JPH07286852A - Drift canceling device for on-vehicle gyroscope - Google Patents

Abstract

PURPOSE:To cancel the DC-level drift of a gyroscope so as to make the gyroscope to output accurate signals by sampling the levels of detected signals only when the differences between the levels of the detected signals and the zero level of a prefixed signal are equal to or smaller than a threshold. CONSTITUTION:The signal output of a piezoelectric vibration gyroscope 11 is inputted to a microcomputer 18 through amplifiers 14 and 15 after a reference voltage from a reference voltage source 12 is subtracted from the signal output at a subtractor 19 and the output of a D/A converter 16 is subtracted from the signal output at another subtractor 1B. The microcomputer 18 samples the levels of detected signals only when the differences between the levels of the detected signals and the zero level of a prefixed signal are equal to or smaller than a threshold by repeatedly referring to the levels of detected signals within a prescribed period of time and calculates the average level of a plurality of levels. When the difference between the calculated average level and zero level is larger than a prescribed value, the microcomputer 19 corrects the zero level based on the average level. Since the zero level is always corrected and controlled in such a way, such a signal that the drift of it still level is offset can be outputted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drift canceling device for a gyro on a vehicle, and particularly, although not intended to be limited to this, a drift canceling device disclosed in Japanese Patent Application No. 5-191189 by the present inventor. Regarding improvement.

[0002]

2. Description of the Related Art Some sensors, such as a piezoelectric vibrating gyro, output a signal containing a direct current component, whose direct current level generally fluctuates according to temperature and humidity. The drift of the DC level of this type of signal becomes remarkable when the gain of the amplifier that amplifies the signal is increased in order to increase the detection sensitivity, thereby increasing the detection error.

In order to suppress the occurrence of an error caused by such a change in the DC level, a capacitor coupling circuit for cutting off the DC component has conventionally been installed at the output of the sensor. Since no DC component appears in the output of this capacitor coupling circuit, a signal with drift removed can be obtained.

[0004]

However, since the capacitor coupling circuit attenuates not only the DC component but also the AC component having a low frequency, when the signal contains an AC component having a very low frequency, it is detected. A portion of the desired signal is missing or attenuated. For example, in the case of detecting a changing angular velocity in the traveling direction of a vehicle using a piezoelectric vibration gyro, if the traveling direction of the vehicle changes slowly, the output of the piezoelectric vibration gyro corresponds to an alternating current with a very low frequency. The ingredients appear. Therefore, it is necessary to directly reference the output signal of the piezoelectric vibrating gyroscope without using a capacitor coupling circuit in order to correctly detect a slow change in the traveling direction, which causes a problem of DC level drift. .

Further, if the time constant of the capacitor coupling circuit is increased, the attenuation of low frequency AC components can be reduced, but there is a limit to increase the time constant. That is, since the capacitance of the capacitor becomes large, the circuit becomes large, and in a circuit having a large time constant, a large voltage fluctuation occurs even if the leakage current of the capacitor is small, and a new DC level drift occurs in the output of the amplifier.

Therefore, it is an object of the present invention to cancel the drift of the DC level of the gyro on the vehicle to output an accurate signal and prevent the low frequency signal component from being attenuated or missing.

[0007]

In order to solve the above problems, in the first invention, a gyro (11) on a vehicle for outputting a detection signal containing a DC component and an AC component; corresponding to the speed of the vehicle Threshold value setting means (30) for setting a threshold value (Vw) that is large at low speed and small at high speed; the level of the detection signal (VS2) is referred to The difference between the zero level (Vc) and the threshold value (V
w) identifying means (2A, 2B) for identifying whether or not the level of the detection signal is repeatedly referred to within a predetermined time, and the difference between the level of the detection signal and a predetermined zero level of the signal is described above. Average calculation means (2C, 2F) for sampling the level of the detection signal only when it is within the threshold value and calculating the average (Mc) of a plurality of levels sampled within the predetermined time; If the difference (ΔVd) between the average and the zero level is greater than or equal to a predetermined value (Vth), the zero level is corrected based on the average, zero level adjusting means (2J, 2K); and the level of the detection signal. And signal output means (22) for outputting a signal corresponding to the difference between the zero level and the zero level.

