JPS58225312A - Output correction of bearing detector - Google Patents

Abstract

PURPOSE:To measure accurate bearing by evaluating the authenticity of a vector locus for calibration determined by turning a bearing detection sensor for detecting magnetic fields for one time. CONSTITUTION:Output of a bearing sensor 10 is inputted into a computer system 50 following a A/D conversion 60 through a signal processing section 20 and the bearing is calculated and displayed 30. The computer system 50 compared the DC component in the vector locus of a magnetic field vector detected from an earth magnetism bearing sensor 10 when it is turned once with the value of the DC component determined from the vector locus previously measured. When the difference therebetween is smaller than a specified value, the DC component most currently measured is used as such for the correction of the bearing. But when it is larger than the value, a specified correction value is added to the DC component used for the previous correction of the bearing and the results are used for the correction of the bearing as DC component.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for correcting the output of a direction detection device used to detect the direction of movement of a vehicle, etc., and more specifically, the present invention relates to a method for correcting the output of a direction detection device used to detect the direction of movement of a vehicle, etc. The present invention relates to output correction of an azimuth detection device.

C) The mounting position of the automobile orientation detection sensor, which applies the principle of measuring the magnetic field vector generated by the earth's magnetism, is as follows:
It should be possible to reduce the amount of metal and the earth's magnetic field inside the vehicle, and to make it easy to fix the orientation detection sensor. Therefore, for example, a body steel plate in the center of the roof is selected. This body steel plate is made of ferromagnetic material such as iron, so residual magnetism is likely to occur.For this reason, even if the vehicle is completely demagnetized at the initial stage, as the vehicle is used for a long time, residual magnetism may be generated from outside the vehicle. The body steel plate is magnetized by the magnetic field caused by the surface current of the vehicle body, or the magnetic field caused by 11Ei, etc., and residual magnetism is formed as a result.For this reason, in the direction detection device, not only the magnetic field vector due to the earth's magnetism but also the body steel plate The magnetic field generated by magnetization is also detected, and accurate orientation cannot be detected due to this I.

Therefore, in order to detect accurate orientation in degrees Celsius, magnetic field components other than the earth's magnetic field must be corrected. The following method has been considered as a means for this purpose.

One method is to demagnetize the vehicle body plate by AC demagnetization or the like during the use of the vehicle.

However, this method has the disadvantage that demagnetization requires time and effort, and demagnetization equipment is required. On the other hand, as another method, a compensation winding is installed near the 6th position detection device or in the direction detection device itself, and an appropriate DC current is passed to create a compensation magnetic field, thereby canceling out the residual magnetic field caused by the magnetization of the body. It's a method. This method requires the addition of a compensation winding, and also requires precise adjustment of the compensation current to adjust the magnitude of the compensation magnetic field every time the magnitude of magnetization m changes. The drawback is that the adjustment is complicated.

By the way, as a sensor for measuring the magnetic field vector due to the earth's magnetic field, there are, for example, a flux gate valve, a flux meter using a Hall element, etc. Two of these detection elements are used, for example, in two directions (X-axis) perpendicular to each other on one plane.

Y-axis direction), and each magnetic field component is determined. After the signals detected from these two detection elements are processed by an appropriate circuit and extracted as a voltage signal, the locus of the signal generally has an elliptical shape. Then, one output component is VX, and the other output component is VX.
When V, x and Vy can be written as the following formulas.

VX = Kx - [3 cos e vy 4y ” Bs1nθ However, ■× is the output of the first magnetic detection element, Vy t,
is the output of the second magnetic detection element, KX and KV are coefficients ψ that depend on the output system of the first and second magnetic detection elements. B is the horizontal component of the earth's magnetic field. θ is the angle between the horizontal component B and the central axis of the first magnetic sensing element. Here Kx
, Ky are adjusted to be equal to each other, the magnetic field vector locus represented by the measured voltage output becomes a circle. FIG. 1 shows this vector locus. In this way, when the magnetic field vector is caused by the earth's magnetic field Gt, the detected vector locus becomes a circle 1 centered on the origin. However, in reality, as mentioned above, the origin of this vector trajectory is not necessarily the origin due to the influence of the magnetic field on the vehicle body. The value of this @magnetic vector is measured in a coordinate system fixed to the vehicle body. Therefore, even if the vehicle rotates with respect to its orientation, the magnetization vector rotates together with the vehicle body, so the output of the magnetic orientation sensor fixed to the vehicle body does not vary with the rotation of the vehicle body.

That is, the influence that the magnetized flat wheel has on the orientation detection range with respect to the rotation of the vehicle is the influence as a DC component.

