WO2016116360A1

WO2016116360A1 - Method for calculating an orientation using a sensor system, and sensor system

Info

Publication number: WO2016116360A1
Application number: PCT/EP2016/050727
Authority: WO
Inventors: Tobias RANKL; Rainer Dorsch; Timo Giesselmann; Gerhard Lammel
Original assignee: Robert Bosch Gmbh
Priority date: 2015-01-21
Filing date: 2016-01-15
Publication date: 2016-07-28
Also published as: DE102015200944A1

Abstract

Disclosed is a method for calculating an orientation using a sensor system which has a micro-mechanical yaw-rate sensor and a calculation unit, wherein a yaw rate is measured using the yaw-rate sensor and the yaw rate is integrated over time in the calculation unit in order to calculate the orientation. A micro-mechanical oscillator of the yaw-rate sensor oscillates at a natural frequency and generates a natural frequency signal, a clock signal for the integration in the calculation unit being derived from the natural frequency signal.

Description

Description title

Method for calculating an orientation with a sensor system and sensor system

State of the art

The invention is based on a method for calculating an orientation with a sensor system according to the preamble of claim 1 and on a sensor system according to the preamble of claim 9.

In the prior art, such methods are used, for example, in inertial navigation systems, with which the orientation of objects and / or persons freely movable in space can be determined. Part of the orientation determination is to calculate angles with respect to given spatial axes. This is a rotation rate - ie a

Angular velocity - z. B. measured with a micromechanical rotation rate sensor and this rotation rate is then integrated over time to obtain the angle. The temporal integration of the rotation rate requires a time reference, which is typically provided as a clock signal. In order to obtain the most accurate time reference, clock signals are derived in the known from the prior art sensor systems of a quartz oscillator.

In such methods, there is the disadvantage that due to the additionally required quartz oscillator results in a complex structure, thereby increasing the cost of the sensor system. It is therefore an object of the present invention, a cost-effective

Calculation of orientation or to enable a low-cost sensor system.

Disclosure of the invention

The inventive method for calculating an orientation with a sensor system and the sensor system according to the invention with the features of the independent claims have the advantage over the prior art that it is not necessary to provide a quartz oscillator for generating a clock signal. Rather, a natural frequency signal is generated with a micromechanical oscillator which is provided anyway in the rotation rate sensor and which oscillates at a natural frequency. From the

Natural frequency signal of the micromechanical oscillator is derived a clock signal for integrating in the calculation unit. By dispensing with a quartz oscillator, a cost-effective calculation of the orientation or a cost-effective implementation of the sensor system is made possible.

The orientation can be calculated, for example, as an angle with respect to a given axis in space.

Advantageous embodiments and developments of the invention are the dependent claims, as well as the description with reference to the

Drawings removable.

According to a preferred embodiment, it is provided that for the derivation of the clock signal, the natural frequency signal is fed to a frequency multiplier, wherein the frequency multiplier the natural frequency of

micromechanical oscillator multiplied by a balance factor. With the aid of the frequency multiplier manufacturing tolerances can be compensated by the frequency of the clock signal is set by multiplication with an adjustable adjustment factor to a predetermined target frequency. Of the

Adjustment factor may take an integer value or a fractional value, so that it is possible with the frequency multiplier to generate a clock signal having a clock frequency which is a full or is a fractional multiple of the natural frequency or which is a quotient of the natural frequency. Particularly preferably, the frequency multiplier is designed as a digital frequency multiplier, so that it can be implemented in a digital electronic circuit.

In this context, an embodiment is advantageous in which the adjustment factor is determined, wherein the natural frequency of the micromechanical oscillator is measured and the measured natural frequency is compared with a predetermined desired frequency. The measurement of the natural frequency and the comparison of the measured natural frequency with the given

Target frequency can be carried out within the framework of the production of the sensor system and / or as part of a calibration process. It is advantageous if the nominal frequency is generated by a quartz oscillator. The adjustment factor is preferably stored in a calibration register of the sensor system.

A further preferred embodiment provides that the

Calculation unit is fed a correction factor to a

Compensate quantization error of the frequency multiplier. Such an embodiment is advantageous for compensating for a deviation of the clock frequency of the clock signal derived from the frequency multiplier from the predetermined setpoint frequency, which is due to the finite resolution of the adjustment factor-that is, the quantization error of the

Frequency multiplier - conditional. By the compensation of the

Quantization error with the correction factor, the accuracy of the frequency multiplier can be reduced, i. the quantization error can be increased without deteriorating the accuracy of the angular orientation.