According to a second aspect of the invention, the average calculating means of the first aspect of the invention is configured to invalidate the calculation result (2E) when the number of samplings within the predetermined time is less than a predetermined number. To do.

In the third aspect of the invention, the gyro (1) on the vehicle that outputs the detection signal containing the DC component and the AC component is used.
1); Amplifying means (15) for inputting and amplifying the detection signal to generate an output signal; Base signal generating means (16) for outputting a DC level base signal; A synthesizing means (1B) for adding or subtracting to the input of the amplifying means;
First discriminating means (24, 25) for discriminating whether or not the difference between the detected signal level and the base signal level is larger than a predetermined first threshold value by referring to the detected signal level and the base signal level;
A first average calculating means (23b, 23b, which calculates an average level (Vb) of the detection signal within the time when the difference is larger than the first threshold value for a first time or more.
27, 28a); Base level changing means (28b) for changing the level of the base signal output by the base signal generating means according to the average level; Low speed corresponding to the speed of the vehicle Threshold value setting means (30) for setting a second threshold value (Vw) that is large and small at high speed; the level (VS2) of the output signal of the amplifying means is referred to, and the level of the signal is predetermined. Second discriminating means (2A, 2B) for discriminating whether or not the difference between the signal and the zero level (Vc) is within the second threshold value (Vw); the level of the output signal is set within the second time. Repeatedly referred to, the level of the detection signal is sampled only when the difference between the level of the output signal and the predetermined zero level of the signal is within the second threshold value, and is sampled within the second time. Calculate the average (Mc) of multiple levels The second average calculation means that (2C, 2
F); if the difference (ΔVd) between the average calculated by the second average calculating means and the zero level is equal to or more than a predetermined value (Vth),
Zero level adjusting means (2J, 2K) for correcting the zero level based on the average; and signal output means (22) for outputting a signal according to the difference between the level of the output signal and the zero level. Set up.

The symbols shown in parentheses are reference numerals of corresponding elements in the embodiments described later, but each constituent element of the present invention is a specific element in the embodiments. It is not limited to only.

[0011]

For example, as shown in FIG. 4, the DC level (Vre
When the sensor outputs the signal Vin including f), generally, only the AC component of the signal Vin is amplified A times and output. When the DC level of the signal Vin matches the reference level (Vref), the signal (A (Vin-Vref)) obtained by amplification is obtained.
The DC level of + Vref) also matches the reference level Vref.
However, when the DC level of the sensor drifts, the DC level of the signal obtained by amplification is displaced from the reference level Vref by a level obtained by multiplying the drift amount by A. When the DC level of the output signal is at the center of the power supply voltage, even if the amplitude of the AC component of the signal is relatively large,
Although signal level saturation does not occur, when the DC level of the signal is largely deviated from the center of the power supply voltage, the amplitude of the AC component is large, and as shown in the middle and lower stages of FIG. Or the level saturates when approaching 0V). Even when saturation does not occur in the amplifier, the correct signal level cannot be detected if the level of the output signal is out of the input range of the circuit (eg, A / D converter) that processes it. In any case, when referring to the signal level without blocking the DC component, a large detection error occurs depending on the variation of the DC level.

According to the first invention, the signal output means is
A signal corresponding to the difference between the level of the detection signal and the zero level is output. This zero level is automatically adjusted by the zero level adjusting means. That is, the level of the detection signal is repeatedly referred to within a predetermined time (for example, 15 seconds), and the average of many levels sampled when the difference between the level of the detection signal and the zero level is within the threshold value (Vw). If the difference from the zero level is greater than or equal to a predetermined value, the zero level is corrected based on the average.

If the zero level coincides with the stationary level (DC level) of the detection signal, and if the threshold value Vw is set to be extremely small, the level of the detection signal when the detection signal does not include an AC component. And the difference between the zero level and the zero level is within the threshold value, and when an AC component (effective signal component) appears in the detection signal, the difference between the level of the detection signal and the zero level exceeds the threshold value. Therefore, the average of a large number of levels sampled when the difference between the level of the detection signal and the zero level is within the threshold value is substantially equal to the actual static level (DC level) of the detection signal. Therefore, when the difference between the average level and the zero level is large, it is considered that the zero level is deviated, and the zero level is corrected based on the average level. As a result, the zero level is constantly controlled so as to match the actual quiescent level (DC level) of the detection signal. Therefore, by outputting the difference between the detection signal and the zero level, a signal that cancels the drift of the quiescent level is output. Can be output.

