KR100450994B1 - Compenastion method of nonlinear thermal bias drift of vibratory gyroscope by using fuzzy logic - Google Patents

Abstract

The present invention relates to a temperature bias drift correction method of a vibration type gyroscope using a fuzzy logic to correct the nonlinear bias drift caused by the temperature change.

The present invention includes a first step of quantifying a degree of temperature change generated by a temperature chamber during a predetermined time by detecting a temperature change detection value from a measured temperature change waveform by a chamber and arranging it to an arithmetic value; A second step of quantifying the bias-triggered output value for the temperature change detected from the output waveform of the pre-correction vibration type gyroscope biased for the same time by being influenced by the degree of temperature change during the predetermined time; A third step of deriving an ideal fuzzy logic of the unquantified dispersion values shown in the output waveforms of the pre-correction vibration type gyroscope by the linear least-squares method; And a fourth step of subtracting the values calculated by substituting the values quantified in the first step into the fuzzy logic derived in the third step from the values quantified in the second step to find a corrected output value.

According to the present invention, the bias drift of the vibration-type gyroscope according to the temperature change can be simply, easily and appropriately corrected in a short time, and thus its utility is doubled.

Description

COMPENSATION METHOD OF NONLINEAR THERMAL BIAS DRIFT OF VIBRATORY GYROSCOPE BY USING FUZZY LOGIC}

The present invention is a vibration type gyroscope that is widely used in self-navigation, automatic flight, camera stabilization device and mobile robot, rehabilitation medical equipment, tunnel construction, high-rise building dustproofing, such as small aircraft, unmanned aerial vehicle, satellite, mobile vehicle, etc. In particular, the present invention relates to a temperature bias drift correction method of a vibrating gyroscope using a fuzzy logic to compensate for nonlinear bias drift caused by temperature change.

In general, a gyroscope (GYROSCOPE) is a kind of angular velocity measuring sensor used in navigation devices and inertial guidance devices, such as aircraft, ships and automobiles, and the like (SPINNING WHEEL); Optical (OPTICAL); Semiconductor type (MEMS); It is roughly classified as a vibration type (resonant type) gyroscope.

Rotating gyroscopes, the most widely used classic gyroscopes, are also known as rotary gyroscopes, and are supported by double or triple gimbals of disc-shaped metal plates that rotate at high speed with round wheels. It can be configured to rotate freely or freely so that the angular velocity can be measured. Since it is based on a mechanical structure, it is expensive because it is bulky and must maintain high precision, and has a short life cycle due to abrasion and loss, and is also easily damaged by shock and vibration.

As shown in FIG. 1, an optical gyroscope exhibits different phases when light interfering in an annular interferometer rotates when the interferometer ring rotates, and the phase difference is proportional to the rotational angular velocity. It is an angular velocity measuring device using the neck (Sagnac) effect. Gyroscopes using this principle include ring lasers and fiber-optic gyroscopes. These gyroscopes have proven excellent performance but have an expensive disadvantage.

Semiconductor type gyroscopes are manufactured using semiconductor technology. Because of the characteristics of semiconductor processing technology, it is difficult to produce a rotating object, and thus a sensor having a vibrating shape is manufactured. That is, semiconductor gyroscopes are classified into bulk type vibration gyroscopes fabricating silicon workpieces and MEMS (Micro Electro Mechanical System) gyroscopes using surface micro machning technology, which is a semiconductor processing technology.

Such a semiconductor gyroscope has a limitation in its performance, so it is suitable for home appliances and electronics, but has a limitation that it cannot be used for an unmanned aerial vehicle and a marine navigation system.

On the other hand, in the case of a vibrating gyroscope, high sensitivity, low cost, and miniaturization are possible, which solves the above-mentioned disadvantages of the rotary, optical, and semiconductor type gyroscopes. From the research results to measure the angular velocity by processing the quartz in the form of tuning fork and detecting the amount of deformation caused by the Coriolis force, it is continuously proportional to the rotational angular velocity while resonating the cylinder-type structure using a piezoelectric element or electromagnetic force. Therefore, the angular velocity can be measured by measuring the characteristic in which the amount of resonance mode change is proportional to the Coriolis force.

