JPH08210855A - Azimuth measurer - Google Patents

Abstract

PURPOSE: To remove a random bias component by mounting and rotating an angular speed sensor for correction on a rotary base, stopping it in the case parallel to an X-axis and a Y-axis, and in the case parallel to the reverse direction, taking in a bias value, and thereby correcting angular speed sensors of the X-axis and the Y-axis. CONSTITUTION: A rotatry base 9 is rotatably supported in parallel to a plane including an X-axis and a Y-axis, and rotatably driven by a drive part 11. An angular speed sensor for correction 12 is provided thereon. The sensors 12 are stopped at total 4 points where the direction of the sensor 12 is the same direction and the reverse direction of the angular speed sensors 4, 5 corresponding to the X-axis and the Y-axis. The detection output of the sensors 4 or 5 and the sensor 12 are read as a bias value at each stop point, an angular speed detection signal on the X-axis and the Y-axis is corrected in a random bias correction part, and input to an azimuth calculation part. As a result, a highly accurate measuring result can be obtained open if a low accurate sensor is used.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an azimuth measuring device for measuring an azimuth angle (deviation angle from the north in the bow direction) of a ship or the like, and particularly using an inexpensive gyro (angular velocity sensor) while reducing the cost. Is to provide a strap-down type azimuth measuring device capable of improving measurement accuracy.

[0002]

2. Description of the Related Art FIG. 5 shows the structure of a conventional strap-down type azimuth measuring device. In the figure, 10 indicates the entire compass. The strap-down azimuth meter 10 is composed of three acceleration sensors 1, 2, 3 and three angular velocity sensors 4, 5, 6. The three acceleration sensors 1, 2 and 3 are mounted in a posture in which their input axes are parallel to the X, Y and Z axes which are orthogonal to each other. The arrow attached to each input shaft indicates the detection signal a output when an acceleration is applied in that direction.
This means that x, ay, and az have positive polarities.

The angular velocity sensors 4, 5 and 6 have X,
It is attached in a posture parallel to the Y and Z axes. The arrow attached to each input shaft is an angular velocity detection signal GX output when a rotation input is given in the direction of each arrow around each input shaft,
This means that GY and GZ are processed as positive polarity angular velocity detection signals. FIG. 6 shows an electrical connection structure of the strapdown type azimuth meter shown in FIG. The detection signals ax, ay, az of the acceleration sensors 1, 2, 3 are directly input to the azimuth angle calculation unit 8 from the respective sensors 1, 2, 3. On the other hand, the detection signals of the angular velocity sensors 4, 5, 6 are input to the azimuth angle calculation unit 8 via the fixed bias correction unit 7.

The azimuth is determined as follows.
First, the gravitational direction is obtained from the acceleration detection data of the acceleration sensors 1, 2 and 3, and the horizontal reference plane is assumed from this gravitational direction. The rotational speed applied to the azimuth meter 10 is obtained from the three angular velocity data, and the horizontal reference plane is held. Further, the azimuth angle calculation unit 8 calculates the earth rotation angular velocity component on the horizontal reference plane, and the direction in which the earth rotation angular velocity component is the minimum is the east-west direction, and the direction in which the earth rotation angular velocity component is the maximum is the north-south direction. , Determine the azimuth.

The accuracy of the measured azimuth depends, of course, on the performance of the angular velocity sensors 4, 5, 6. The performance of the angular velocity sensors 4, 5, 6 is mainly determined by an error called bias (also called drift). The bias means the value of the output signal when the angular velocity is not input to the angular velocity sensors 4, 5, 6 (stationary state). Incidentally, the permissible value of the bias error for obtaining the azimuth angle accuracy of 0.1 ° in the vicinity of Japan must be an error of about 0.02 ° / (time) or less.

That is, assuming that the latitude around Japan is 35 degrees, the rotation angular velocity ω ₃₅ about the north-south axis of the horizontal plane can be obtained by multiplying the rotation angular velocity ω ₀ on the equator by the cos latitude.
If ω ₀ = 15 ° / hour, then ω ₃₅ = 15 ° / H × cos ₃₅ ° = 12.3 ° / H the rotational angular velocity component ω at a deviation angle from the east of 0.1 °
_0.1 is ω _0.1 = 15 ° / H × sin _0.1 ° = 0.02 ° / H Therefore, if the angular velocity sensors 4, 5 and 6 have a bias error of 0.02 ° / H, the azimuth is 0.1 °. Will have an error of.