An embodiment is preferred, wherein the correction factor corresponds to the ratio of the clock frequency of the clock signal to the nominal frequency. The quotient of the actual frequency of the clock signal and the nominal frequency indicates the quantization error of the frequency multiplier.

In this context, an embodiment is advantageous in which the clock frequency of the clock signal is measured to determine the correction factor becomes. The measurement can, for example, in the context of the production of the

Sensor system and / or carried out as part of a calibration process. Alternatively, it is possible to measure the natural frequency of the natural frequency signal and to calculate the clock frequency based on the adjusted adjustment factor.

A further preferred embodiment provides that the correction factor is stored in a correction register of the sensor system, so that the correction factor for calculating the orientation can be retrieved by the calculation unit. The correction register is particularly preferably designed as a non-volatile correction register, so that the determined correction factor is maintained even when the power supply of the sensor system is turned off.

Also advantageous is an embodiment in which the calculation unit for calculating the orientation is detected with an acceleration sensor detected acceleration values and / or with a magnetic field sensor

Magnetic field values are supplied. The acceleration values and / or magnetic field values can be taken into account in the calculation of the orientation, in particular the calculation of an angle. It is particularly advantageous if a Kalman filter is used to calculate the orientation, to which the rotation rate, an acceleration value and a magnetic field value determined with the yaw rate sensor are preferably supplied as input variables.

According to a preferred embodiment of the invention

Sensor system is provided that the sensor system a

Frequency multiplier, which is designed such that the

Natural frequency of the micromechanical oscillator can be multiplied by a predetermined adjustment factor and that the sensor system

Correction register in which a correction factor is stored, which can be supplied to the calculation unit to correct a quantization error of the frequency multiplier. Such a development of the sensor system has the advantage that the accuracy of the time base for the calculation of the orientation is increased. Embodiments of the present invention are illustrated in the drawings and explained in more detail in the following description.

Brief description of the drawings

FIG. 1 shows a block diagram of a sensor system according to FIG

exemplary embodiment of the present invention.

FIG. 2 shows various measured and predetermined frequencies for illustrating the processes during a calibration process.

FIG. 3 shows two examples of calculated time profiles of FIG

Orientation according to the method of the invention.

Embodiments of the invention

In the different figures, the same parts are always the same

Reference numerals provided and are therefore usually named or mentioned only once in each case.

1 is a block diagram of an embodiment of a

Sensor system 1 according to the invention shown, which can be used in an inertial navigation system to determine the orientation of an object and / or a person in the room. The sensor system 1 has a first micromechanical rotation rate sensor 2, via which a rotation rate ω-that is to say an angular velocity about a predetermined first axis-is measured. The first rotation rate sensor 2 has at least one

Micro-mechanical oscillator, which is excited to a vibration with a natural frequency. The micromechanical oscillator can

For example, be designed as a seismic mass, which is suspended movably. In addition to the first rotation rate sensor 2 are in the

Sensor system 1 is provided two further rotation rate sensors, which measure rotation rates about two axes arranged perpendicular to the first axis. The two further rotation rate sensors are not shown in FIG. 1 for reasons of clarity. To calculate the orientation of the sensor system 1 in space, angles φ 'relative to the three mutually perpendicular axes are calculated from the measured rotation rates ω of the three rotation rate sensors 2. The calculation of the angles φ 'takes place in a calculation unit 6 of FIG

Sensor system 1, which is arranged in a common housing with the first rotation rate sensor 2 and preferably with the two other rotation rate sensors. In addition, in the common housing, one or more acceleration sensors for measuring acceleration data and / or one or more magnetic field sensors for measuring

Magnetic field data to be arranged.

In the calculation unit 6, the rotation rate ω measured with the first rotation rate sensor 2 is integrated over an integration time to obtain a first angle φ '. The integration of the rate of rotation co over time requires one

Time reference, which is provided in the sensor system 1 as a clock signal 13 available. To derive the clock signal 13, a natural frequency signal 11 is generated by the rotation rate sensor 2, which has the natural frequency 21 of the micromechanical oscillator of the rotation rate sensor 2. The

Natural frequency signal 11 is supplied to a digital frequency multiplier 4.