For this reason, as shown in Figure 1, a circular vector locus is drawn centered at a point Δvector away from the origin.Therefore, when determining the direction from this vector locus, there is an influence of magnetization. In this case, simply the arctangent of the ratio of the components in the X direction or Yh direction of the measured magnetic field vector is:
I can't ask for it. That is, the direction must be determined using the following formula.

G=+an −' ((Vx −Box> / (V
V -13°y ) Here, BoX and BoV are the X and Y direction components of the magnetization vector. The output signal detected from the direction detection sensor installed in the vehicle in this manner must be corrected by the amount of the magnetization vector. This @magnetic field torque amount is given by the vector m at the center of the circle of the vector locus drawn by the output signal when the vehicle body is rotated once in the range of 0 to 6360 degrees. Therefore, it is necessary to find the center of the vector locus using an appropriate method and then correct it as shown in the above equation to find the h position θ.

However, even in θ corrected by such a method, errors occur due to the influence of external disturbances.

That is, if the measurement data used to find the center vector of the circle of the vector locus includes a short-time disturbance, it will clearly contain an error. For example, suppose that the vector locus detected by the direction detection sensor when the vehicle rotates once becomes as shown in FIG. The center vector of this vector locus is usually calculated based on the average (10) of the maximum and minimum values of the X component of the measured values, or the average of the maximum and minimum values of the Y component. -!1(ll+ <If a short-term disturbance occurs at measurement points P1 and P2, the center solid 1 to 1 calculated from the average of points 11 and 13 and the average of points 12 and P4
If a vector locus like that shown in Fig. 2 is obtained, the magnetization vector, generally the DC component vector with respect to the direction, is , AO]. Therefore, in such a case, the center of the vector locus should be set to A
If it is one vector, there will be no change in the magnetization vector in reality, which will cause a large error in the measurement of the orientation.

Therefore, when the center vector of a circle is extracted in this way, the present invention does not immediately use that value to correct the measurement data as a direct current component, but instead calculates it based on the data of the vector locus measured last time. The novelty of the newly obtained value of the center vector A1 is evaluated in relation to the center solid vector A1. In the present invention, the magnetization vector is
Based on the premise that the DC component (magnetization vector) does not fluctuate in a short cycle equal to or longer than the period of one rotation of the vehicle, the DC component (magnetization vector) is calculated and compared with past DC component measurement data, and it is determined that two consecutive measurements are approximately the same. If the value is obtained, use the value determined from the most recent measurement value as the DC component of the direction correction, otherwise,
By correcting the DC component of the previous direction correction by a predetermined amount i) By correcting the corrected value as the DC component of the direction correction, it is possible to prevent the influence of disturbances that occur in a short period of time, thereby accurately measuring the direction of movement of the vehicle body. Define what you want to do as your purpose.

That is, the present invention obtains a DC component unrelated to the alignment direction of the geomagnetic direction detection sensor υ in the Peltle locus of the magnetic field vector detected from the geomagnetic direction detection sensor υ, and extracts the DC component from the detected magnetic field Peltle. In the method of determining the direction using the corrected value, the DC component obtained from the most recently measured vector trajectory is compared with the DC component obtained from the previously measured vector trajectory, and the difference is determined by a predetermined value. When it is smaller than the value, the DC component of the latest measurement is used as the DC component used for direction correction, and when the difference is more than a predetermined value, the predetermined correction m is applied to the DC component used for the previous direction correction. This consists of a method for correcting the output of the direction detecting device, in which the value obtained by adding (J) is corrected as a direct current component used for direction correction.

Here, as a method for determining the center vector of the circle of the vector locus, a mathematical method for determining the center of the circle can be used, but the method is not limited to this method. For example, as mentioned above, the average value of the maximum and minimum values of the measured value X component and the measured value Y
The average value of the maximum value and minimum value in the direction can be determined as the center point with the X and Y components, respectively. In addition, as shown in the example, the average value of two points where a straight line parallel to The components of the center vector can be found. It is also possible to find two sets of perpendicular bisectors of any two points on the circumference and find the intersection of the straight lines. In this way, one vector locus can be obtained by rotating seven cars once. The center vector is determined based on the obtained value.