The bias value of the angular velocity sensor generally fluctuates with the passage of time, and if the number of sampling points is increased, the bias value obtained at each sampling point will be normally distributed. The average value of the bias values (hereinafter referred to as a fixed bias) can be calculated and corrected from the data acquired during the inspection of the angular velocity sensor. The fixed bias correction unit 7 corrects this correction.

[0008]

By the way, the bias value of the angular velocity sensor fluctuates with the passage of time, as described above. Hereinafter, this fluctuation component will be referred to as a random bias. The fluctuation range of random bias cannot be predicted. Therefore, in order to improve the azimuth accuracy, it is necessary to use an angular velocity sensor with a small random bias component.

A ring laser gyro is mentioned as an angular velocity sensor having a small random bias component.
By using the ring laser gyro, an azimuth meter with high azimuth accuracy can be constructed. However, the ring laser gyro is expensive, and since the strap-down type azimuth meter must use three expensive angular velocity sensors, the entire azimuth meter is extremely expensive. .

An object of the present invention is to provide a strap-down type azimuth meter which can measure an azimuth accurately while using an inexpensive angular velocity sensor.

[0011]

According to the present invention, in a strapdown type azimuth meter composed of three acceleration sensors and three angular velocity sensors, three angular velocity sensors are used, for example, optical fiber coils. Replace with the configured angular velocity sensor. The angular velocity sensor formed by using the optical fiber coil is inexpensive, but has a characteristic that the random bias component is large.

Therefore, in the present invention, a correction angular velocity sensor is provided to remove the random bias component. This correction angular velocity sensor is mounted on the turntable and provided with X
The rotary table is rotated so that the angular velocity sensor having the input shaft arranged in the direction parallel to the axis and the Y-axis can be changed to the posture in the same direction and the opposite direction. The rotary base is stopped every time the orientation of the correction angular velocity sensor reaches the posture in the same direction or the opposite direction with respect to the angular velocity sensor arranged in the direction parallel to the X-axis and the Y-axis, and the correction angular velocity is stopped at the stop position. Take the bias value from the sensor. The bias value thus taken in is subtracted from the measured values of the angular velocity sensors provided in parallel with the X-axis and the Y-axis, and the random bias component is removed from the detection signals of these angular velocity sensors.

Therefore, according to the configuration of the present invention, it is possible to obtain an advantage that the azimuth can be accurately measured while using the inexpensive angular velocity sensor.

[0014]

1 shows the structure of an azimuth measuring apparatus 20 according to the present invention. The parts corresponding to those in FIG. 5 are designated by the same reference numerals. In the present invention, a rotary base 9 rotatably supported with a plane parallel to the plane including the X axis and the Y axis as a rotary surface, a rotary drive unit 11 for rotationally driving the rotary base 9 and a correction base mounted on the rotary base 9. An angular velocity sensor 12 is newly provided. Including the correction angular velocity sensor 12, other angular velocity sensors 4,
5 and 6 use low-priced angular velocity sensors configured using, for example, optical fiber coils.

The rotary drive unit 11 can be constituted by, for example, a pulse motor, and, for example, by a control command output from the random bias correction unit 13 shown in FIG.
When the orientation of the correction angular velocity sensor 12 is in the same direction as the orientation of the angular velocity sensor 4 (hereinafter, this rotational position is referred to as 0 ° position), the orientation of the correction angular velocity sensor 12 is the angular velocity sensor 5
The same direction as the direction of
The position of the correction angular velocity sensor 12 is opposite to the direction of the angular velocity sensor 4 (hereinafter, this rotational position is referred to as 1).
It is controlled to stop in a state where the orientation of the correction angular velocity sensor 12 is opposite to the orientation of the angular velocity sensor 5 (hereinafter, this rotational position is referred to as the 270 ° position).

