The present application is based on and claims the benefit of priority of Japanese Patent Application No. 2018-073832, filed on Apr. 6, 2018, the disclosure of which is incorporated herein by reference.
The present disclosure relates to a vibrating structure or vibration type gyroscope.
Microelectromechanical (MEMS) devices such as MEMS resonators may be used as components in vibration-type (“vibration type”) gyroscopes due to the availability and cost of MEMS resonators. However, such MEMS devices may be prone to accuracy issues due to manufacturing errors. As such, vibration type gyroscopes are subject to improvement.
The present disclosure describes a vibration type gyroscope that can dynamically perform a highly accurate phase correction of signals during signal processing.
Objects, features, and advantages of the present disclosure will become more apparent from the following detailed description made with reference to the accompanying drawings, in which:
Microelectromechanical systems (MEMS) or MEMS devices such as MEMS resonators may be used in vibrating structure type gyroscopes due the relative wide availability and low cost of MEMS resonators. However, MEMS resonators may be prone to manufacturing error and as a result cause accuracy issues in the gyroscope (e.g., sensor values, output values).
An input drive signal may provide a drive voltage to the MEMS resonator at the Drive In terminal. When the drive voltage is applied to the MEMS resonator, the MEMS resonator vibrates/oscillates along its drive axis. The oscillation of the MEMS resonator causes a displacement of the MEMS resonator along the drive axis, and such a displacement causes an electric current in the output drive signal at the “Drive Out” terminal. Such an electric current can be used to detect and measure the displacement of the MEMS resonator during oscillation.
A drive axis oscillation loop is formed from the Drive Out terminal and loops to the Drive In terminal of the MEMS resonator. The drive axis oscillation loop can be used to control the frequency and amplitude of the MEMS resonator displacement along the drive axis direction so that the MEMS resonator displacement is driven with the same frequency and amplitude.
When an angular velocity is applied to the MEMS resonator in a direction perpendicular to the drive axis and sense axis, a displacement that is proportional to the Coriolis force and having the same frequency as the input drive signal is generated along the sense axis. Such a displacement also causes an electric current that is output at the Sense Out terminal. As such, the sense signal is an amplitude modulated (AM) signal proportional to the Coriolis force with a carrier frequency that is the same as the input drive signal frequency.
The drive axis oscillation loop includes a charge amplifier (CA), a phase locked loop (PLL) (not shown), and an automatic gain control (AGC) circuit. Since an angular velocity signal “Rate” in the sense signal is generated by the Coriolis force and subjected to amplitude modulation (AM) by a resonance frequency, the value of the angular velocity signal Rate can be derived by using a synchronously detected output drive signal and by cancelling the harmonic frequency component using a low-pass filter (LPF).
The sense signal output from the Sense Out terminal normally includes a Quadrature Bias Error (QBE) and an In-phase Bias Error (IBE), in addition to the angular velocity signal, due to manufacturing errors in the MEMS resonator. Since a quadrature bias error signal is orthogonal to the angular velocity signal, the quadrature bias error signal can be eliminated by synchronously detecting the sense signal with the drive signal. In reality, however, as shown in
In order to solve such a problem, the phase difference θmis may be corrected to zero. For example, there are methods of correcting the errors by calibrating the gyroscope at a reference temperature T0, which uses a quadrature phase error empirically identified during the manufacturing and/or the calibration of the gyroscope, by using a vector operation.
However, by using such a method, it is necessary to specify and obtain the correction information in advance. In other words, in instances where no configuration information is available and/or element characteristics of the gyroscope fluctuate due to aging, a high-accuracy phase correction cannot be performed. Such calibration may also be time consuming when shipping the gyroscopes/MEMS resonators from the factory, which leads to an increase in manufacturing costs.
A vibration type gyroscope in one aspect of the present disclosure may use a resonator that includes a drive signal input terminal, a drive signal output terminal, and a sense signal output terminal for outputting a modulated sense signal corresponding to an externally-applied angular velocity as modulated by a drive signal. A quadrature demodulator may quadratically demodulate the modulated sense signal, and first and second offset canceller circuits may respectively cancel a direct current (DC) offset component included in an in-phase signal and a quadrature signal of the sense signal.
An error detector may detect a quadrature error based on a signal input from the first and second offset canceller circuits, and output an error signal. A phase corrector may receive an input of the in-phase signal and the quadrature signal of the sense signal, and output a phase signal having a corresponding phase to the error signal.
