JPS59218914A - Method for correcting hybrid of direction sensor - Google Patents

Abstract

PURPOSE:To correct respective error factors of an angular speed sensor and an earth magnetism sensor respectively properly by correcting the error factor of one sensor so that the difference between detecting angles by both the sensors is removed. CONSTITUTION:At the running of a car, an angular speed omega due to the directional change of the car is continously detected by the angular speed sensor 1, the detecting signal is amplified 2, the amplified signal is sampled every unit time and converted into a signal indicating declination DELTAtheta corresponding to the advancing of the car by an A/D converter 3, and then the converted signal is applied to a signal processing circuit 9. Simultaneously, a signal indicating relative angle thetaFG between the horizontal component direction of earth magnetism detected by the earth magnetism sensor 4 and the advancing direction of the car is applied to the circuit 9. A pulse signal P outputed from a distance sensor 5 in accordance with the unit travelling distance of the car is sent to a travelling distance counting circuit 7 and a distance signal L is applied to the circuit 9. The frequency of the P output is detected by a frequency detecting 8 and a detecting signal (f) is applied to the circuit 9. When the signal (f) is turned to zero, the circuit 9 detects the stop of the car.

Description

DETAILED DESCRIPTION OF THE INVENTION TECHNICAL FIELD The present invention relates to a hybrid correction method for a direction sensor that detects the moving direction of a mobile object such as an automobile.

Prior Art Recently, direction detectors are used to detect changes in direction as the vehicle moves, and distance detectors are used to detect changes in the direction of the vehicle as it moves. IJIJ detects the distance traveled by the vehicle and uses the detected values to determine the current position on the vehicle's travel route.
An fc travel route display device has been developed that provides guidance by calculating the current location information and displaying it on the screen while comparing it with a road map.

Generally, the direction detector in this type of travel route display device is a rate-type gyroscope that detects the unidirectional change Q angular velocity of the floating object or a gas rate sensor (the displacement of the gas flow caused by the direction change of the moving object). A rate sensor such as a monopoly is used, which detects the change in the amount of heat sensitive by the sensor part and detects the angular velocity at that time.However, this type of rate sensor has good accuracy in detecting changes in direction. However, it has the disadvantage of causing cumulative errors due to drift.

In addition, there is a slight change over time in the detection sensitivity of the rotation angle of the rate sensor, and there is an error that when a moving object runs multiple times, the traveling trajectory of the moving object displayed on the traveling route display device bends around that point. will occur.

Further, as a direction detector, a geomagnetic sensor is used that senses the magnetic field and detects the relative angle between the direction of its horizontal component and the moving direction of the moving object. Direction detectors using geomagnetic sensors always know the current absolute direction, so unlike rate sensors, there is no cumulative drift error. 9.
There are drawbacks.

Therefore, we focused on the advantages of rate sensors and geomagnetic sensors, and chose rate sensors for short time (time differential).

A moving Garo detection method has been proposed in which five combinations of geomagnetic sensors are used for a long time, and the outputs of the image sensors are combined at an appropriate ratio depending on the movement status of the moving body.

However, this method of detecting the direction of movement simply combines the outputs of the rate sensor and the geomagnetic sensor, so if there is a deviation in the sensors, the deviation will remain forever. There are 9 problems. that time,
For example, if a geomagnetic sensor goes awry due to railroad crossing magnetization, the angle detected by the erroneous geomagnetic sensor will appear in the final output.

Purpose The present invention has been made in consideration of the above points, and provides a direction sensor that performs hybrid correction of error factors of each sensor while comparing angles respectively detected by a rate sensor and a geomagnetic sensor. The present invention provides a hybrid correction method.

groove bottom An embodiment of the present invention will be described in detail below.

In the direction sensor hybrid correction method according to the present invention, the first hybrid correction is performed after comparing the detection angles of the rate sensor and the geomagnetic sensor.
The second method is to correct the offset error due to disturbance of the geomagnetic sensor after assuming that the angle detected by the rate sensor is accurate. A means for correcting the drift error of the sensor, and thirdly a means for correcting the sensitivity of the rate sensor due to changes over time, assuming that the angle detected by the geomagnetic sensor is accurate.