By the way, the vibratory gyroscope has acquired the possibility of high sensitivity, low cost, and miniaturization, but the nonlinear bias drift phenomenon is caused by the unavoidable change of the cylinder's mechanical characteristics due to manufacturing tolerances, material non-homogeneity, and temperature change. The disadvantage was that it caused a significant decrease in the reliability of the measurement.

2 is a graph illustrating such a phenomenon as an example, and it can be seen that a significant nonlinear bias drift phenomenon is caused especially by a temperature change.

Conventionally, a lookup table method or a polynomial curve fitting method is used to correct this phenomenon. The lookup table method stores an error due to experimental data in a memory. In the table state, correction is possible by subtracting the error value due to the operating temperature from the measured value of the vibrating gyroscope, that is, stored in the memory and the tabled value.In this case, the range of the operating temperature is enormous In order to store all the error values of the experimental data corresponding to each of them in the memory, an enormously large amount of memory is required, and its practical value is denied.

Polynominal curve fitting method is a kind of data analysis method, HYPER GRAPH, which is the closest higher-order polynomial that minimizes the error by using powerful engine including more than 100 mathematical functions and operators according to the distribution of experimental data. After making the waveform of the vibration type gyroscope, the waveform of the vibration type gyroscope is compared with that of the higher-order polynomial to compensate for the difference, and the sharpness of the accuracy is increased. For example, in the graph of FIG. 3, the gyroscope output value of FIG. Examples of corrections made under the polynominal curve fitting method are shown.

However, even in the case of the polynominal curve fitting method, a higher order polynomial that is suitable for the distribution of experimental data, which is a criterion for measurement, is required, and an excessive calculation time for finding the higher order polynomial is excessively overlooked. I can't.

The present invention was created in view of the above-described problems of the prior art, and solves this problem, using a fuzzy logic, a high-order polynomial for improving the accuracy or accuracy of a large-capacity memory, and a temperature without delay of calculating a secondary time. An object of the present invention is to provide a temperature bias drift correction method of a vibration type gyroscope using a fuzzy logic to easily and efficiently correct nonlinear bias drift of a vibration type gyroscope according to a change.

1 is a schematic block diagram showing an example of a conventional gyroscope.

Figure 2 is a deformation graph of the output value according to the temperature change of the conventional vibration type gyroscope.

Figure 3 is an output graph showing the correction value of the vibration type gyroscope using the prior art.

Figure 4 is a measurement conceptual diagram showing the configuration of a vibrating gyroscope for explaining the present invention.

5 is a block diagram illustrating a control method according to the present invention;

Figure 6 is a graph showing the experimental data and the extraction fuzzy logic of the temperature bias drift for explaining the present invention.

Figure 7 is a graph showing the output value of the vibration type gyroscope corrected in accordance with the method of the present invention.

<Description of Symbols for Main Parts of Drawings>

10: chamber 12: temperature change waveform by the chamber

20: (vibration type) gyroscope 22: output waveform of vibration type gyroscope before correction

40: fuzzy logic 50: corrected output waveform

The above object of the present invention can improve the performance of the vibration type gyroscope by correcting the performance deterioration factor of the vibration type gyroscope by the temperature bias drift using the fuzzy logic.

That is, in order to achieve the above object, the present invention provides a first process of quantifying a temperature change detection value from a measured temperature change waveform by a measured chamber and quantifying the degree of temperature change generated by a temperature chamber for a predetermined time. and; A second step of quantifying the bias-triggered output value for the temperature change detected from the output waveform of the pre-correction vibration type gyroscope biased for the same time by being influenced by the degree of temperature change during the predetermined time; A third step of deriving an ideal fuzzy logic of the unquantified dispersion values shown in the output waveforms of the pre-correction vibration type gyroscope by the linear least-squares method; And a fourth process of finding the corrected output value by subtracting the values calculated by substituting the values quantified in the first process into the fuzzy logic derived in the third process from the values quantified in the second process. A temperature bias drift correction method of a vibrating gyroscope using a fuzzy logic is provided.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

Figure 4 is a schematic diagram showing the operating principle of the vibration type gyroscope for explaining the present invention.

First, referring to FIG. 4, the basic operation principle of the vibratory gyroscope is as follows.

If the cylinder is assumed to be perfectly uniform and symmetrical and excited at its natural frequency, the first mode will occur in the central axis where the excitation force is applied, and the node point will occur in the 45 ° direction of the biaxial axis.