If a quadrature phase error emir exists in the drive signal and the sense signal, the in-phase error and the quadrature error can be eliminated by cancelling the DC offset component from the demodulated signal, since the in-phase error and the quadrature error are included in the quadratically-demodulated signal as DC offset components in an angular velocity signal, which is an alternating current (AC) component. Then, by using the error-cancelled in-phase and quadrature signals, the quadrature phase error θmis can be calculated. Consequently, by outputting the phase signal having the corresponding phase to the error signal based on the in-phase signal and the quadrature signal of the sense signal, the phase corrector can dynamically perform the phase correction for eliminating the influence of the in-phase error and the quadrature error, and is capable of more accurately detecting the angular velocity.
Other embodiments described herein may reduce rounding errors and minimize the finite word length effects of digital processing, so that the phase correction can performed with an even higher accuracy.
Prior to the description of the vibration type gyroscope 1, the principle of phase correction is described. The principle of phase correction in the present embodiment assumes for the in-phase bias error (IBE) signal and the quadrature bias error (QBE) signal that “ω” is the drive axis vibration frequency and “ARexp(jωt)” is the rate signal based on an angular velocity, where the IBE signal and QBE signal may be defined as:
IBE signal: AIBexp(jωt)
QBE signal: AQBexp(jωt+jτ/2)
A modulated rate signal “Smod” that is output from a sense axis is represented by equation (1).
S mod=AR exp(jωt)+AIB exp(jωt)+AQB exp(jωt+jτ/2) Equation (1)
When the quadrature phase error θmis exists in the drive signal and in the sense signal, the quadratically-demodulated signals Isig and Qsig are represented by equations (2) and (3).
Isig=AR cos θmis+(AIB cos θmis−AQB sin θmis) Equation (2)
Qsig=AR sin θmis+(AIB sin θmis+AQB cos θmis) Equation (3)
In equations (2) and (3), the AR cos θmis and AR sin θmis signal components are alternating current (AC) components, and the IBE and QBE signal components, i.e., (AIB cos θmis−AQB sin θmis) and (AIB sin θmis+AQB cos θmis), are direct current (DC) components.
The in-phase signal Isig_dcoc and the quadrature signal Qsig_dcoc are represented by equations (4) and (5) after cancelling the DC offset from the in-phase signal Isig and the quadrature signal Qsig.
Isig_dcoc=AR cos θmis Equation (4)
Qsig_dcoc=AR sin θmis Equation (5)
Equations (4) and (5) are information that includes quadrature errors. The amount of the quadrature phase error θmis can be calculated from equation (6) using equations (4) and (5).
θmis=tan−1(Qsig_dcoc/Isig_dcoc) Equation (6)
When the level of the Rate signal detected by a signal detector exceeds a threshold value for a preset period of time, the amount of the quadrature phase error θmis can be calculated by equation (6). By correcting the sense signal by an angle ϕ=−θmis using an I signal and a Q signal, the quadrature error can be corrected by equation (7).
Ical=Isig·cos ϕ−Qsig·sin ϕ Equation (7)
The above calculations limit and/or eliminate the influence of the IBE signal and the QBE signal so that the angular velocity can be more accurately detected.
The vibration type gyroscope 1 further includes a reference signal process circuit 31, a sense signal process circuit 32, a quadrature demodulator 33, and a quadrature error correction circuit 34.
An impedance conversion amplifier 3 (i.e., a transimpedance amplifier (TIA) 3) is connected to the Drive Out output terminal. The TIA 3 converts the change of electrostatic capacitance in the MEMS sensor 2 to a voltage value. The voltage value converted by the TIA 3 is further converted to digital data by an A/D converter “ADC_D” 4 in the reference signal process circuit 31. The digital data is then input to a high-pass filter (HPF) 5.
The output voltage of the TIA 3 (i.e., the converted voltage value) is a drive signal having a resonance frequency Fd of the MEMS sensor 2. In addition to being input into the A/D converter 4, the converted voltage value is also input to an automatic gain control (AGC) circuit 6.
The output signal of the AGO circuit 6 is input to the Drive In input terminal of the MEMS sensor 2 via a driver 7. When the gyroscope 1 begins operation, the AGC circuit 6 compares an amplitude of a reference voltage “VRef” with an amplitude of the converted voltage from the TIA 3, and then performs a gain control based on the comparison result to stabilize the amplitude of the converted voltage from the TIA 3. By doing such, the amplitude of the drive signal can be maintained. The above description describes an oscillation loop of the drive axis of the MEMS sensor 2. The resonance frequency Fd is, for example, roughly 10 kHz to 20 kHz.