First, the offset error correction means of the geomagnetic sensor will be explained below.

When the rate sensor and the geomagnetic sensor start working at the same time, the angle detected by the rate sensor is zeroed to the angle detected by the geomagnetic sensor, so that the angle error between the two sensors becomes zero. In addition, the angle (direction) determined by the rate sensor
Detection involves continuously detecting the angular velocity ω that changes in one direction as the moving body moves, and converting the angular velocity ω signal into a unit time Δt.
By sampling each time, the declination angle Δ0 (Δθ=ω・Δt) corresponding to the direction change is obtained, and the declination angle Δθω
This is done by cumulatively adding up the signals every moment to determine the current moving direction of the moving object.

Based on the angle detected by the geomagnetic sensor (absolute ten thousand degrees), the azimuth plane is divided into 16 equal parts, for example, as shown in Figure 1. The difference between the angle and the angle detected by the rate sensor at each point in time is averaged and stored.In addition, as for the averaging process, for example, the angle at each point in time is This is performed by low-pass filtering, which sequentially weights and averages the differences.

At this time, assuming that the angle detected by the rate sensor is accurate, errors as shown in FIGS. 2(a) and 2(b) occur according to offset errors in each of the x and y directions of the geomagnetic sensor. Therefore, for the offset error ΔX0FF in the X direction shown in (8) of the same figure, Y
If the X-direction offset error ΔX0FF is changed so that the 160,000-position error becomes symmetrical with respect to I+11, the X-direction offset error ΔXO of the geomagnetic sensor
It becomes possible to completely correct FF. In addition, each azimuth error εX when the offset error ΔX□ F F in the X direction is changed is the original azimuth error ε
(, where θ is the averaged angular difference, is given by the following equation.

t X ” e 'X+ΔX0FF −CO5θ
......(1) Also, for the offset error ΔyoB・F in the Y direction shown in FIG.
If the Y-direction offset error ΔYOFF is changed so that the 60,000-position error becomes symmetrical, the Y-direction offset error ΔYOFF in the geomagnetic sensor can be completely corrected. Note that each azimuth error εY when the Y-direction offset error ΔYOFF is changed is 4. When the averaged angular difference is θ, it is given by the following equation.

e Y""e'-z+ΔYOFF' sinθ
(2) Therefore, by correcting each offset error in the X and Y directions in the geomagnetic sensor while making a comparison with the angle detected by the rate sensor, hybrid correction of the direction sensor with good FlllK is performed. It becomes more than possible.

Specifically, when an excessive magnetic field detection means is used to detect an excessive magnetic field that disturbs the geomagnetic sensor when passing through a railroad crossing, etc., it clears the previous 160,000-degree error table (at the same time averages the angular difference). 1 corresponding to the new offset error correction.
If the 60,000-degree error table is incorporated, the offset error of the geomagnetic sensor can be corrected immediately after the magnetization of the railroad crossing.

In addition, in the present invention, the difference between the angle detected by the rate sensor and the angle detected by the geomagnetic sensor is averaged and corrected to the offset error of the geomagnetic sensor (a gentle force). Since the performance of sensors varies depending on time and location, it is desirable to average the sampled values for each fixed distance traveled.

Next, the means for correcting the drift error of the rate sensor will be explained below.

Each time the moving object stops, an offset adjustment is made to zero the declination signal Δθ detected by the rate sensor at that time.After the offset adjustment is performed one after another, the angle detected by the rate sensor and the angle detected by the rate sensor are adjusted at regular intervals. The average difference between the angle and the angle detected by the geomagnetic sensor is calculated.