The frequency multiplier 4 multiplies the natural frequency 21 of the

Natural frequency signal 11 with a balance factor 10, which is given as the quotient n / m of two integer values n, m. The adjustment factor 10 is in the form of the integer values n, m in a trim register 3 of the

Sensor system 1 deposited. If n / m> 1, the clock frequency is 23 of the

Clock signal 13 greater than the natural frequency 21 of the micromechanical oscillator. If n / m <1, then the clock frequency 23 of the clock signal 13 is smaller than the natural frequency 21. With the frequency multiplier 4, the clock frequency 23 of the clock signal 13 can be adjusted to a predetermined nominal frequency. Since the

Adjustment factor 10 has a finite resolution, but the clock frequency 23 of the clock signal 13, as a rule, a deviation 24 from the target frequency, which represents the quantization error of the frequency multiplier 4. To correct the quantization error is in a correction register 5 of the Sensor system 1 a correction factor fcorr deposited, which is the

Calculation unit 6 is supplied and which in integrating the

Rate of rotation ω is taken into account, as will be explained in more detail below. The correction factor fcorr corresponds to the ratio of the clock frequency 23 of the clock signal 13 to the setpoint frequency 22.

In addition, acceleration values 15 detected by an acceleration sensor and magnetic field values 16 detected by a magnetic field sensor are supplied to the calculation unit 6 for calculating the orientation.

Based on the illustration in Figure 2, the operations for

Adjustment of the clock frequency 23 of the clock signal 13 will be explained. The adjustment takes place in the sensor system 1 as part of a calibration process, which, for example, immediately after the production of the sensor system 1 or as needed before the installation of the sensor system 1 in an inertial

Navigation system is performed.

In FIG. 2, the frequency fT generated by the frequency multiplier 4 is plotted against the frequency fE of the natural frequency signal 11 obtained from the micromechanical oscillator. With the

Reference numeral 19 denotes the frequency range in which the measured value of the natural frequency 21 of the natural frequency signal 11 due to the manufacturing tolerance is presumed. In this frequency range 19, it is possible by selecting a suitable adjustment factor 10 to match the clock frequency 23 to the reference frequency 22. To the respective frequency f E of the

Natural frequency signal corresponding adjustment factors 10 are exemplified in the figure 2 by the reference numerals 41 to 47. These symbolize a certain ratio of the values n and m. The resulting from the choice of the adjustment factor 10 clock frequency fT follows due to the

Frequency multiplier 4 quantization error of a zigzag line.

During the calibration process, the natural frequency 21 of the

Micromechanical oscillator measured. To match the clock frequency 23 to a predetermined setpoint frequency 22, the corresponding

Adjustment value 10 selected. The adjustment factor 10 is selected, at which the frequency difference 24 between the resulting from the application of the adjustment factor 10 clock frequency 23 of the clock signal 13 and the target frequency 22 is the lowest. In this example, as

Adjustment factor 10, the value 44 is chosen, which stands symbolically for a certain ratio of the values n and m, for example, 15/16 and this coded. The adjustment factor 10 is stored in the adjustment register 3.

In order to determine the correction factor fcorr, the quotient of the clock frequency 23 of the clock signal 13 and the setpoint frequency 22 is formed. The correction factor fcorr is stored in a non-volatile correction register 5 of the sensor system 1. To determine the correction factor fcorr, the clock frequency 23 can be measured. Alternatively, it is possible to calculate the clock frequency 23 on the basis of the measured natural frequency 21 and the selected adjustment factor 10.

Based on the illustration in Figure 3, the calculation of the

Orientation of the sensor system 1 will be explained. FIG. 3 shows the calculated angle φ over a first time axis t, where

Quantization error of the frequency multiplier 4 is not corrected and the calculated angle φ 'over a second time axis t', wherein the

Quantization error is corrected.