The phase of the data output by the HPF 5 is shifted by 90 degrees by the phase shifter 8. The phase shifter 8 may include, for example, a Hilbert transformer circuit, by which a quadrature signal D_Q is generated. The phase shifter 8 also gives the output data by the HPF 5 a time delay equal to the time used to perform the above-described phase shift, and outputs the time-delayed signal as an in-phase signal D_I.
When the angular velocity is applied to the MEMS sensor 2 in a state where the oscillation loop of the drive axis is operating and the oscillation of the drive signal is in a steady state after the gyroscope 1 begins to operate, the application of the Coriolis force causes an AM modulated sense signal to be output from the output terminal Sense Out. The sense signal is modulated by the frequency of the drive signal that is input at the Drive In terminal of the MEMS sensor 2.
When the modulated sense signal is output from the sense signal output of the MEMS sensor 2, the modulated sense signal is input to a sense signal process circuit 32, where the signal passes through a charge amplifier (CA) 9, an analog-to-digital converter ADC_S 10, a high-pass filter (HPF) 11, and a delay (Dly) circuit 12.
The Dly circuit 12 sets a delay time that takes both the delay in the phase shifter 8 and a fixed offset delay from the Sense Out output terminal into consideration. That is, the delay may take into consideration the same delay time as the one given to the in-phase signal inside the phase shifter 8 and a fixed offset delay time from the output terminal Sense Out of the MEMS sensor 2 to the sense signal process circuit 32. The fixed offset delay time can be set by a microcontroller or like processing device by a preset or fixed delay time stored in memory. After passing through the Dly circuit 12, the processed modulated sense signal is input to the quadrature demodulator 33.
The quadrature demodulator 33 includes multipliers (i.e., mixers) 13I and 13Q, and low-pass filters (LPF) 14I and 14Q. The in-phase signal D_I and the quadrature signal D_Q from the phase shifter 8 are input respectively to the mixers 13I and 13Q of the quadrature demodulator 33. The mixers 13I and 13Q perform synchronous detection (i.e., demodulation) by respectively multiplying the modulated sense signal by the in-phase signal D_I and the quadrature signal D_Q. The detected signal (i.e., the modulated sense signal that has been multiplied by the in-phase signal and quadrature signal) is then quadratically demodulated by the LPFs 14I and 14Q and output as a sense signal.
From the quadrature demodulator 33, the sense signal is input to the quadrature error correction circuit 34. In the quadrature error correction circuit, as shown in
The first and second offset canceller circuits 16I and 16Q are described in greater detail with reference to
The signals Isig_dcoc and Qsig_dcoc are then input to both a quadrature error detector 20 and to a detector 21. The quadrature error detector 20 calculates the quadrature phase error θmis by equation (6), and outputs the quadrature phase error θmis to a control circuit 22. The quadrature error detector 20 may also be referred to simply as the error detector 20. The detector 21 calculates a signal power. Signal power, as shown in
The control circuit 22 may be a digital circuit such as, for example, a microcontroller, a field-programmable gate array (FPGA), and a sequencer, and the control circuit 22 is configured to calculate an average value. A memory 23 is connected to the control circuit 22. The memory can store an amount of the quadrature error correction ϕ as an initial value. The control circuit 22 reads the initial value stored in the memory 23 immediately after the gyroscope 1 begins operating, and outputs the initial value to the IQ correction circuit 15. Thereafter, when the detector 21 outputs a high-level “H” output signal, i.e., a trigger signal, the control circuit 22 performs an averaging process of the quadrature phase error θmis for a preset period of time. In other words, the control circuit 22 is configured to calculate the average of a plurality of error signals that are input from the error detector 20 for a preset period of time in response to receiving the trigger signal. As such, the control circuit 22 may also be referred to as the average calculator 22. The stored average value is then set as the quadrature error correction amount ϕ at a preset interval (i.e., ϕ=−θmis). The quadrature error correction amount ϕ is then output to the IQ correction circuit 15.
As shown in
The correction signal generator 24 may include, for example, a microcontroller or a digital signal processing (DSP) circuit (not shown). Alternatively, a table with the sine and cosine values for each angle ϕ may be prepared in advance and stored in memory or a register, and, when an angle ϕ is input, a corresponding sin and/or cos value to the angle ϕ may be read from the table and output.
The DC offset component included in the correction signal Ical is cancelled by a DC offset correction circuit (DCOC2) 27 disposed on an output side of the IQ correction circuit 15, and the DCOC2 outputs the value of the angular velocity signal Rate from the quadrature error correction circuit 34. While the DCOC116I may be referred to as the first offset canceller circuit 16I, and the DCOC116Q may be referred to as the second offset canceller circuit 16Q, the DCOC227 may be referred to as the third offset canceller circuit 27.