According to the second aspect of the invention, when the number of data sampled by the average calculating means within the predetermined time is smaller than a predetermined number, the data sampled at that time becomes invalid and the average level is zero. No correction process is performed. That is, when the number of data sampled within the predetermined time is small, the period in which the effective AC component is included in the detection signal is long and the error in calculating the static level (DC level) increases. , This error is prevented from adversely affecting the zero level accuracy.

According to the third aspect of the invention, the amplifying means amplifies the signal obtained by adding or subtracting the detection signal and the base signal. When the difference between the level of the detection signal and the level of the base signal is larger than the first threshold value (ΔV) for a long time (for example, 25 seconds), the detection signal within the time Is calculated by the first average calculating means, and the level of the base signal is automatically changed corresponding to this average level. The signal output means outputs a signal according to the difference between the level of the output signal and the zero level. This zero level is automatically adjusted by the zero level adjusting means. That is, the level of the output signal is repeatedly referred to within a predetermined time (for example, 15 seconds), and a large number of levels sampled when the difference between the level of the output signal and the zero level is within the second threshold value (Vw). If the difference between the average (Mc) of (VS2) and the zero level is greater than or equal to a predetermined value,
The zero level is corrected based on the average (Mc).

By the way, since the position change speed of the vehicle is proportional to the vehicle speed, the position and direction change speeds are different in proportion to the vehicle speed even if the direction changes at the same angular velocity. Therefore, at high speeds, the piezoelectric vibration gyro output is as fine as possible. It is preferable to be able to extract fluctuations (changes in the level of the angular velocity signal). So the first
Each of the third to third inventions includes a threshold value setting means (30) for setting a threshold value (Vw) that is large at low speed and small at high speed, corresponding to the vehicle speed (V). As a result, minute output fluctuations (changes in the angular velocity signal level) of the piezoelectric vibration gyro are extracted at high speeds. At high speeds, the turning time of the vehicle is short, and changes in the drift level during this period are virtually unthinkable, so there is virtually no possibility of mistakenly extracting drift as an angular velocity signal at high speeds. At low speeds, the extraction sensitivity of the angular velocity signal is relatively lower than at high speeds, but since it is low, the vehicle position change speed and direction change speed due to the vehicle speed and angular speed (turn speed) are slow. , Vehicle position recognition error is minimal.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an example of the structure of an apparatus for carrying out the present invention. The device shown in FIG. 1 is an azimuth measuring device that can be mounted on a vehicle and used. Referring to FIG. 1, the device includes a piezoelectric vibration gyro 11, a reference voltage source 12, a buffer 13, amplifiers 14, 15, a D / A converter 16, an A / D converter 17,
Microcomputer 18, subtractors 19, 1B, adders 1A, 1C, antenna ANT, GPS receivers 1D and G
The PS demodulator 1E is provided.

The piezoelectric vibrating gyro 11 used in this embodiment is a model ENC-05 manufactured by Murata Manufacturing Co., Ltd.
S, which has a reference voltage input terminal R, a signal output terminal S, a power supply terminal Vcc, and an earth terminal G. The piezoelectric vibrating gyro 11 outputs to the signal output terminal S a voltage that is proportional to the angular velocity of rotation about one axis. In this embodiment, in order to detect the azimuth, the piezoelectric vibrating gyro 11 is arranged with the axis of rotation oriented vertically.

Since the rotation angle of the piezoelectric vibrating gyro 11 (that is, the rotation angle of the automobile) is obtained by integrating the detected angular velocity, the reference azimuth detected at a certain time,
The azimuth (direction of travel) of the automobile can be obtained based on the rotation angle obtained from the angular velocity. In this embodiment, GPS positioning is performed using the GPS receiver 1D and the GPS demodulator 1E to detect the accurate reference azimuth.
However, since it takes a relatively long time to carry out GPS positioning, the azimuth that changes momentarily cannot be detected in real time only by carrying out the GPS positioning. Therefore, in this embodiment, real-time direction detection is realized by combining the detection of the reference direction by GPS positioning and the detection of the rotation angle using the piezoelectric vibration gyro 11.