Then, memorize the amount of rotation, and then select the center solid 1 of the vector trajectory that will be measured when the vehicle rotates one more time.
When . In the present invention, if the difference between the previously calculated uncorrected center vector and the most recently calculated uncorrected center vector is smaller than a predetermined value, then the newly calculated center vector is calculated. The value of the vector is assumed to be true. That is, if one rotation of the vehicle is one recycle, and if consecutive or similar center vectors are found in two consecutive cycles, the value is considered to be correct and is set to 4q. This is because disturbances other than magnetization have a short period compared to the rotation period of the vehicle, so the probability that a disturbance will occur at the same measurement phase point twice in a row is low, and magnetization fluctuations This is because the period can generally be considered to be sufficiently longer than the rotation period of the vehicle. On the other hand, if the value of the difference is larger than a predetermined value, it is determined that one of the uncorrected central vectors is affected by a disturbance! yi'! I will. Therefore, in that case, a new DC component (magnetization vectors 1 to 1) [11 Add a predetermined small correction component to the DC component that has been used for correction up to now]
Make it something that has added to it. Here, the predetermined correction component may be, for example, a vector I to Le whose direction is equal to the vector of the difference and whose absolute value 111 is a predetermined value. Further, as shown in the embodiment, the difference between the deviation vector and a predetermined value can be determined independently for the X component and the Y component. In this case, if the X component is smaller than a predetermined value, that value is adopted; if the X component of the difference is larger than the predetermined value, the Correct by adding a small amount of correction. The same applies to the Y component. Correcting the X and Y components independently is effective when one local independent disturbance occurs near the measurement phase point (a short-period disturbance as shown in Figure 2). ). In this manner, the present invention is intended to prevent the influence of disturbances to a minimum and to accurately determine the orientation.

Hereinafter, the present invention will be explained in more detail based on Examples.

FIG. 3 is a block diagram 12-gram showing a single-function device for implementing the method of the present invention. 10 is Senna in 6th place. The orientation sensor 10 has an annular core 14, and four excitation windings 11 are arranged around the core 14. Both ends of the excitation winding 11 are connected to an oscillator 21. . On the other hand, an output winding 12 that is wound around the opposing body portions of the annular core 14 and an output winding 13 that is similarly wound around the annular core 14 in a direction perpendicular to this are arranged. ing.

The output horn winding 12 inputs both ends thereof to an amplifier 238,
Both ends of the output winding 13 are input to another amplifier 23b. The outputs of these amplifiers are input to sample and hold circuits 24a and 24b, respectively, and the gate inputs of these sample bold circuits are connected to input the signal from the timing generation circuit 22. The outputs of the sample and hold circuits are each input to an AD converter 60, converted into a digital quantity, and input to the computer system 50. The output of the combination system 50 is output to a display drive circuit 40, which drives the display device 30 to display the direction of travel of the vehicle. In such a configuration, the operation will be explained as follows.

Since the annular core 14 is made of magnetically hard material, it can be easily saturated.

The alternating current of frequency F applied to the excitation winding 11 saturates the magnetic flux within this annular core.

Since the earth's magnetic field, which is a direct current magnetic field, is inside this annular core, the magnetic field created inside is a high-frequency oscillation that is biased by the earth's lion and field acid components. Then, the high frequency oscillation of the saturated magnetic flux modulated by this earth's magnetic field component causes an electric current 1i M Me in the two output windings,
Output signals are generated at both ends. The amplitude of the second harmonic component of the excitation frequency F of the moon is proportional to the magnitude of the earth's magnetic field, which is the DC component. Therefore, only the second harmonic is amplified in the amplifier, and its peak value is sampled and held in the sample bold circuit.

The output signal (VX, Vν) detected by the output winding 1j112 and the output winding 13 perpendicular to it is expressed as
The trajectory of solid 1 to 1, which is the component, is as described above when the car is 1.
When rotated, it draws a circle as shown in Figure 14.

Next, software for realizing the output correction method, which is the gist of the present invention, using a computer system will be explained. FIG. 5 is a flowchart showing the method. Ste 0. The knob 100 is a step for so-called data initialization.

Here, XC and YC represent the X component and Y component of the center vector after performing a predetermined correction process on the vector locus. Therefore, the value of (XC, YC) is an amount that contributes to direction determination.