The node point is no change in the node point if no external input, that is, the angular velocity is applied. That is, the displacement and velocity are zero.

When an input angular velocity is applied to this excitation mode, a second mode occurs at the node point. That is, a physical change occurs at the node point.

At this time, the first mode (DRIVE MODE) (D), the second mode is called the sensing mode (RESPONSE MODE (R) and the axis of each mode is the excitation axis (q _d ) and the sensing axis (q _r ), And sensing uses electromagnetic force.

At this time, the phase-locked control loop is used to excite the amplitude and the natural frequency, and the mechanical structure in which the excitation forces act in opposition to each other is applied to increase the excitation force. ).

If no angular velocity is applied to the gyroscope from the outside, the maximum displacement occurs on the axis where the excitation force is applied, and when the angular velocity is applied, the Coriolis force as shown in Equation 1 is applied to the excitation mode (D) and the sensing mode (R). This will occur.

(Equation 1)

F = 2mΩ × v

Where m is the mass of the object, Ω is the input angular velocity, v is the relative velocity of the object with respect to the rotational coordinate system, and x is the product of the vector.

Therefore, since the magnitude of the Coriolis force is proportional to the angular velocity, the angular velocity can be measured.

That is, the angular velocity can be calculated from the equation of motion of the ideal gyro as shown in Equation 2 below.

(Equation 2)

Where ω is the natural frequency, Wow Is the damping ratio of the two modes q _d and q _r , G is the effective mass affecting both modes by Coriolis force, Ω is the input angular velocity, and F is the excitation force.

On the other hand, in addition to the method of measuring the angular velocity by measuring the magnitude of the vibration of the sensing mode (R), a closed loop controller for canceling the physical change of the node (node) to increase the bandwidth and linearity of the gyroscope You can also use

When a gyroscopic angular velocity is applied and a Coriolis force is generated, a physical change occurs at the nodal point.At this time, the angular velocity is measured by measuring the magnitude of the electromagnetic force, which is a control action to offset the physical change occurring at the nodal point. Can be measured.

In the present invention, a complex nonlinearity is extracted after extracting a linear error model for several temperature intervals using a linear least squares method using a gyroscope measuring the angular velocity by operating according to the above-described operating principle. Each linear error for all operating temperatures through fuzzy logic (FUZZY LOGIC), which divides the fuzzy set in the system and configures the rule base according to each area to obtain better performance than the existing nonlinear controller. The organic mixing of the model results in a simple but accurate fuzzy logic to compensate for complex nonlinear bias drift over the entire operating temperature range.

For example, the linear least-squares method can draw various functions according to the point of view when the experimental data are distributed, and the best method is to find a line that is closest to the data by applying a statistical method. For a set of n experimental values (x1, y1), (x2, y2), ........ (xn, yn), a linear regression by means of least squares If you draw a straight line and get n deviations d1, d2, ...... dn (d = difference between the experimental data value and the linear value), the optimal straight line through n data points is The linear least squares method is called the linear least squares where the sum of the squares of the deviations is the minimum and thus the polynomial with the least error is optimized.

In addition, fuzzy control is modeling 'manipulation method performed by human beings that is flexible and excellent adaptability with ambiguous information' as control rule and realizes such control in calculator using fuzzy inference. As the knowledge accumulated in the process of operation and the characteristics of the process are accumulated, their knowledge is stored in the form of language, and IF- which concludes the process when the condition is satisfied is the conditional proposition. It is easily expressed in THEN form (if ~).

In this control rule, it is a feature of fuzzy control that the condition judgment criteria of the process and the contents of the operation are processed as ambiguous contents and quantified by a membership function. For example, the knowledge of reducing fuel when the temperature starts to rise is expressed by the control rule as follows.

IF TEMP = ML THEN FUEL = NS

TEMP is the temperature, FUEL is the variable representing the fuel, ML is a slightly increased, NS is a fuzzy variable means a slight decrease, the content is quantified by the membership function.

That is, the IF part is composed of a condition (condition) and the THEN part (action), each rule is executed when the IF part condition is true, the operation of THEN is executed.

A control rule indicating a control strategy of the process is prepared, and when an input amount is given, an operation output is determined by performing a numerical operation called fuzzy inference based on the control rule.