By the way, according to the specification, the piezoelectric vibrating gyro 11 outputs a DC voltage Vo within the range of Vref ± 500 mV to the output terminal S when there is no rotation, and a voltage of Vo + 72 mV when rotating at the maximum angular velocity. Is output.
Also, since a relatively large drift occurs in the output voltage,
It is a standard usage to cut the direct current component through the capacitor coupling circuit and take out the signal at the output terminal S. However, when a capacitor coupling circuit is used, a signal component having a very low frequency is significantly attenuated or missing, so that it is convenient for an application that also needs a signal component that changes slowly such as a rotational angular velocity of an automobile. Is bad. Therefore, in this embodiment, a special control is performed by using the microcomputer 18, and the drift of the output level of the piezoelectric vibration gyro 11 is canceled without providing the capacitor coupling circuit.

The signal Vin output from the signal output terminal S of the piezoelectric vibrating gyro 11 passes through the buffer 13 and is applied to the subtracters 19 and 1B, respectively. In the subtractor 19, a voltage obtained by subtracting a constant reference voltage (DC voltage) Vref output from the reference voltage source 12 from the signal Vin is input to the amplifier 14 and amplified. In the subtractor 1B, the signal V
A voltage obtained by subtracting the base signal Vbase output from the D / A converter 16 from in is input to the amplifier 15 and amplified. In this example, the gain A1 of the amplifier 14 is 1.8 and the gain A2 of the amplifier 15 is 28.6. Adder 1A
In, the result of adding the output signal of the amplifier 14 and the reference voltage Vref is the signal VS1, and the result of adding the output signal of the amplifier 15 and the reference voltage Vref is the signal VS2. These signals VS1 and VS2 are respectively A /
It is input to the microcomputer 18 via the D converter 17.

The reference voltage Vref is 2.5V in this example. Originally, the signal Vin has the same voltage as Vref when there is no rotation, as shown in the uppermost stage of FIG. 7, the voltage increases when the angular velocity in the forward rotation direction is detected, and the voltage increases when the angular velocity in the reverse rotation direction is detected. Decrease. However, in reality, since a drift occurs, the voltage of the signal Vin when there is no rotation is displaced from the reference voltage Vref as shown in the middle and lower stages of FIG. . Therefore, for example, (Vin-Vref) is amplified A1 times and then V
When ref is added, as shown in FIG. 7, the static level of the voltage after amplification (the level at the time of non-rotation) greatly varies with the drift of the voltage of the signal Vin, so that when a large rotational angular velocity appears, The signal reaches the saturation level and the correct level cannot be detected. In any case, when the stationary level of the signal fluctuates, the correspondence between the voltage of the signal and the rotational angular velocity is deviated, so that a detection error of the rotational angular velocity occurs.

Therefore, in this embodiment, first, even if the static level of the signal Vin fluctuates, the actual static level of the signal Vin is prevented so that the level of the signal VS2 obtained at the output of the amplifier 15 having a large amplification degree is not saturated. A base signal Vbase having a level close to is generated, and the offset of the amplifier 15 is automatically adjusted by using this base signal Vbase. As a result, even when the static level of the signal Vin is largely displaced from the reference voltage Vref as shown in FIG. 8, it is possible to prevent the level of the signal having a large amplitude from being saturated. In this embodiment, in order to detect the static level of the signal Vin, the signal VS1 obtained by amplifying the signal Vin by the amplifier 14 is referred to, but the level of the signal Vin may be directly referred to.

The operation of the microcomputer 18 is shown in FIGS. FIG. 3 is a main routine, and FIG. 2 is a timer interrupt 1 routine. When the power supply of the device is turned on, in step 31, initialization is first executed, in which a pulse interrupt ((a) in FIG. 4) and a timer interrupt 2 in response to the vehicle speed synchronization pulse for calculating the vehicle speed V are performed. ((B) of FIG. 4 is permitted. Next, GPS positioning calculation is executed,
The output invalidation flag is cleared and timer interrupt 1 (Fig. 2) is enabled.

