahedron tool, the level of the rotary table is adjusted, the hexahedron tool is placed according to the posture (1) -posture (4) in the eleventh step, and pulse output F of the first accelerometer, the second accelerometer and the third accelerometer is recorded respectively_ax ¹⁺(k)、F_ay ¹⁺(k)、F_az ¹⁺(k) K is 1, 2, 3, 4; putting the hexahedral tool according to the posture (5) -posture (8) in the eleventh step, and respectively recording pulse output F of the first accelerometer, the second accelerometer and the third accelerometer_ax ^1-(k)、F_ay ^1-(k)、F_az ^1-(k) And k is 1, 2, 3, 4. The scaling factor error of the x-direction adding meter and the error coefficients of the y-direction adding meter input shaft and the z-direction adding meter can be obtained:

wherein epsilon_a1Is the average value of the nonlinear errors of the scale factors of the first accelerometer from the attitude (1) to the attitude (8), theta_a21The projection average value theta of the installation deviation of the second accelerometer in the direction of the first accelerometer from the attitude (1) to the attitude (8)_a31The projection average value of the installation deviation of the third accelerometer in the direction of the first accelerometer from the attitude (1) to the attitude (8), K_axFor the first accelerometer scale factor least squares fit value, K_ayScaling factor least squares fit for second accelerometerSum of K_azFor the third accelerometer scale factor least squares fit value, g₀To test the nominal value of the local gravitational acceleration.

Step thirteen: putting the hexahedral tool in the sequence of posture (9) -posture (12) in the eleventh step, and respectively recording pulse output F of the first accelerometer, the second accelerometer and the third accelerometer_ax ²⁺(k)、F_ay ²⁺(k)、F_az ²⁺(k) K is 1, 2, 3, 4; putting the hexahedral tool in the sequence of posture (13) -posture (16) in the eleventh step, and respectively recording pulse output F of the first accelerometer, the second accelerometer and the third accelerometer_ax ^2-(k)、F_ay ^2-(k)、F_az ^2-(k) And k is 1, 2, 3, 4. The scale factor error of the y-direction adding meter and the error coefficients of the input shafts of the x-direction adding meter and the z-direction adding meter can be obtained:

wherein epsilon_a2Is the mean value of the nonlinear error of the scaling factor of the second accelerometer from the attitude (9) to the attitude (16), theta_a12Is the projection average value of the installation deviation of the first accelerometer in the direction of the second accelerometer from the attitude (9) to the attitude (16), theta_a32The projection average value of the installation deviation of the third accelerometer in the direction of the second accelerometer from the attitude (9) to the attitude (16).

Fourteen steps: putting the hexahedral tool in the sequence of posture (17) -posture (20) in the eleventh step, and respectively recording pulse output F of the first accelerometer, the second accelerometer and the third accelerometer_ax ³⁺(k)、F_ay ³⁺(k)、F_az ³⁺(k) K is 1, 2, 3, 4; putting the hexahedral tool in the sequence of posture (21) -posture (24) in the eleventh step, and respectively recording pulse output F of the first accelerometer, the second accelerometer and the third accelerometer_ax ^3-(k)、F_ay ^3-(k)、F_az ^3-(k) And k is 1, 2, 3, 4. The scaling factor of the y-direction addition can be obtainedNumber error and x-direction, z-direction add meter input axis error coefficients:

wherein epsilon_a3Is the average value of the nonlinear error of the scaling factor of the third accelerometer from the attitude (17) to the attitude (24) ([ theta ])_a13Is the projection average value of the installation deviation of the first accelerometer in the direction of the third accelerometer from the attitude (17) to the attitude (24), theta_a23The projection average value of the installation deviation of the second accelerometer in the direction of the third accelerometer from the attitude (17) to the attitude (24).

Step fifteen: writing an addition installation matrix [ A ] of the IMU according to the installation parameters calculated in the step twelve, the step thirteen and the step fourteen]_CAOThe following were used:

sixthly, the steps are as follows: orthogonalizing the formula (12) in the step fifteen according to row vectors, and adding and installing a matrix [ A ] of the IMU]_CAThe following were used:

wherein the constructed row vector orthogonal basis is

Seventeen steps: according to the formula (8) in the step ten and the formula (13) in the step sixteen, the gyroscope input shaft coordinate system, the adding input shaft coordinate system and the mechanical coordinate system are converted to obtain the following installation matrix:

wherein, g_ij、a_ijAre respectively a matrix [ G]_mG、[A]_mAThe elements of (a) and (b),mounting matrix C for gyroscope_mCGThe inverse of the matrix of (a) is,for adding the meter, installing matrix C_mCAInverse matrix ofWhen the gyroscopic input shaft and the totalizing input shaft are in the same design (i.e. the orientation of the gyroscopic input shaft and the totalizing input shaft is consistent)) The transformation matrix C can be obtained according to the following formula_mC：