In addition, Xold and -'(1+cl indicate the values of the Set it to the value O. In other words, set the ACC switch to the Since it is not moving, wait until it is turned on.Next step 300
Now, since each of the circuits described above generates an output voltage, the output voltages vx and vy are read. The next step 400 is a step of selecting data for finding the center of the vector locus. The data for finding this center are XE, XW, YN, and YSk: read/read. I will explain these numbers here. In FIG. 4, the vector locus 80 is the previously measured vector]
・The vector trajectory 81 is the vector trajectory 81 measured this time.
1-le locus. Then, the coordinates of the corrected center of the previous vector trajectory are XC1,
It is given in YCl. Parallel to the X axis (XC1, YCl
), the X coordinates of the intersection with the vector locus 81 are XW and XE, respectively. . Similarly (XC1,
A straight line and vector locus passing through YCI) and parallel to the Y axis8
The Y coordinates of the intersection with 1 are YN and YS, respectively. Once these values are read, the process moves to the next step 500, where the center coordinates (Xnc
w, Ynew). X new is (XE+
Y new is calculated using the Azusa formula (YN+YS)/2. This determines the center of the newly measured vector trajectory. Next, the process moves to step 600, and it is determined whether the magnetization vector given by the center vector obtained in this way is true. In other words, the central vector correction process according to the gist of the method of the present invention is ignored. This determination is shown in the flowchart shown in FIG. At step 601, if 83 is 1, then
Since this means that the central data has been found, the process moves to the next step 602. In step 602, if the difference between the X component of the new center vector and the X component of the center vector before correction processing determined in the previous measurement does not satisfy the expression lXnew -XOId1≦n, the process moves to step 603. Step 603
Now, let the X component of the new center vector found this time be the X component of the magnetization vector. On the other hand, if the judgment condition is not satisfied in step 602, this means that the new data is significantly deviated from the center X coordinate obtained from the previous measurement value. 604 is corrected using the formula 1). Here, the new center X component is the sum of the center solid X component (XC) used for the direction correction up to now, plus a minute amount approximately 1/10 of the radius of the vector locus.

Then, proceed to the next step 605, step 606,
At 607, a similar correction is made to the Y coordinate in °C. As a result, the components of the center vector after new correction processing are determined. Then, the subsequent direction determination is performed using the center plane (XC, YC) after this correction process. Step 608 is a step in which the center vector before correction is stored as (Xold, Yold). Next, proceeding to step 700, the traveling direction θ of the vehicle is calculated using the following formula.

θ=tan −' ((Vx −XC)/(VV
-Yc)) After the calculation, in step 8''OO, calculation for displaying the orientation is performed, and in step 900, a signal for driving the display device is generated.

Note that the values such as (XC, YC) and (Xold, Yold) are stored even if the main switch is turned off. The present invention can be concretely implemented as described above.

In summary, the present invention provides a method for accurately determining orientation by determining a magnetization vector and correcting measurement data.
This is an attempt to eliminate magnetic vector measurement errors as much as possible. Therefore, if the magnetization vector obtained from the measured value of the magnetic field shows the same value twice in a row, 1. , we adopted that value as the magnetization vector, and if those values deviated significantly, we decided to slightly correct the magnetic field vector used for azimuth correction. In order to adopt such a configuration, according to the method of the present invention, short-period disturbances are eliminated,
It is possible to obtain an extremely accurate DC component, ie, a magnetized flat signal. Moreover, if the magnetization vectors obtained by a predetermined method from two consecutive measurements Ils show similar values,
Since this value is adopted, it has the effect of being able to follow rapid changes in the @magnetic vector in a fast cycle. Therefore, if the direction is determined by employing the magnetization vector which is the DC component corrected as described above, an accurate determination can be made.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram of a vector locus of a signal detected from a μ-j position detection sensor. Figure 2 shows solid 1 to 1 when the signal detected by the direction detection range includes disturbance at J3.
This is the locus. FIG. 3 is a block diagram showing the configuration of a control device that specifically implements the Gosui Komei method. FIG. 4 is an explanatory diagram of a method for determining the center solida. Figure 5 shows computer system 1 used in J3 in the same example.
This is a flowchart for the first draft of the program. FIG. 6 is a detailed 70-chart of step 600 in the same process. 10... Azimuth range 20... Signal processing section 50.
...Computer System Special W [Applicant Nippondenso Co., Ltd. Agent Patent Attorney Hirodo Okawa Patent Attorney Shudo Hizuki Patent Attorney Akio Maruyama Figure 1

Claims (1)

[Claims] In the Peltor locus of the magnetic field detected by the geomagnetic direction detection sensor, a direct current component unrelated to the orientation direction of the geomagnetic direction detection sensor is determined, and the detected ! In the method of determining the direction from the value after correcting the i-field Peltor with the DC component, the DC component determined from the most recently measured vector trajectory and the DC component determined from the previously measured vector trajectory are calculated. If the difference is smaller than a predetermined value, the DC component of the latest measurement will be used for direction correction, and if the difference is greater than a predetermined value, the DC component used for the previous direction correction will be used. An output correction method for an orientation detection device that corrects a value obtained by adding a predetermined correction amount to a DC component as a DC component used for orientation correction.