The gyroscope 1 as a whole may be composed of application-specific integrated circuits (ASICs) or like hardware. For example, each of the reference signal processing circuit 31, the sense signal processing circuit 32, the quadrature demodulator 33, and the quadrature error correction circuit 34 and/or their internal components may each be realized as one or more ASICs specifically configured to perform the processing, functions, and calculations associated with each of these circuits, as described above.
As described above, according to the present embodiment, the gyroscope 1 includes the MEMS sensor 2 including the input terminal Drive In, the output terminal Drive Out, and the output terminal Sense Out. The quadrature demodulator 33 quadratically demodulates the modulated sense signal, and the offset canceller circuits (DCOC1) 16I and 16Q respectively cancel the direct current offset component included in the in-phase signal and the quadrature signal of the sense signal. The quadrature error detector 20 detects the quadrature error θmis based on the signals input from the offset canceller circuits (DCOC1) 16I and 16Q, and outputs the error signal. The IQ correction circuit 15 receives the in-phase signal Isig and the quadrature signal Qsig of the sense signal as inputs, and outputs the phase signal Ical having the phase ϕ based on the error signal.
The phase shifter 8 generates the in-phase signal of the drive signal and the quadrature signal of the drive signal, where the quadrature signal of the drive signal has a phase difference of 90 degrees from the in-phase signal of the drive signal. The quadrature demodulator 33 multiplies the modulated sense signal by the in-phase signal and the quadrature signal of the drive signal to demodulate the sense signal. The demodulated in-phase signal and the demodulated quadrature signal of the sense signal are input both to the offset canceller circuits 16I and 16Q and to the IQ correction circuit 15, and the offset canceller circuit (DCOC2) 27 cancels the DC offset component included in the phase signal Ical that is output from the IQ correction circuit 15, and outputs the angular velocity signal Rate. By using the above-described configuration, the angular velocity can be made highly accurate and better detectable by dynamically performing the phase correction to eliminate the influence of the in-phase error IBE and the quadrature error QBE.
In the description of the current embodiment, like features and elements from the previous embodiment may be used and be denoted by the same reference numbers used in the description of the previous embodiment. As such, a repeat description of the like features and elements from the previous embodiment may be omitted in the description of the current embodiment. The description of the current embodiment focuses on the differences between the current embodiment and the previous embodiment.
As shown in
According to the second embodiment configured as described above, since the in-phase signal Isig_dcoc and the quadrature signal Qsig_dcoc are input to the IQ correction circuit 15 with the DC offset component already having been cancelled, the DCOC2 circuit 27 from the first embodiment may be eliminated from the current embodiment to reduce the overall area and number of parts of the quadrature error correction circuit 42, and to reduce the overall size of and simplify the gyroscope 41.
As shown in
The analog AGC circuit 6 of the previous embodiments may use a filter for compensating a phase margin that may cause a tradeoff between the gain control stability and the high-speed pull-in. On the other hand, if the AGC circuit 52 is configured as the digital circuit described in the third embodiment, the phase compensation filter may have more flexibility in its design. Consequently, the drive-axis oscillation loop with the digital AGC 52 in the present embodiment can have both increased stability and high-speed pull-in without having to sacrifice one for the other.
As shown in
By using the above-described configuration, the noise in the drive signal can be reduced/limited by the LF 65 in the PLL circuit 63. Consequently, the gyroscope 61 in the fourth embodiment can reduce signal noise and thus has low-noise characteristics.
As shown in
As shown in
Using the fifth embodiment configured as described above can reduce the rounding errors in the digital processing of drive axis signals having large amplitudes and reduce the finite word-length-effect of digital processing. As a result, a higher degree of accuracy may be achieved in the phase correction processing.
As shown in
In other words, the sense signal processing circuit 83 has the phase shifter 8 disposed at a position on an output side of the high-pass filter (HPF) 11, and the phase shifter 8 outputs the in-phase signal S_I and the quadrature signal S_Q of the modulated sense signal to the quadrature demodulator 33. The reference signal processing circuit 82 has the Dly circuit 12 disposed at a position on an output side of the HPF 5, and outputs a drive signal having an added time delay to the quadrature demodulator 33.
By using the configuration in the above-described current embodiment, signal processing similar to the signal processing in the third embodiment can be performed. Since the noise superposed on the sense signal can be cancelled by the filtering characteristics of the HPF 11 and the phase shifter 8, by using the configuration of the current embodiment, the gyroscope 81 in the current embodiment can reduce signal noise and thus has low-noise characteristics.