Wherein:the normal line pitch angle of the product mounting surface is theta_XThe normal yaw angle of the product installation surface is phi_XThe normal pitch angle of the product reference leaning surface is theta_ZThe normal yaw angle of the product reference leaning surface is phi_Z。

The pitch and yaw angles of the gyro and the accelerometer can be determined according to the following table.

TABLE 1 calibration results for each gyro and accelerometer input axis

Adding meter	Pitching angle (deg)	Yaw angle (deg)
			G1	arcsin(g₁₁)	arctan(-g₁₂/g₁₃)
G2	arcsin(g₂₁)	arctan(-g₂₂/g₂₃)+180
			G3	arcsin(g₃₁)	arctan(-g₃₂/g₃₃)-180
A1	arcsin(a₁₁)	arctan(-a₁₂/a₁₃)
			A2	arcsin(a₂₁)	arctan(-a₂₂/a₂₃)+180
A3	arcsin(a₃₁)	arctan(-a₃₂/a₃₃)-180

Eighteen steps: recording the transformation matrix from the mechanical system (IMU body system) m to the reference mirror coordinate system b asThe calculation method is as follows. In the reference mirror coordinate system b there areTwo mutually non-parallel light path vectors C1 and C2, whose corresponding measured unit observation vectors in the body coordinate system m are t1 and t2, respectively. Assume that the first vector basis is:

the second vector basis is a unit vector that is perpendicular to both observation vectors and can be expressed as follows:

the third vector basis is chosen as:

thus, a transformation matrix from the mechanical system (IMU body system) m to the reference mirror coordinate system bComprises the following steps:

nineteen steps: according to the formula (14) in the step 17 and the formula (19) in the step eighteen, a conversion matrix from the gyro input shaft coordinate system G, the totalizing input shaft coordinate system A to the reference mirror coordinate system b can be obtainedAndrespectively as follows:

and (3) installing an IMU component: as shown in fig. 1, the hexahedral fixture is taken out of the packaging box, and the mounting surfaces and the leaning surfaces of the hexahedral fixture and the IMU assembly are cleaned by alcohol cotton balls. And placing the hexahedral tool on the operating platform, enabling the IMU mounting plane to be horizontal, carefully placing the IMU assembly into the hexahedral datum, and enabling the IMU datum leaning surface to be in contact with the IMU positioning surface of the hexahedral tool. Three metal washers were installed in the IMU snubber holes as shown in fig. 2 and the IMU was secured to the tooling (lightly held, not tightened) with 3M 5 screws. And (5) enabling the IMU reference leaning surface to lean against the tool positioning surface. The mounting screws are screwed (screws opposite to the leaning surface are screwed firstly to avoid the leaning surface from sliding), and are screwed for multiple times, so that the mounting errors of the reference leaning surface and the mounting surface are ensured to be as small as possible. And (4) checking whether the mounting surface and the leaning surface are tightly combined or not, and if necessary, using a feeler gauge or a flashlight for polishing and checking.

Installing a hexahedron tool: the large end face of the hexahedral tooling is placed on the table top of the inner ring of the double-shaft rotary table, and the output shaft of the tested gyroscope is close to the main shaft of the inner ring of the rotary table as much as possible. The hexahedral tooling was fixed with 4L-shaped compacts and 4M 8 screws, respectively. The locking pin of the pitching shaft (outer frame) is opened, the pitching shaft (outer frame) of the turntable is balanced, and the pitching shaft (outer frame) of the turntable can stay at any position.