When the vehicle speed synchronization pulse of 1 pulse is given to the microcomputer 18 for the rotation of the axle by a predetermined small angle, and the pulse interruption and the timer interruption 2 are permitted,
The computer 18 executes the interrupt process shown in FIG. 4A every time one vehicle speed synchronization pulse arrives. That is, the pulse count register RVC is incremented by 1 (the data in the register is incremented by 1) (41). When the timer interrupt 2 is enabled, the microcomputer 18 starts the interrupt timer 2 (program timer), and responds to the interrupt timer 2 time-over when it responds to the interrupt timer 2. Figure 4 after restarting
The timer interrupt 2 shown in (b) is executed. That is, the vehicle speed V is calculated from the data RVC of the pulse count register RVC.
Is calculated and written in the vehicle speed register V (42), and the pulse count register RVC is cleared (43). Assuming that the time limit value of the interrupt timer 2 is T _t2 , the vehicle speed V is
_It is proportional to the number of vehicle speed synchronization pulses (RVC) during _t2 . Based on this relationship, the vehicle speed V is calculated in step 42 and written in the vehicle speed register V. This update write is T
_Since it is performed at _t2 , the latest vehicle speed is always secured in the vehicle speed register V. Referring again to FIG. 3, in the GPS positioning calculation, the information received from three or more satellites is sent to the GPS.
It is input via the receiver 1D and the GPS demodulator 1E, a predetermined calculation is performed based on these information to obtain the azimuth in the traveling direction, and the azimuth obtained here is set as the reference azimuth ψref.
When the timer interrupt 1 is permitted in step 31, the timer interrupt 1 rule shown in FIG.
Chin is executed. In this embodiment, the timer interrupt 1 request is generated periodically every 20 msec.

The routine of the timer interrupt 1 will be described with reference to FIG. In step 21, the two signals VS1 and V
S2 is sampled by the A / D converter 17 and
Input the result of / D conversion. In step 22, the result of subtracting the zero level Vc from the value of the sampled signal VS2, that is, the value corresponding to the rotational angular velocity ω is set in the register Rω.
Store at. In the next step 23a, the register Rr
And the contents of the register Rω are added and the result is stored in the register Rr. In the next step 23b, the contents of the register Rt and the value of the sampled signal VS1 are added and the result is stored in the register Rt. That is, the register Rr holds the integrated value of the rotational angular velocity ω, and the register Rt holds the integrated value of the signal VS1.

In the following step 24, the sampled value of the signal VS1 is compared with the upper threshold level Vbase + ΔV, and in step 25, the value of the signal VS1 is compared with the lower threshold level Vbase−ΔV. The threshold ΔV is
In this example, it is set to 56 mV. And the signal VS1
If the value is within the upper and lower threshold values, the process proceeds to step 26, and if it is out of this range, the process proceeds to step 27. That is, the value of the signal VS1 and the base signal Vbase
When and are substantially equal, the process proceeds to step 26, and when the difference between the two is large, the process proceeds to step 27.

In step 27, after adding 1 to the content of the register Tbase, the content of Tbase is referred to, which is 25
Identify whether it is more than a second. Timer interrupt cycle is 20 mse
c, and the content of Tbase is incremented by 1 every 20 msec, so in step 27, the content of Tbase and 1250 are actually added.
(= 50 × 25) is compared. When the content of Tbase is less than 25 seconds, the process proceeds to step 2A. When the content of Tbase is 25 seconds or more, the processes of steps 28a, 28b and 29 are executed, and then the process proceeds to step 2A. If the value of the signal VS1 falls within the upper and lower threshold values before 25 seconds have passed, the register Tbase is determined in step 26.
And the contents of Rt are cleared. That is, if the value of the signal VS1 is outside the range of the upper and lower threshold values for 25 seconds, the process proceeds to step 28a.

In step 28a, the signal VS held in the register Rt and sampled for 25 seconds is used.
The integrated value of 1 is the sampling number N25 (= 1250) during that period.
The result, that is, the average value Vb of the signal VS1 for 25 seconds is obtained, and the value obtained by converting the average value Vb into the average level of the input signal Vin is stored in the register Vb2. In the next step 28b, the contents of the register Vb2 are D / A
Output to the converter 16. As a result, the base signal Vba
The voltage of se is updated. That is, the average value of the signal Vin for 25 seconds becomes the new voltage of Vbase.