At that time, assuming that the angle detected by the geomagnetic sensor is accurate, if there is a drift error of x0/sec in the angle detected by the rate sensor, each angle sampled at a fixed time Δt as shown in Figure 3. Difference θ1. θ2
．． θ3. The averaged angular difference characteristic shown by the broken line in the figure is obtained, and the averaged angular difference θm is given by the following equation.

θm=x? /see Xt (see)
--(3) Therefore, by offsetting the detection angle U of the rate sensor by -x'/sec The drift error can be corrected, and even a relatively low-accuracy rate sensor can perform accurate direction detection over a long period of time.

Next, a description will be given below of the sensitivity correction means based on changes in the rate sensor over time.

When the rate sensor and the geomagnetic sensor start working at the same time, the angle detected by the rate sensor is set to match the angle detected by the J1u magnetic sensor so that the angular error between the two sensors becomes zero.

After that, the moving object is made to travel many times, for example, 1 of 360 degrees.
For each rotation, the average difference between the angle detected by the rate sensor and the angle detected by the geomagnetic sensor is calculated.

At that time, assuming that the detection angle of the geomagnetic sensor is accurate, the sensitivity difference due to the rate sensor's change over time is 1%.
If so, as shown in FIG. 4, the valley angle difference θ'1. which is sampled every 1101 rotations of the moving body. θ′2. θ′
5. The averaged angular difference characteristic shown by the broken line in the figure is obtained, and the averaged angular difference θ is given by the following equation.

0' = 360°Xx%Xn rotation...
(4) Therefore, the sensitivity of the rate sensor is corrected by -x% so that the angular difference θ□ becomes zero.
The sensitivity of the rate sensor can be corrected, and even a relatively low-accuracy rate sensor can perform accurate direction detection over a long period of time.

The above-described hybrid corrections for geomagnetic sensor offset error correction, rate sensor drift error correction, and rate sensor sensitivity correction can be performed individually or by combining two or more types of hybrid corrections. They may be executed simultaneously.

In addition, when performing these valley hybrid corrections, the correction process is performed based on the assumption that the angle detected by the -10,000 direction sensor is correct, so there may be a deviation in the direction sensor of both swords. However, in reality, the rate sensor and the geomagnetic sensor have completely different error patterns, so errors from both may be mixed during correction. There are very few cases.

As described above, according to the hybrid direction sensor hybrid correction method according to the present invention, the detected angle l by the rate sensor and the detected angle j by the geomagnetic sensor are compared and compared to each other.
For example, the detection angles of the rate sensor and the geomagnetic sensor can be combined into a correction coefficient according to the movement status of the moving object at an appropriate ratio weighted by J:9. When detecting the moving direction, the precision of the direction detection can be increased synergistically, rather than simply adding the increased precision of the rate sensor and the geomagnetic sensor.

In addition, each hybrid correction of the rate sensor and the geomagnetic sensor explained above is specifically performed in a signal processing section consisting of a microcomputer in the travel route display device, which processes the output signal from the rate sensor and the output from the geomagnetic sensor. Each command is executed by reading the signal.

As described above, in the hybrid correction method for direction sensors according to the present invention, while comparing the angle detected by the rate sensor and the angle detected by the geomagnetic sensor, it is assumed that the angle detected by the -10,000 sensor is accurate. Then, the error factors of the other sensor are corrected so that there is no difference in the detection angle between the two sensors] In this way, each error factor of the rate sensor and the geomagnetic sensor is optimally corrected by the hybrid method. It has the great advantage of being able to

[Brief explanation of the drawing]