As shown in FIG. 2, the random bias correction unit 13 is provided after the fixed bias correction unit 7, and the random bias correction unit 13 corrects the angular velocity detection signals gx and gy on the X axis and the Y axis. Corrected angular velocity detection signal ω
x and ωy are input to the azimuth angle calculation unit 8. That is, the detection outputs ax, ay, az of the acceleration sensors 1, 2, 3 are directly input to the azimuth angle calculation unit 8 as in the conventional case. The detection outputs of the angular velocity sensors 4, 5, 6 and the correction angular velocity sensor 12 are input to the fixed bias correction unit 7, and the fixed bias prepared and stored in advance is subtracted and removed. That is, the fixed bias correction unit 7 includes fixed bias values dx and d obtained by measuring the angular velocity sensors 4, 5, 6 and 12 at the time of inspection.
Y, dz, and dw are stored and prepared. The detection outputs of the angular velocity sensors 4, 5, 6 and 12 are GX, GY, GZ and G.
Let W be the value after fixed bias correction gx, gy, g
z and gw are as follows: gx = GX-dX gy = GY-dY gz = GZ-dZ gw = GW-dW.

The random bias correction unit 13 has a random bias correction value d for the angular velocity sensors 4, 5, and 12.
x, dy, dw are stored and prepared. Initially, zero is stored as the random bias correction values dx, dy, and dw.
The angular velocity signals after random bias correction are ωx, ωy, ω
If w, then ωx = gx−dx ωy = gy−dy ωw = gw−dw.

Random bias correction values dx, dy, dw
The calculation method of is performed as follows. The rotary table 9 is stopped at the 0 ° position, and in this state, the detection outputs of the angular velocity sensor 4 and the correction angular velocity sensor 12 are set to ω.
It is read and stored as x ₀ , ω w ₀ . The rotary table 9 is stopped at the 90 ° position, and in this state, the detection outputs of the angular velocity sensor 5 and the correction angular velocity sensor 12 are read and stored as ωy ₉₀ and ωw ₉₀ .

The rotary table 9 is stopped at the 180 ° position,
In this state, the detection outputs from the angular velocity sensor 4 and the correction angular velocity sensor 12 are read and stored as ωx ₁₈₀ and ωw ₁₈₀ . The rotary table 9 is stopped at the 270 ° position, and in this state, the detection outputs of the angular velocity sensor 5 and the correction angular velocity sensor 12 are read and stored as ωy ₂₇₀ and ωw ₂₇₀ .

The following calculation is performed. A = ωx ₀ −ωw ₀ B = ωx ₁₈₀ + ωw ₁₈₀ C = ωy ₉₀ −ωw ₉₀ D = ωy ₂₇₀ + ωw ₂₇₀ The residual δx after the random bias correction of the angular velocity sensor 4 in this state and the random angular velocity sensor 12 for correction The residual δwx after bias correction is obtained by δx = (A + B) / 2 δwx = (B−A) / 2.

Angular velocity sensor 5 and correction angular velocity sensor 12
In the same manner as above, δy = (C + D) / 2 δw _y = (D−C) / 2. Here, δx and δwx are (A + B) / 2,
The reason obtained by the calculation of (BA) / 2, (C + D) / 2, and (DC) / 2 will be described with reference to FIG.

Only the pair of angular velocity sensors 4 and 12 will be described here. Assuming that the physical rotational angular velocity about the input axis X at the time of measuring ωx ₀ and ωw ₀ with the rotary table 9 stopped at the 0 ° position is ω ₀ , then ωx ₀ = ω ₀ + δx ωw ₀ = ω ₀ + δwx Let ω ₁₈₀ be the physical rotational angular velocity about the input axis X when the rotary table 9 stops at the 180 ° position, then ω ₁₈₀
Is output with the opposite polarity to ωX ₁₈₀ as shown in FIG. 3, but the residual δwx is output with the same polarity as when 0 °.

[0023] _{ωx 180 = ω 180 + δx ωw} 180 = -ω 180 + δwx A = ωx 0 -ωw 0 = ω 0 + δx-ω 0 -δwx B = ωx 180 + ωw 180 = ω 180 + δx-ω 180 + δwx the A and B by displaying by inverting the Omegadaburyu ₁₈₀ in FIG. 3, it can be understood at a glance to be the a = δx-δwx B = δx + δwx.