For example, as shown in FIG. 8, the static level of the signal Vin deviates from the reference voltage Vref due to the drift of the signal Vin, but the average value of the voltage of the signal VS1 is calculated as the base signal Vba.
It is output as se, and the base signal Vba
By sequentially updating se, the voltage of the base signal Vbase can be brought close to the static level of the signal Vin (and VS1), whereby only the signal obtained by removing the DC component from the signal Vin is amplified by the amplifier 15. Therefore, the quiescent level of the signal VS2 approaches the reference voltage Vref. Therefore, even when the amplitude of the signal Vin is large (the angular velocity is large), the saturation of the voltage as shown in FIG.
It is unlikely to occur in S2.

Referring again to FIG. Step two
When the base signal Vbase is updated in 8b, the output invalid flag is set in the next step 29 and the register Rr,
Clear the contents of Rt and Tbase. That is, the base signal V
When the base changes, the quiescent level of the signal VS2 changes greatly in a stepwise manner, and the data before updating the base signal Vbase is likely to include a relatively large error. Disable the affected data.

In step 30, a threshold value Vw corresponding to the vehicle speed V is calculated and set. Vehicle speed V is vehicle speed register V
Read from the calculation formula Vw set in the program
= Vwo- (Vwo-a) V / Vv is used to calculate the threshold value Vw and is written in the threshold value register Vw. FIG. 5 shows the relationship between the vehicle speed V and the threshold value Vw according to this calculation formula.

In step 2A, the value of the signal VS2 is compared with the threshold level Vc + Vw, and in the next step 2B,
The value of signal VS2 is compared to a threshold level Vc-Vw.
Vw is a value indicated by the data of the register Vw. VS2> V
If c + Vw, proceed from step 2A to 2D, VS2> V
If c-Vw, the process proceeds from step 2B to 2D. Also VS2
Is within the range between the threshold levels Vc-Vw and Vc + Vw, the routine proceeds to step 2C. The threshold value Vw is set to 7.3 mV in this example.

In step 2C, first, the register Rt2
The value of the sampled signal VS2 is added to the contents of the above, the result is stored in the register Rt2, and 1 is added to the contents of the register Cs. The register Rt2 holds the integrated value of the value of the signal VS2, and the register Cs holds the data number of VS2 integrated in Rt2.

At the next step 2D1, 1 is added to the content of the register Tw. The number of samples is verified in the next step 2D2. This content is shown in FIG. Here, 1 is added to the content of the reference count register Cso (4
4). Then, the ratio Cs / Cso of the number of times Cs of sampling of the register Cs to the number of times of reference Cso of the register Cso is calculated, and it is checked whether this is 56.2% or more (4
5). If so, the number of samples for the zero level calibration is sufficient, and therefore the process proceeds to the next check 2D3 (FIG. 2).
If it is less than 56.2%, for example, the high level VS2 to be extracted shown in FIG. 9 within the current time period of 15 seconds is shown.
There is (a level exceeding Vc + Vw), and the reference count register Cso is counting up during that time, but the sample count register Cs is not counting, so the number of zero level samples (the level below Vc + Vw in FIG. 9 is extracted). Since the number of times) is too small, the register Rt2 is used to cancel the present 15-second time count section and the previous data therein and start a new 15-second time count and data collection. Clear Cs, Cso and Tw (46) and return to the main routine.

Now, in the above verification (2D2), Cs / C
When so is 56.2% or more, the content of the register Tw is compared with 15 seconds in the next step 2D3 of FIG. Actually, the content of Tw is compared with 750 (= 50 × 15). If Tw ≧ 15 seconds, then step 2E
Proceed to.

In step 2E, the contents of register Cs and 422 are compared. That is, register Rt within 15 seconds
The data number of VS2 accumulated in 2 is compared with 422.
If Cs> 422, the process proceeds to step 2F, and if not, the process proceeds to step 2G. In step 2F, the contents of the register Rt2 are divided by the contents of the register Cs, and as a result, that is, Cs pieces of data sampled in 15 seconds.
The average value of the data is stored in the register Mc.