Figure 1 is a diagram in which the azimuth plane is divided into 16 equal parts, Figure 2 (- is a diagram showing the state of one offset difference in the X direction of the geomagnetic sensor,
Figure (b) is a diagram showing the offset error state of the geomagnetic sensor in the Y direction. FIG. 4 is a diagram showing angular difference characteristics obtained by sampling and averaging each angular difference between the rate sensor and the geomagnetic sensor every rotation of the moving body. Applicant's agent Kiyoshi Torii Procedural amendment June 27, 1980 Kazuo Wakasugi, Commissioner of the Japan Patent Office ``Sei 1, Indication of the case 1982 Patent Application No. 094529 3, Relationship with the amended person case Patent Applicant: 6-27-8-4 Jingumae, Shibuya-ku, Tokyo, Agent: 1 year, month, 1920 (Delivery date: month, day, 1927) 6. Amendment (details of the number of inventions to be submitted, title of the invention, and Hybrid correction method Claim 1: Compare the angle detected by the angular velocity sensor and the square plate detected by the geomagnetic sensor, and determine the difference between the angles detected by both sensors, assuming that the angle detected by the -10,000 sensor is accurate. A hybrid correction method for a direction sensor, in which the error factor of the other sensor is corrected so that the error factor is eliminated. 2. The angle detected by the angular velocity sensor is assumed to be accurate, and the offset error caused by the disturbance of the geomagnetic sensor is )) Ibrid correction method for the directional sensor according to the description in the m1 section above, which is characterized in that the correction is performed. 3. It is assumed that the angle detected by the geomagnetic sensor is accurate, and the drift error of the angular velocity sensor is A hybrid direction sensor correction method according to the above item 1, in which the cf characteristic is that the correction is performed. 4. It is assumed that the angle detected by the geomagnetic sensor is accurate, and the summer sensitivity correction is performed based on the change over time of the angular velocity sensor. A hybrid correction method for a direction sensor according to item 1 above, characterized in that: Related to hybrid correction method.Prior art Recently, a direction detector is used to detect changes in direction as the vehicle moves, and a distance detector is used to detect the travel distance of the vehicle according to the vehicle speed. Obtained by calculating the current position on the car's travel route,
A driving route display device has been developed that provides guidance by displaying constantly changing current position information on a screen while comparing it with a road map. In general, the direction detector in this type of travel route display device is a rate-type gyroscope that detects the angular velocity of a moving object, or a gas rate sensor (a gas rate sensor that detects the angular velocity of a moving object). Sensors that use angular velocity sensors, such as those that detect the angular velocity at that time by detecting changes in the amount of heat sensitive to the displacement of the flow, are used. However, although such an angular velocity sensor has good accuracy in detecting changes in direction, it has the disadvantage of causing cumulative errors due to drift. In addition, the angular velocity sensor has a slight change in the detection sensitivity of the rotation angle over time, and when a moving object travels many times, the normal trajectory of the moving object displayed on the driving route display device will bend around that point. This caused an error. As the 7cs direction detector, a geomagnetic sensor is used that senses the geomagnetic field and detects the relative angle between the direction of its horizontal component and the moving direction of the moving object. Direction detectors using geomagnetic sensors always know the current absolute direction, so unlike angular velocity sensors, there is no accumulation of drift errors. The disadvantage is that it is greatly influenced. Therefore, considering the advantages of the angular velocity sensor and the geomagnetic sensor, we decided to use the angular velocity sensor for the short time (poetry differentiation) and the geomagnetic sensor for the long time, so that the output of both sensors can be adjusted according to the movement status of the moving object. A moving Garo detection method has been proposed in which the signals are combined in an appropriate proportion.However, such a moving Garo detection method only combines the outputs of the angular velocity sensor and the geomagnetic sensor. -There is a problem that if there is any deviation in Senna, that deviation will remain forever.In that case,
For example, if a geomagnetic sensor becomes erroneous due to magnetization at a railroad crossing, the angle detected by the erroneous geomagnetic sensor will be skewed in the final direction. Purpose The present invention has been made in consideration of the above points, and provides a direction in which hybrid correction of the error factor of the valley sensor is performed while comparing the angles respectively detected by the angular velocity sensor and the geomagnetic sensor. A hybrid sensor correction method is provided. Configuration An embodiment of the present invention will be described in detail below. In the hybrid correction method for a direction sensor according to the present invention, when performing hybrid correction while comparing each detection angle of an angular velocity sensor and a geomagnetic sensor, firstly,
A means for correcting an offset error due to disturbance of the geomagnetic sensor after assuming that the angle detected by the angular velocity sensor is accurate; A means for correcting the drift error of the angular velocity sensor, and thirdly, a means for correcting the sensitivity of the angular velocity sensor due to a change over time at a point where the angle detected by the geomagnetic sensor is deemed to be accurate. First, the correction means for the offset error of the geomagnetic sensor will be explained below. When the angular velocity sensor and the geomagnetic sensor start working at the same time, the angle detected by the angular velocity Iff sensor is adjusted to match the angle detected by the geomagnetic sensor. Set so that the angle error between lengths is zero.In addition, when detecting the angle (direction) using an angular velocity sensor, the angular velocity ω that changes in one direction due to the movement of the moving object is continuously detected, and the angular velocity ω Signal unit time Δ
The declination angle Δθ (Δθ = ω・Δt) corresponding to the direction change is obtained by sampling 1- every t, and the declination angle Δθ
This is done by calculating the current movement of the moving body by cumulatively adding the ω signal every moment. In addition, based on the angle detected by the geomagnetic sensor (absolute azimuth), the azimuth plane is divided into 16 equal parts as shown in Figure 1 (generally, it is divided into 16 or 32 equal parts, etc.). ), and the difference between the angle detected by the angular velocity sensor at each point in time is averaged and stored. Note that the averaging process