At this time, rather than selecting the operation method that most closely matches the first half proposition, the master operator calculates the total amount of operation by superimposing the grade of the first half proposition of each control rule. The method is the same as the method for determining the operation amount.

As such, fuzzy control actively uses ambiguous information and collects a lot of information accordingly, and uses the knowledge that expresses the control of the multi-input small output that determines the amount of manipulation. Thus, it is possible to construct a control system that is easily familiar to the operator.

Therefore, when applied to the present invention, as shown in Figure 5 as an embodiment, the degree of temperature change generated inside the chamber 10 appears as a temperature change waveform 12 by the chamber, which is not shown By means of a temperature change detector, the temperature change detection value 14 is quantified in real time to an arithmetic value.

On the other hand, the temperature change inside the chamber 12 affects the vibratory gyroscope 20 to cause a nonlinear bias drift, which is represented by the output waveform 22 of the pre-correction vibratory gyroscope as shown in the graph.

The output waveform 22 of the vibration-type gyroscope before correction is in a state in which the measured value is distorted considerably due to the temperature change of the chamber 10. As shown in the graph, the detection values are variously distributed and distributed. It can be seen as a value.

Therefore, if there are the most ideal function values that these dispersion values follow, such as the maximum value and the minimum value, the value that zeroes the deviation is found and the values and the temperature change detection values 14 inside the chamber 10 are found. By calculating the difference from the values derived from the correlation with the ideal function, it is possible to infer the measured value, that is, the correction value when the vibration type gyroscope 20 is not affected by the temperature change of the chamber 10. will be.

To this end, the fuzzy logic 40, which is an ideal function value they follow, from the variance values can be found by linear least squares method, which has already been described above.

In summary, first, the degree of temperature change generated by the temperature chamber 10 for a predetermined time is determined by detecting the temperature change detection 14 value from the temperature change waveform 12 by the measured chamber and quantifying it to an arithmetic value. You will do step 1.

Subsequently, the bias-trimmed output value y for the temperature change detected from the output waveform 22 of the pre-correction type vibration gyroscope 20 biased drift for the same time by being influenced by the degree of temperature change during the predetermined time. A second process of quantification is performed.

Thereafter, a third process of deriving the ideal fuzzy logic 40 following the unquantified dispersion values shown in the output waveform 22 of the pre-correction type vibration gyroscope 20 through the linear least-squares method is performed. .

When the first, second, and third processes described above are performed, values calculated by substituting the values quantified in the first process into the fuzzy logic 40 derived in the third process are then quantified in the second process. Corrected output value subtracted from By carrying out the fourth process of finding), the desired correction value can be calculated.

In further detail, the bias drift of the gyroscope 20 caused by the temperature change of the chamber 10 is obtained by using the linear least squares method of FIG. The same ideal fuzzy logic is derived, and each linear error model is then organically mixed with the fuzzy logic according to the temperature deviation, resulting in a simple but accurate fuzzy logic to compensate for complex nonlinear bias drift over the entire operating temperature range. It becomes possible.

7 is a graph illustrating a corrected output value output after correcting a bias drift of the vibratory gyroscope 20 using the purge logic of FIG. 6 according to the present invention described above.

As described in detail above, according to the present invention, the bias drift of the vibratory gyroscope according to the temperature change can be simply, easily and appropriately corrected in a short time, and thus its utility is doubled.

Claims (1)

A first step of detecting the temperature change detection value 14 from the temperature change waveform 12 by the measured chamber and quantifying the degree of temperature change for a predetermined time generated by the temperature chamber 10 into an arithmetic value; Quantified by the bias-trimmed output value (y) of the temperature change detected from the output waveform 22 of the pre-correction vibration type gyroscope 20 biased drift for the same time by being influenced by the degree of temperature change during the predetermined time. A second process of doing; A third step of deriving an ideal fuzzy logic 40 following the unquantified dispersion values shown in the output waveforms 22 of the oscillating gyroscope 20 before correction through a linear least-squares method; The output value corrected by subtracting the values calculated by substituting the values quantified in the first process into the fuzzy logic 40 derived in the third process from the values quantified in the second process ( The temperature bias drift correction method of a vibrating gyroscope using a fuzzy logic, characterized in that it comprises a fourth step of finding a).