RU2598155C1 - Method for compensation of systematic components of drift of gyroscopic sensors - Google Patents

Abstract

FIELD: instrument making.

SUBSTANCE: invention relates to instrument-making and can be used in designing single- and three-axes meters of movement parameters - angular velocities and linear accelerations for inertial navigation systems and flight control systems for mobile objects. Disclosed is a method for compensation of temperature dependence of systematic components of drift of gyroscopic sensors, including measurement in factory conditions in the process of debugging sensitive elements of values of systematic components in the form of zero signals and scaling factors at fixed values of a row of temperatures within the operating range, description of piecewise linear or polynomial approximation of dependence of zero signal and scaling factor from the temperature. Herewith the measurement of systematic components in the form of zero signals and scaling factors at fixed values of the row of temperatures within the operating range is carried out during at least two runs of the sensitive elements. Calculated are average values of zero signals and scaling factors at fixed values of the row of temperatures within the operating range obtained during the runs. Obtained average values are used to determine coefficients of piecewise linear or polynomial approximation of temperature dependence. Then those coefficients are recorded into a microcontroller for the possibility of algorithmic compensation of the temperature dependence of zero signals and scaling factors during operation.

EFFECT: technical result is the increase of accuracy characteristics of gyroscopic sensors.

1 cl, 3 dwg

Description

The invention relates to the field of instrumentation and can be used to build uniaxial and triaxial measuring instruments for motion parameters — angular velocities and linear accelerations for inertial navigation systems and aerobatic control systems for moving objects.

One of the main sources of measurement error of gyroscopic sensors - gyroscopes and accelerometers is the temperature dependence of the systematic components of the drift - zero signals and scale factors.

There are two ways to eliminate this dependence - temperature stabilization and algorithmic compensation [1].

Temperature stabilization requires additional energy consumption, increasing the size and weight of the motion parameters meter. However, this method does not eliminate the instability of taxonomy arising in the sensors of angular velocities and linear accelerations from start to start [2].

Less energy-consuming and with great opportunities for improvement is the method of algorithmic compensation of the temperature dependence of the systematic components of the drift of gyroscopic sensors [3], which is a prototype of the claimed invention. In this method, in the factory, during the debugging of the gyroscope, the systematic components are determined in the form of zero signals and scale factors at fixed values of a number of temperatures in the operating range. The dependence of the zero signal and the scale factor on temperature is approximated by a polynomial:

- for a zero signal:

- for scale factor:

where: B ₀ , B ₁ , B ₂ , B ₃ are the coefficients of a polynomial approximating the temperature dependence of the zero signal U ₀ (T); S ₁ , S ₂ , S ₃ - coefficients of a polynomial approximating the temperature dependence of the scale factor SF _K (T); T is the current temperature.

The coefficients of polynomials describing the temperature dependence of the zero signal and the scale factor are recorded in the microcontroller of the sensing element. During operation of the device for the current temperature values according to the algorithms (1), (2), the values of zero signals and scale factors are calculated. The calculation of the measured angular velocity Ω taking into account the dependence of the zero signal U ₀ (T) and the scale factor SF _K (T) on temperature is carried out in the microcontroller of the sensing element according to the formula:

where: U (Ω) is the output signal of the gyroscope in analog form.

The main disadvantage of the algorithmic compensation implemented in the prototype is the lack of consideration of the instability of the systematic components of the gyro drift from start to start. As the results of experimental studies show, the instability of systematic components from start-up to start-up is significant and can exceed the instability of systematic components in a start-up.

The claimed invention solves the problem of reducing the magnitude, instability of systematic components, while achieving such a technical result as improving the accuracy characteristics of gyroscopic sensors.