In the next step 2H, it is checked whether or not there is valid data in the register Mp. When the data is present, the process proceeds to step 2I, otherwise proceeds to step 2L. At step 2L, the value of the register Mc is stored in the register Mp.

In step 2I, the absolute value of the difference between the contents of the register Mc (current average value) and the contents of the register Mp (previous average value) is set to ΔVd, and the next step 2J
Then, the difference ΔVd obtained in step 2I is set to the threshold value Vth.
Compare with. The threshold value Vth is 2.44 m in this example.
It is set to V. ΔVd> Vth, that is, 1 of the signal VS2
If the average value of the data within the range of Vc ± Vw for 5 seconds has changed by Vth or more as compared with the previous time, then step 2K is executed. In step 2K, the contents of the register Mc, that is, the average value of this time is set as a new zero level Vc and stored in the register Mp. In addition, the registers Rt2, C
Clear the contents of s, Cso, and Tw.

The data (VS2) accumulated in the register Rt2 in step 2C is zero level V as shown in FIG.
When the angular velocity signal component appears in a very narrow range (± Vw) with respect to c, the level of the signal VS2 is Vc.
Since it is out of the range of ± Vw, the average value (Mc) obtained in step 2F becomes the average level when the signal VS2 is at the stationary level for 15 seconds. By drifting
For example, if the quiescent level increases or decreases over time,
The difference (ΔVd) between the current average value (Mc) and the previous average value (Mp) becomes larger than the threshold value Vth, and the zero level Vc is automatically updated in step 2K. Therefore, the zero level Vc is constantly adjusted so as to approach the actual static level. Therefore, the data stored in the register Rω in step 22 is the rotational angular velocity detected by the piezoelectric vibration gyro 11.

Description will be made with reference to FIG. When the output invalidation flag is not set, step 34 is executed. In step 34, the integrated value of the rotational angular velocities held in the register Rr, that is, the rotational angle is added to the reference azimuth .psi.ref obtained by the GPS positioning calculation, and the result is set as the current azimuth data. Then, this azimuth data is output to the outside.

If the difference between the stationary level and the base signal Vbase becomes large due to the drift of the signal Vin and the output invalid flag is set, step 33 is executed. In step 33, interrupts are first prohibited, the contents of various registers are cleared, then GPS positioning calculation is executed, the zero level Vc is initialized, the output invalidation flag is cleared, and interrupts are permitted. The rotation angle (contents of Rr) is cleared,
A new reference bearing ψref is detected by the GPS positioning calculation.

FIG. 10 shows the measurement results of the changes in the detected orientations when the drift is detected and the zero level Vc is sequentially corrected and when the Vc is not updated by the apparatus of the above embodiment. This experiment was carried out under the condition that the apparatus was installed in a stationary state and only the influence of drift appeared as a change. With reference to FIG. 10, the device of the embodiment sufficiently suppresses the influence of drift, and provides an accurate rotation angle (azimuth).
It can be seen that

In the above embodiment, the case where the azimuth in the traveling direction of the automobile is detected using the piezoelectric vibration gyro has been described, but various other detections that output a signal including a DC component that causes drift similarly. The present invention can be implemented in the same manner as in the above-described embodiment when processing a signal of a container.

It is possible to omit the amplifier 14 shown in FIG. 1 and refer to the signal Vin instead of the signal VS1, but in that case, the control error of the base signal Vbase increases to some extent.

[0047]

As described above, according to the present invention, since the drift of the DC level of the detector output is canceled without installing a capacitor coupling circuit, slow signal changes other than drift are lost or attenuated. It is possible to prevent an error and read an accurate signal.

A threshold value setting means (30) for setting a threshold value (Vw) that is large at low speed and small at high speed corresponding to the vehicle speed (V) is included. As a result, minute output fluctuations (changes in the angular velocity signal level) of the piezoelectric vibration gyro are extracted at high speeds. At high speeds, the steering time of the vehicle is short,
Since the change of the drift level during this period is virtually unthinkable, there is substantially no possibility of mistakenly extracting the drift as an angular velocity signal at high speed. When the speed is low, the extraction sensitivity of the angular velocity signal is
Although it is relatively low as compared with the case of high speed, since it is low speed, the vehicle position change speed and the direction change speed due to the vehicle speed and the angular speed (turn speed) are slow, so that the vehicle position recognition error remains small.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an apparatus according to an embodiment.