Calibrating a gyro input shaft: after various preparation and inspection operations are finished, the cable is connected, and the product is powered on. And leveling the table top of the rotary table by using a level meter, wherein the levelness error of the table top of the rotary table is required to be less than 5 arc seconds. The calibration test was performed according to the test sequence shown in fig. 4:

PSTG 01: the input shaft of the first top is vertical to the table top of the rotary table and faces downwards;

PSTG 02: the input shaft of the first top is vertical to the table top of the rotary table and faces upwards;

PSTG 03: the input shaft of the second top is vertical to the table top of the rotary table and faces downwards;

PSTG 04: the input shaft of the second top is vertical to the table top of the rotary table and faces upwards;

PSTG 05: the input shaft of the third top is vertical to the table top of the rotary table and faces downwards;

PSTG 06: the input shaft of the third top is vertical to the table top of the rotary table and faces upwards;

the main shaft of the rotary table is driven to rotate at the speed of +7.5 degrees/s, and 5 groups of 9.6s gyro pulse output F are recorded after the rotary table operates stably_g ⁱ⁺The main shaft of the rotary table is driven to rotate at a speed of-7.5 degrees/s, and 5 groups of 9.6s gyro pulse output F are recorded after the rotary table operates stably_g ^i-The calculation is performed according to the formulas (1) to (8).

Adding a meter to calibrate an input shaft: after various preparation and inspection operations are finished, the cable is connected, and the product is powered on. And leveling the table top of the rotary table by using a level meter, wherein the levelness error of the table top of the rotary table is required to be less than 5 arc seconds. The calibration test was performed according to the test sequence shown in fig. 5:

PSTA01-PSTA 04: the input shaft of the first accelerometer is vertical to the table top of the rotary table and faces downwards;

PSTA05-PSTA 08: the input shaft of the first accelerometer is vertical to the table top of the turntable and faces upwards;

PSTA09-PSTA 12: the input shaft of the second accelerometer is vertical to the table top of the rotary table and faces downwards;

PSTA13-PSTA 16: the input shaft of the second accelerometer is vertical to the table top of the turntable upwards;

PSTA17-PSTA 20: the input shaft of the third accelerometer is vertical to the table top of the rotary table and faces downwards;

PSTA21-PSTA 24: the input shaft of the third accelerometer is vertical to the table top of the turntable and faces upwards;

and recording the three adding pulse outputs at twenty-four positions respectively, calculating the adding pulse output in unit time, and calculating according to the formula (9) to the formula (13).

Calculating the pitch angle and yaw angle of the gyroscope and the adder: setting the normal line pitch angle of the product mounting surface as theta_XThe normal yaw angle of the product installation surface is phi_XThe normal pitch angle of the product reference leaning surface is theta_ZThe normal yaw angle of the product reference leaning surface is phi_ZWhen the orientation design of the gyro input shaft and the direction design of the adding input shaft are consistent, the pitch angle and the yaw angle of the gyro and the adding are calculated according to the corresponding relations of a formula (14), a formula (15) and a table 1.

Conversion of the reference mirror coordinate system: and (3) calculating an installation matrix of the 3-axis gyroscope and the 3-axis accelerometer input axis coordinate system relative to the reference mirror coordinate system according to the formulas (16) to (20), so that an external test and an Inertial Measurement Unit (IMU) inertial navigation simulation test based on an inertial system can be conveniently carried out.

Optimizing an IMU calibration test flow: in order to improve the working efficiency of IMU calibration test, the working conditions of the double-shaft rotary table in the same pose are integrated, the optimized calibration test flow is shown in figure 6, the efficiency is improved by more than 50% compared with the subentry test efficiency, and the product card loading times are reduced by 3 times.

In the embodiment, a double-shaft turntable is adopted to carry out the integrated test of the gyro installation matrix and the totalizing installation matrix in the IMU product. Through the combined design of the test sequence, the rotary error of the rotary table, the inclination error of the table top, the machining error of the tool and the like in the IMU installation matrix calibration test process are eliminated, so that the uncertainty of the measurement is better than 10 arc seconds. Through the installation matrix orthogonalization processing, under the condition of not changing an installation error angle, different orthogonal bases are constructed, and the gyro installation matrix and the adding installation matrix are respectively enabled to be orthogonal in vector. In addition, a combined double-shaft light tube double-vector attitude determination technical scheme is adopted, so that the accurate conversion from an IMU reference mirror coordinate system to an IMU component body system is realized, an accurate and orthogonalized installation matrix from a gyroscope, an accelerometer input shaft to the reference mirror coordinate system is established, and the attitude accuracy, the speed accuracy and the position accuracy of IMU navigation are effectively improved. The method can also effectively improve the calibration test efficiency of the IMU product through the compilation of an automatic test sequence.

Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.

33页详细技术资料下载

上一篇：一种医用注射器针头装配设备

下一篇：一种基于高精度三轴转台的陀螺标度因数测试方法

Method for quickly calibrating installation matrix based on error cancellation

相关技术

网友询问留言