The claimed technical result is achieved by a method of compensating for the temperature dependence of the systematic components of the drift of gyroscopic sensors, including measurement at the factory, during the debugging of sensitive elements, the values of the systematic components in the form of zero signals and scale factors at fixed values of a number of temperatures in the operating range, a piecewise linear or polynomial approximation of the dependence of the zero signal and the scale factor on temperature, at m the measurement of systematic components in the form of zero signals and scale factors at fixed values of a number of temperatures in the operating range is carried out during at least two starts of sensitive elements, calculate the average values of zero signals and scale factors at fixed values of a number of temperatures in the working range obtained in starts , the obtained average values determine the coefficients of a piecewise linear or polynomial approximation of the temperature dependence, then these coefficients are recorded in the microcontroller for the possibility of algorithmic compensation of the temperature dependence of zero signals and scale factors during operation.

During the start-up of the sensing element, the value of the zero signal at the start temperature is determined, the difference between the measured value of the zero signal and its value calculated by the algorithmic compensation coefficients is calculated, and subsequently, when the temperature changes, the calculated values of the zero signal are corrected according to the algorithmic compensation coefficients, taking into account differences between the measured and calculated values of the zero signal at the start.

The problem is solved by taking into account the instability of the systematic components of the drift of gyroscopic sensors from start to start as follows.

In the factory, during the debugging of sensitive elements, the systematic components are measured in the form of zero signals U _oi (T _b ) and scale factors k _mi (T _b ) at fixed values of a number of temperatures T _b in the operating range in the process of multiple (n), at least of two starts of sensitive elements, the average values of zero signals are calculated

and scale factors

and they determine the coefficients of the piecewise linear k _T (q _j ), k _tM (q _j ) or polynomial k ₀ , k ₁ , k ₂ ... k _m ; B ₀ , B ₁ , B ₂ ... B _m approximations of their temperature dependence, and these coefficients are introduced into the microcontroller of sensitive elements, with the help of which the algorithmic compensation of the temperature dependence is carried out during operation:

m-th polynomial:

zero signals

scale factors

or piecewise linear approximation:

zero signals

scale factors

where: B ₀ , B ₁ , B ₂ ... B _m are the coefficients of a polynomial approximating the temperature dependence of the zero signal; T _b - the basic temperature values (T ₁ , T ₂ , T ₃ , T ₄ and so on) at which the measurement of zero signals and scale factors in the factory; T - current operating temperature values at which the values of zero signals and scale factors in operating conditions are determined; k ₀ , k ₁ , k ₂ ... k _m are the coefficients of a polynomial approximating the temperature dependence of the scale factors; k _To (q _j ) and k _Tm (q _j ) are the coefficients of the linear temperature approximation of the zero signals and scale factors in the sections q between their basic values:

where: U ₀ (T _b ), U ₀ (T _{b + 1} ), k _m (T _b ), k _m (T _{b + 1} ) - zero signals and scale factors at base temperatures at the beginning (T _b ) and c the end (T _{b + 1} ) of the linear section q _j , aj is the number of the interval.

In addition, in the process of starting the sensing element, the value of the zero signal at the start temperature is determined, the difference ΔU _{op is} calculated between the measured value of the zero signal U ₀ (T _p ) and its value calculated by the algorithmic compensation coefficients U _op (T _p ) according to formulas 6, 8.

and further, when the temperature changes, the calculated values of the zero signal are adjusted according to the algorithmic compensation coefficients, taking into account the difference between the measured and calculated values of the zero signal at the start:

when approximated by an m-th polynomial:

with piecewise linear approximation:

The invention is illustrated by graphs:

FIG. 1 is a graph of the temperature dependence of the averaged values of zero signals U _o (T) approximated by an m-th polynomial;

FIG. 2 is a graph of the temperature dependence of the averaged values of the scale factor k _m (T) with piecewise linear approximation;

FIG. 3 is a graph for determining the correction value of the zero signal ΔU _op at startup and its accounting during operation at operating temperatures.

The invention is implemented as follows.