2 is a timer interrupt 1 of the microcomputer of FIG.
It is a flowchart showing the contents of.

3 is a flowchart showing a main routine of the microcomputer shown in FIG.

4A is a flow chart showing the contents of pulse interruption of the microcomputer of FIG. 1, and FIG. 4B is a flow chart showing the contents of timer interruption 2. .

5 shows V calculated by "calculate Vw" 30 shown in FIG.
It is a graph which shows the relationship between w and vehicle speed V.

FIG. 6 is a flowchart showing the contents of “verification of sample number” 2D2 shown in FIG.

FIG. 7 is a waveform diagram showing an example of a signal output by the piezoelectric vibration gyro 11 shown in FIG. 1 and a signal obtained by amplifying the signal.

8 is a waveform diagram showing the levels of signals Vin, VS2 and a base signal of the apparatus of FIG.

9 is a waveform diagram showing a signal VS2, a zero level Vc, and a threshold level of the device of FIG.

10 is a graph showing measurement results of drift of the apparatus of FIG.

[Explanation of symbols]

11: Piezoelectric vibrating gyro 12: Reference voltage source 13: Buffer 14, 15: Amplifier 16: D / A converter 17: A / D converter 18: Micro computer 19, 1B: Subtractor 1A, 1C: Adder 1D: GPS receiver 1E: GPS demodulator VS1, VS2: signal Vref: reference voltage Vbase: base signal Vc: zero level

Front Page Continuation (72) Inventor Tomio Hota 2-1-1 Asahi-cho, Kariya City, Aichi Prefecture Aisin Seiki Co., Ltd.

Claims (3)

[Claims] 1. A gyro on a vehicle that outputs a detection signal containing a DC component and an AC component; threshold setting means for setting a threshold value that is large at low speed and small at high speed in correspondence with the speed of the vehicle; Discriminating means for discriminating whether or not the difference between the level of the detection signal and a predetermined zero level of the signal is larger than the threshold value set by the threshold value setting means by referring to the signal level; a predetermined time The reference level of the detection signal is repeatedly referred to within, and only when the difference between the level of the detection signal and the predetermined zero level of the signal is within the threshold value, the level of the detection signal is sampled, and the predetermined time is passed. An average calculation means for calculating an average of a plurality of levels sampled within; if the difference between the average calculated by the average calculation means and the zero level is more than a predetermined value, the zero level is calculated based on the average. Modify the zero level adjusting means; and said signal output means for outputting a signal corresponding to the difference between the level and the zero level of the detection signal; vehicle on gyro drift canceling apparatus comprising a. 2. The drift canceling device for a gyro on a vehicle according to claim 1, wherein the average calculating means invalidates the calculation result when the number of samplings within the predetermined time is less than a predetermined number. 3. A gyro on a vehicle for outputting a detection signal containing a DC component and an AC component; an amplifying means for inputting and amplifying the detection signal to generate an output signal; a base for outputting a DC level base signal. -Base signal generating means; combining means for adding or subtracting the base signal to or from the input of the amplifying means; reference is made to the level of the detection signal and the level of the base signal, and the difference between them is predetermined. First discriminating means for discriminating whether or not the difference is larger than a first threshold value; when the difference is larger than the first threshold value for a first time or more, the detection signal within the time First average calculating means for calculating an average level; base level changing means for changing the level of the base signal output by the base signal generating means according to the average level; Correspondingly, at low speed it is big at high speed Threshold setting means for setting a second threshold; referring to the level of the output signal of the amplifying means, the difference between the level of the signal and a predetermined zero level of the signal is the second threshold. Second discriminating means for discriminating whether or not the value is within a value; the level of the output signal is repeatedly referred to within a second time, and the difference between the level of the output signal and a predetermined zero level of the signal is the second level. Second average calculation means for sampling the level of the detection signal only when it is within a threshold value and calculating an average of a plurality of levels sampled within a second time; calculated by the second average calculation means. And a signal corresponding to the difference between the level of the output signal and the zero level when the difference between the average and the zero level is more than a predetermined value, the zero level is corrected based on the average. Signal output means for outputting Drift cancellation apparatus for a vehicle on the gyro.