[E] For example, this is performed by low-pass filter processing in which the angular difference at each time point is weighted with jlfi and then averaged. At that time, if the angle detected by the angular velocity sensor is accurate, 1
Then, according to the offset error of the geomagnetic sensor in each of the X and Y directions, the errors will occur as shown in Figure 2 (a) and (b), respectively. j For the offset error ΔX0FF in the X force direction, if the offset error ΔX0FF in the It becomes possible to completely correct ΔX0FF.In addition, when changing the offset error ΔX0FF between X directions, each direction) 14 degree error (X is the angle obtained by averaging each direction error by 8zH) When the difference is θ, it is given by the following formula: Here, M is the radius of the circle, that is, the magnitude of the earth's magnetism. Also, the offset error ΔYO of Y Galo shown in Figure (b)
By changing the offset error ΔYOFF of Y Galo so that the 16 force position angle error is symmetrical with respect to FF and the X axis, it is possible to completely correct the offset error ΔY OFF of Y Galo in that geomagnetic sensor. become able to. In addition, the valley azimuth angle summer error ε when changing the offset error ΔYOFF of Y Galo is, when each original azimuth angle error is 6'Y and the averaged angular difference is θ.
It is given by the following equation. Therefore, by correcting the X and Y gallo offset errors in the geomagnetic sensor while comparing the detected angle with the angle detected by the angular velocity sensor, it becomes possible to perform hybrid correction of the direction sensor with high accuracy. Specifically, when an excessive magnetic field detection means is used to detect an excessive magnetic field that disturbs the geomagnetic sensor when passing through a railroad crossing, etc., the previous 16 force position angle error tables are cleared (
At the same time, we will lighten the weighting for averaging the angle differences, and add a 16 force position angle error table that corresponds to the new offset error correction. It becomes possible to correct the offset error of the geomagnetic sensor. In addition, in the present invention, in particular, the difference between the detected angle by the angular velocity sensor and the detected angle by the geomagnetic sensor is averaged to gently correct the offset error of the geomagnetic sensor. Since performance varies depending on time and location, it is desirable to average the sampled values for each distance traveled by the -foot. Next, the means for correcting the drift error of the angular velocity sensor will be explained below. Every time the moving body stops, an offset adjustment is made to zero the declination signal Δθ detected by the angular velocity sensor at that time, and after the offset adjustment is performed sequentially, the angle detected by the angular velocity sensor is adjusted every - foot time. The average difference between the angle detected by the geomagnetic sensor and the angle detected by the geomagnetic sensor is calculated. At that time, assuming that the angle detected by the geomagnetic sensor is accurate, if there is a drift F difference of x'/see in the angle detected by the angular velocity sensor, then the angle detected by the angular velocity sensor is sampled every foot time Δt, as shown in Figure 3. Each angle difference θ4. θ2
．． θ5. The averaged angular difference characteristic shown by the broken line in the figure is obtained, and the averaged angular difference θm is given by the following equation. θm= x0/sec X t(sec)
e value 3) Therefore, in order to make the angular difference θ□ zero, J With this, the drift error of the angular velocity sensor can be corrected, and accurate detection can be performed over a long period of time even with a relatively low degree angular velocity sensor.Next, The sensitivity correction means for changing the angular velocity sensor over time will be explained below. When the angular velocity sensor and the geomagnetic sensor work simultaneously, the angle between the two sensors is adjusted so that the angle detected by the angular velocity sensor is adjusted to the angle detected by the geomagnetic sensor. Set so that the error is zero.Then, make the moving object travel many times, for example, 1 of 360 degrees.