In the process of calibrating the device in the factory, the zero signals U _oi (T _b ) of FIG. 1 and scale factors k _mi (T _b ) of FIG. 2 for a number of fixed values of T ₁ , T ₂ , T ₃ , T ₄ , T ₅ , T ₆ , T _{7 of the} base temperature T _b in several starts (in Fig. 1, 2, as an example, for four starts). By formulas 4, 5 are the average values of zero signals U ₀ (T _b ) and scale factors k _m (T _b ) for fixed values of the base temperatures. Using the data obtained, the temperature dependence of zero signals and scale factors can be described either by an m-th polynomial in formulas 6, 7, or using a piecewise linear approximation in formulas 8, 9.

When approximating the temperature dependence by a polynomial, the coefficients of the polynomial В ₀ , В ₁ , В ₂ , В ₃ ... B _{m are} recorded in the memory of the microcontroller of the sensing element for the algorithmic compensation of the zero signal and the coefficients k ₀ , k ₁ , k ₂ , k ₃ ... k _m - for algorithmic: compensation of the scale factor during operation of gyroscopic sensors.

With a piecewise linear approximation, the values of the base temperatures (T _b ) and zero signals U _o (T _b ) of scale factors k _m (T _b ) calculated according to formulas 4 and 5 at these base temperatures are recorded in the memory of the microcontroller. In addition, the coefficients of the temperature dependence of the zero signal k _To (q _j ) and the scale factor k _Tm (q _j ) (Fig. 2) in the linear sections of q between the base temperatures calculated in the factory according to formulas 10 are also recorded in the microprocessor's memory.

During operation of gyroscopic sensors, the algorithmic compensation of the temperature dependence of the scale factors is carried out according to algorithms 7 or 9, and zero signals - according to algorithms 6 or 8, depending on the adopted method of approximating the temperature dependence.

For additional compensation of the zero signal during operation of the gyroscopic sensors at the start, the difference ΔU _op between the measured value of the zero signal U _o (T _p ) and its value calculated by the algorithmic compensation coefficients U _op (T _p ) is determined according to formulas 6, 8 - FIG. . 3. The specified correction ΔU _op is determined again with each new start of the gyroscopic sensor.

After starting during operation, the calculated correction is taken into account during the algorithmic compensation of the temperature dependence of the zero signal using formulas 12, 13.

Literature

1. Nekrasov Ya.A., Micromechanical gyroscope, RU 2535248 C1.

2. Efremov M.V., Gubanov A.G., Romanov A.V., Karpov M.N. Kruglov S.A. Fiber optic gyroscope with temperature compensated digital output, RU 2448325.

3. V. Logozinsky, I. Saftulin, V. Solomatin. Fiber-optic rotation sensor with digital corrected output / VI St. Petersburg International Conference on Integrated Navigation Systems, May 28-30, 2001

Claims (2)

1. A method of compensating the temperature dependence of the systematic components of the drift of gyroscopic sensors, including measurement at the factory, during the debugging of sensitive elements, the values of the systematic components in the form of zero signals and scale factors at fixed values of a number of temperatures in the operating range, a description of a piecewise linear or polynomial approximation the dependence of the zero signal and the scale factor on temperature, characterized in that the measurement of systematic which are in the form of zero signals and scale factors at fixed values of a number of temperatures in the operating range are carried out in the process of at least two starts of sensitive elements, calculate the average values of zero signals and scale factors at fixed values of a number of temperatures in the working range obtained in these starts, the obtained average values determine the coefficients of a piecewise linear or polynomial approximation of the temperature dependence, then these coefficients are written into the microcontroller for the possibility of algorithmic compensation of the temperature dependence of zero signals and scale factors during operation. 2. The method according to 1, characterized in that in the process of starting the sensing element, the value of the zero signal at the start temperature is determined, the difference between the measured value of the zero signal and its value calculated by the algorithmic compensation coefficients is calculated, and then, when the temperature changes, an adjustment is made calculated values of the zero signal according to the coefficients of algorithmic compensation, taking into account the difference between the measured and calculated values of the zero signal at the start.

Images