As shown in
If the drive signal includes a harmonic component, signal noise from aliasing may be generated by the frequency conversion in the quadrature demodulator 33, and the signal-to-noise (S/N) ratio may become worse (i.e., deteriorate) in some cases. By using the above-described configuration of the current embodiment, or more specifically, by arranging the IQ correction circuit 73 in front of the quadrature demodulator 33, the quadrature error correction can be performed in a good S/N ratio state.
The resonator is not limited to a MEMS resonator. The quadrature error correction circuits in the previous embodiments may be configured so as to make the detector 21, the control circuit 22, and the memory 23 as optional elements. Specific numerical frequency values may be changed based on the design of the device/system. The embodiments may be combined with one another to make new embodiments where elements from each of the embodiments may be added and/or removed from the combined embodiment.
Although the present disclosure has been made in accordance with the respective embodiments, it is understood that the present disclosure is not limited to those embodiments and structures. The present disclosure covers various modification examples and equivalent arrangements.
Number | Date | Country | Kind |
---|---|---|---|
JP2018-073832 | Apr 2018 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
5889193 | Pfaff | Mar 1999 | A |
6062082 | Guenther | May 2000 | A |
6553833 | Funk | Apr 2003 | B1 |
9574902 | Okon | Feb 2017 | B2 |
9575089 | Cazzaniga et al. | Feb 2017 | B1 |
9702898 | Lipka | Jul 2017 | B1 |
20050283330 | Laraia | Dec 2005 | A1 |
20060174684 | Betz et al. | Aug 2006 | A1 |
20060260382 | Fell et al. | Nov 2006 | A1 |
20070177316 | Hotelling | Aug 2007 | A1 |
20070240486 | Moore | Oct 2007 | A1 |
20100132463 | Caminada | Jun 2010 | A1 |
20100294039 | Geen | Nov 2010 | A1 |
20110017868 | Arthurs et al. | Jan 2011 | A1 |
20110192226 | Hayner | Aug 2011 | A1 |
20120096942 | Hayashi | Apr 2012 | A1 |
20120146740 | Furuta | Jun 2012 | A1 |
20130031950 | Donadel | Feb 2013 | A1 |
20130152664 | Pyo | Jun 2013 | A1 |
20130199294 | Townsend | Aug 2013 | A1 |
20130268228 | Opris | Oct 2013 | A1 |
20130269413 | Tao | Oct 2013 | A1 |
20140260619 | Shaeffer | Sep 2014 | A1 |
20140373595 | Glueck | Dec 2014 | A1 |
20150057959 | Ezekwe | Feb 2015 | A1 |
20150128699 | Maki | May 2015 | A1 |
20150160632 | Buhmann | Jun 2015 | A1 |
20150276406 | Rastegar | Oct 2015 | A1 |
20150276407 | Bhandari | Oct 2015 | A1 |
20160018243 | Okon | Jan 2016 | A1 |
20160109258 | Boser | Apr 2016 | A1 |
20160161256 | Lee | Jun 2016 | A1 |
20160187136 | Murakami et al. | Jun 2016 | A1 |
20160231112 | Stewart | Aug 2016 | A1 |
20160277011 | Gil | Sep 2016 | A1 |
20160327393 | Shaeffer | Nov 2016 | A1 |
20170023364 | Gregory | Jan 2017 | A1 |
20170167875 | Bhandari | Jun 2017 | A1 |
20170307374 | Hughes | Oct 2017 | A1 |
20170322029 | Townsend | Nov 2017 | A1 |
20170328712 | Collin | Nov 2017 | A1 |
20170350722 | Senkal | Dec 2017 | A1 |
20180017385 | Shirvani | Jan 2018 | A1 |
20180058854 | Suzuki | Mar 2018 | A1 |
20180245946 | Weinberg | Aug 2018 | A1 |
20180259359 | Mansfield | Sep 2018 | A1 |
20180274924 | Mecchia | Sep 2018 | A1 |
20180274941 | Quartiroli | Sep 2018 | A1 |
20190120657 | Senkal | Apr 2019 | A1 |
20190145773 | Collin | May 2019 | A1 |
20190186950 | Dakshinamurthy | Jun 2019 | A1 |
20190265036 | Arndt | Aug 2019 | A1 |
20200400433 | Aaltonen | Dec 2020 | A1 |
Number | Date | Country |
---|---|---|
2008-082924 | Apr 2008 | JP |
2012-080320 | Apr 2012 | JP |
Number | Date | Country | |
---|---|---|---|
20190310106 A1 | Oct 2019 | US |