Therefore, (A + B) / 2 = δx (BA) / 2 = δwx is obtained. The residuals δx and δwx and δy and δwy are fed back to the random bias correction values dx, dwx and dy, dwy.

Dx _n = dx _n-1 + δx dwx _n = dwx _n-1 + δwx dy _n = dy _n-1 + δy dwy _n = dwy _n-1 + δwy Since δwx and δwy are random bias values of the sensor 12, , The values are almost equal to each other, and either of them may be used as the correction of the sensor 12.

Note that n represents the relationship before and after the feedback. The above measurement, calculation, and feedback are performed regularly, for example, every several tens of seconds. FIG. 4 shows how the random bias correction values dx and dy are updated. Residuals δx and δwx and δ
Each of y and δwy is calculated as a pair by the bias correction residual calculation unit 13A. The residuals δx and δwx and δy and δwy are added to the random bias correction values dx _{n -1} and dy _n-1 previously obtained and stored by the correction value update units 13B and 13C to obtain the random bias correction value dx _n. And dy _n are updated.

Updated random bias correction value dx _n
G an angular velocity signal fixed bias correction using the dy _n
Random bias correction is performed on x and gy, and angular bias signals ωx and ωy subjected to random bias correction are obtained. In addition, although the example in which the correction angular velocity sensor is provided only for the X axis and the Y axis has been described above, the correction angular velocity sensor may be provided for the Z axis as well. In this case, the correction angular velocity sensor may be mounted on a turntable that can switch between the upward and downward postures.

[0028]

As described above, according to the present invention, the correction angular velocity sensor 12 is provided, and the posture of the correction angular velocity sensor 12 is the angular velocity sensor 4 having the X and Y axes as the input axes GX and GY. 5 is changed in the same direction and in the opposite direction, the random bias is measured every time the posture is changed, and the operation of updating the random bias correction value is periodically repeated. Therefore, the random bias value gradually changes. Even if a so-called drift occurs, the drift can be removed and a correct measurement result can always be obtained.

Therefore, even if an inexpensive angular velocity sensor having a relatively low accuracy is used, there is a great advantage that a highly accurate azimuth measurement result can be obtained.

[Brief description of drawings]

FIG. 1 is a layout view for explaining an embodiment of the present invention.

FIG. 2 is a block diagram showing an electrical circuit configuration of the present invention.

FIG. 3 is a graph for explaining the operation of the main part of the present invention.

FIG. 4 is a block diagram for explaining the operation of the main part of the present invention.

FIG. 5 is a layout diagram for explaining a conventional technique.

FIG. 6 is a block diagram for explaining the operation of the conventional technique.

[Explanation of symbols]

 1,2,3 Acceleration sensor 4,5,6 Angular velocity sensor 7 Fixed bias correction unit 8 Azimuth angle calculation unit 9 Rotation table 11 Rotation drive unit 12 Correction angular velocity sensor 13 Random bias correction unit

Claims (1)

[Claims] 1. A. First Embodiment Three acceleration sensors that detect accelerations applied in mutually orthogonal X-axis, Y-axis, and Z-axis directions; and B. Similarly, three angular velocity sensors that detect rotation about each of the X axis, Y axis, and Z axis, and C.I. A turntable supported with a plane including the X axis and the Y axis as a plane of rotation; It is mounted on this turntable, and is stopped each time the rotation input shaft reaches the postures parallel to the X-axis and the Y-axis and the opposite postures parallel to the rotation of the turntable. A correction angular velocity sensor for detecting the angular velocity, and E. The detection signal of the correction angular velocity sensor is subtracted from the detection signals of the X-axis and the Y-axis in the posture parallel to the detection axis of the correction angular velocity sensor and in the opposite parallel posture, and the X-axis and the Y-axis are set as the center. A random bias correction unit that removes a random bias error included in the detected angular velocity value; An azimuth measuring device comprising: an azimuth angle calculation unit that calculates an azimuth based on the angular velocity signal from which the random bias error is removed by the random bias correction unit and the acceleration obtained from the acceleration sensor. .