If the correction factor fcorr is not taken into account in the calculation unit 6, then the calculated angle φ is loaded with the inaccuracy caused by the quantization error of the frequency multiplier 4. The rotation rate ω measured with the first rotation rate sensor 2 is integrated over an integration time to obtain a first angle φ. This relationship is shown in the following equation:

If the correction factor is used in the calculation unit 6 to calculate the angle φ ', the quantization error is compensated and the accuracy of the angle φ' is improved. This will be done in the

Calculation unit 6, the discrete time points with the correction factor fcorr multiplied, which is shown in the following equations as an example for the times t ₀ and ί _τ . to '= t ₀ * fcorr

ti '= ^f i * fcorr

For the angle φ 'the following equation results:

The above calculation may differ from that

Embodiment for all three spatial directions are performed, are determined for the rotation rate in the sensor system 1. Furthermore, in the calculation of the angle, additional acceleration data 15 and / or

Magnetic field data 16 are consulted. In that sense, in the

Calculation unit 6 a fusion of the data of several sensors are performed.

In the method described above for calculating a

Solid angle φ 'with a sensor system 1, which has a micromechanical rotation rate sensor 2 and a calculation unit 6, wherein a

Rate of rotation ω is measured with the rotation rate sensor 2 and for calculating the solid angle φ 'the rotation rate ω is integrated in the calculation unit over time, oscillates a micromechanical oscillator of the rotation rate sensor 2 with a natural frequency 21 and generates a natural frequency signal 11, wherein from the natural frequency signal 11 a Clock signal 13 for integrating in the

Calculation unit 6 is derived. As a result, a cost-effective calculation of the solid angle φ 'and a cost-effective sensor system 1 is made possible.

Claims

claims

1. Method for calculating an orientation with a

Sensor system (1), which is a micromechanical

Rotation rate sensor (2) and a calculation unit (6), wherein a rate of rotation (co) with the rotation rate sensor (2) is measured and to calculate the orientation of the rotation rate (ω) in the

Calculation unit (6) is integrated over time, thereby

marked that

a micromechanical oscillator of the rotation rate sensor (2) oscillates at a natural frequency (21) and generates a natural frequency signal (11), wherein a natural time signal (13) for the integration in the calculation unit (6) is derived from the natural frequency signal (11).

2. The method according to claim 1, characterized in that for

Deriving the clock signal (13), the natural frequency signal (11) is supplied to a frequency multiplier (4), wherein the

Frequency multiplier (4) the natural frequency (21) of the

micromechanical oscillator with a matching factor (10)

multiplied.

3. The method according to any one of the preceding claims, characterized

characterized in that the adjustment factor (10) is determined, wherein the natural frequency (21) of the micromechanical oscillator is measured and the measured natural frequency (21) is compared with a predetermined nominal frequency (22).

4. The method according to any one of the preceding claims, characterized

characterized in that the calculation unit (6) a

Correction factor (fcorr) is supplied to compensate for a quantization error of the frequency multiplier (4).

5. The method according to any one of the preceding claims, characterized

in that the correction factor (fcorr) corresponds to the ratio of the clock frequency (23) of the clock signal (13) to the setpoint frequency (22). equivalent.

Method according to one of the preceding claims, characterized in that the clock frequency (23) of the clock signal (13) for determining the correction factor (fcorr) is measured.

Method according to one of the preceding claims, characterized in that the correction factor (fcorr) in a

Correction register (5) of the sensor system (1) is stored.

Method according to one of the preceding claims, characterized in that the calculation unit (6) for calculating the orientation detected with an acceleration sensor

Acceleration values (15) and / or magnetic field values (16) detected by a magnetic field sensor are supplied.

Sensor system with a micromechanical rotation rate sensor (2) for measuring a rate of rotation (co) and with a calculation unit (6) for calculating an orientation by integration of the measured rate of rotation (ω) over time, characterized in that

the rotation rate sensor (2) has one with a natural frequency (21)

having oscillating, micromechanical oscillator, with which a natural frequency signal (11) can be generated, wherein from the

Natural frequency signal (11), a clock signal (13) for integrating in the calculation unit (6) is derivable.

Sensor system according to claim 9, characterized in that the sensor system (1) comprises a frequency multiplier (4), which is designed such that the natural frequency (21) of the

Micromechanical oscillator with a predetermined

Adjustment factor (10) is multiply and wherein the sensor system (1) comprises a correction register (5) in which a correction factor (fcorr) is stored, which the calculation unit (6) can be supplied to correct a quantization error of the frequency multiplier (4).