r! ,! For each rotation, the average difference between the gushing angle detected by the angular velocity sensor and the angle detected by the geomagnetic sensor is calculated. At that time, assuming that the angle detected by the geomagnetic sensor is accurate, if the difference in sensitivity due to the angular velocity sensor's change over time is
Each angle difference θ'1 sampled for each rotation. θ′2.
To θ', the averaged angle difference characteristic shown by the broken line in the figure is obtained, and 70) The averaged 2' angle 1 current difference θ1
n is given by the following equation. θ:n = 360°×x%×n rotations...(
4) Therefore, it is possible to correct the sensitivity of the moon angular velocity sensor by -X% so that the angular difference θn becomes zero, and to correct the sensitivity of the rainbow angular velocity sensor. This also makes it possible to carry out highly accurate single detection over a long period of time. The above-described hybrid corrections of offset error correction of the geomagnetic sensor, correction of the drift error of the angular velocity sensor, and sensitivity correction of the angular velocity sensor can be performed individually, or two or more types of hybrid correction may be performed. They may be combined and executed simultaneously. In addition, when executing each of these hybrid corrections, the correction process is performed after assuming that the angle detected by the 10,000 Galo sensors is correct, so the direction sensor of both swords may be slightly off. If the correction does not converge to the correct value, there is a concern that the error may be incorrect.However, in reality, the angular velocity sensor and the geomagnetic sensor have completely different error patterns, so when correcting both There are very few cases where errors occur. As described above, according to the hybrid correction method for a Garo sensor according to the present invention, it is possible to perform correction to improve the accuracy of each other while comparing the angular waves detected by the angular velocity sensor and the angle detected by the geomagnetic sensor. When detecting the direction of movement by combining the detection angles of the angular velocity sensor and the geomagnetic sensor at an appropriate ratio determined by a correction coefficient M according to the movement status of the moving object, the accuracy of the detection The precision of the angular velocity sensor and the geomagnetic sensor can be increased synergistically rather than by adding only ζ. Note that each hybrid correction of the angular velocity sensor and the geomagnetic sensor described above is This is executed by reading the output signal from the angular velocity sensor and the output signal from the geomagnetic sensor in a signal processing unit consisting of a light and a microcomputer in the productivity route display device. FIG. 5 shows a scheme for concretely implementing the hybrid correction method for a Garo sensor according to the present invention.
An angular velocity sensor 1 that detects the angular velocity of the yaw 1, a rate converter 2 that amplifies the detected output, A-1) conversion 153 that samples the amplified signal, and a geomagnetic sensor 1. 4, a distance sensor 5 that generates a pulse signal for each unit movement distance of the automatic vehicle, a waveform shaping circuit 6 that shapes the waveform of the pulse output, and a travel distance that counts the waveform-shaped pulse signal. A counting circuit 7, a frequency detection circuit 8 for detecting the output frequency of the waveform shaping circuit 6, and the A-D converter 3. The signal processing circuit 9 includes a microcomputer that performs predetermined calculations based on the outputs of the mileage counting circuit 7 and the frequency detection circuit 8. In a device configured in this way, when the car is running, the angular velocity ω of the car as it changes in one direction is continuously detected by the angular velocity sensor l, and after the detection signal is amplified by the rate amplifier 2, D The signal is sampled by the converter 3 every unit time Δt and converted into a signal with a declination angle Δθ (Δθ=ω・Δt) according to the direction of travel of the vehicle,
The declination signal Δθ is provided to the signal processing circuit 9. At the same time, the relative angle θ between the horizontal component of the geomagnetic field detected by the geomagnetic sensor 4 and the traveling direction of the vehicle
The FG signal is given to the Q processing circuit 9. Also, the pulse signal P output from the distance sensor 5 according to the unit mileage of the car is sent to the mileage counting circuit 7 through the waveform shaping circuit 6. The distance signal counted there is given to the signal processing circuit 9, and the frequency detection circuit 8 detects the output frequency of the pulse signal P, and the detected signal f is given to the signal processing circuit 9. Signal processing circuit 9
When the detection signal f becomes zero, it becomes possible to detect that the vehicle is in a stopped state. FIG. 6 shows a flowchart of the main routine in the signal processing circuit 9. Further, FIG. 7 shows a flowchart 10-1 of the hybrid correction processing of the geomagnetic sensor in the signal processing circuit 9, and FIG. 8 shows a flowchart of the hybrid correction processing of the angular velocity sensor. As described above, in the hybrid direction sensor correction method according to the present invention, while comparing the angle detected by the angular velocity sensor and the angle detected by the geomagnetic sensor, it is assumed that the detected angle by one sensor is accurate, and then, The error factors of the other sensor are corrected so that there is no difference in the angle detected by both sensors, and the hybrid method allows each error factor of the angular velocity sensor and the geomagnetic sensor to be optimally corrected. It has this excellent advantage. Brief explanation of the drawings Figure 1 is a diagram dividing the azimuth plane into 16 equal parts, Figure 2 (a) is a diagram showing the offset error state of the geomagnetic sensor in the A diagram showing the offset error state, Figure 3 is a diagram showing the angular difference characteristics obtained by sampling and averaging each angular difference between the angular velocity sensor and the geomagnetic sensor every unit time, and Figure 4 is a diagram showing the angular difference characteristics between the angular velocity sensor and the geomagnetic sensor. A diagram showing the angular difference characteristics obtained by sampling each angular difference every rotation of the moving body and averaging it. FIG. 7 is a flowchart of the main routine in the signal processing circuit I of the embodiment, FIG. 7 is a flowchart of the hybrid correction processing of the geomagnetic sensor in the same embodiment, and FIG. 8 is a flowchart of the hybrid correction processing of the angular velocity sensor. . ]...Angular velocity sensor 2...Rate amplifier 3...
・A-D converter 4... Geomagnetic sensor 5...
Distance sensor 6...Waveform shaping circuit 7...Distance argument circuit 8...Frequency detection circuit 9...Signal processing Question 1 Applicant's agent Masaru Karasui Figure 5 Figure 6

Claims (1)

[Claims] Compare the angle detected by the 1, t, '-) sensor with the angle detected by the geomagnetic sensor, and assume that the angle detected by -10,000 centimeters is accurate. A hybrid correction method for a direction sensor, which takes means for correcting error factors of the other sensor so as to eliminate the difference between the two sensors. 2. The hybrid direction sensor correction method according to item a<1 above, wherein the angle detected by the rate sensor is assumed to be accurate, and an offset error due to disturbance of the geomagnetic sensor is corrected. 3. The direction sensor hybrid correction method according to item 1 above, wherein the angle detected by the geomagnetic sensor is assumed to be accurate and the drift error of the rate sensor is corrected. 4. Direction sensor Q) hybrid correction method according to item 1 above, characterized in that the angle detected by the geomagnetic sensor is assumed to be accurate, and sensitivity correction is performed based on changes in the rate sensor over time.