Gyro sensor and demodulation circuit
The gyro sensor uses a MEMS structure with a phase locked loop and Hilbert transform demodulation to enhance accuracy and speed, addressing inaccuracies in existing PLL-based demodulation methods.
Patent Information
- Application Number
- JP2024042643
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-03-18
- Publication Date
- 2025-10-01
AI Technical Summary
Existing gyro sensors using phase-locked loops (PLL) for demodulation suffer from inaccurate amplitude and inclination measurements of the elliptical orbit and interference between PLL control loops, leading to suboptimal performance.
A gyro sensor employing a MEMS structure with a phase locked loop circuit for phase tracking and a demodulation circuit using a Hilbert transform to calculate driving forces based on position information, eliminating the need for synchronous detection and reducing demodulation errors.
The proposed solution improves demodulation accuracy and response speed, reducing errors and interference, thereby enhancing the overall performance of the gyro sensor.
Smart Images
Figure 2025142985000001_ABST
Abstract
Description
[Technical Field]
[0001] FIELD An embodiment of the present invention relates to a gyro sensor and a demodulation circuit. [Background technology]
[0002] In gyro sensors, such as direct angle detection gyro sensors or rate integrating gyroscopes (RIGs) made with MEMS (Micro Electro Mechanical Systems), the oscillator is continuously vibrated in a straight line with a desired amplitude in the xy plane by controlling the voltage applied to the electrodes on the same element as the oscillator, thus realizing a pseudo-Foucault pendulum. This control requires demodulation of the oscillator's trajectory in the xy plane (the amplitude of the major and minor axes of the elliptical orbit, the inclination of the elliptical orbit, and the oscillator's phase on the elliptical orbit) from the detected oscillator position signals along the x and y axes. One common demodulation method is the synchronous detection method using a phase-locked loop (PLL) to detect the oscillator's phase. However, this method can result in inaccurate demodulation of the amplitude of the major and minor axes of the elliptical orbit and the inclination of the elliptical orbit until the PLL converges, or interference between the PLL control loop and the oscillator's excitation control loop. [Prior art documents] [Non-patent literature]
[0003] [Non-Patent Document 1] D. D Lynch. “Vibratory gyro analysis by the method of averaging”. In: Proc. 2nd St. Petersburg Conf. on Gyroscopic Technology and Navigation, St. Petersburg. 1995, pp. 26-34. Summary of the Invention [Problem to be solved by the invention]
[0004] A gyro sensor and a demodulation circuit capable of improving performance are provided. [Means for solving the problem]
[0005] The gyro sensor of the embodiment includes a MEMS including a vibrator, a spring element supporting the vibrator, a sensor detecting the position of the vibrator on a two-dimensional plane, and an actuator applying a driving force to the vibrator, and a processing circuit including a phase locked loop circuit for tracking the phase φ of the vibrator on the two-dimensional plane and calculating the driving force to be applied to the actuator based on position information corresponding to the position detected by the sensor.
[0006]
number
[0007] The driving force is calculated based on the energy E, the quadrature amplitude Q, the vibration main axis angle θ, and the error δφ. [Brief explanation of the drawings]
[0008] [Figure 1] FIG. 1 is a block diagram showing a configuration of a gyro sensor according to an embodiment. [Figure 2] 3A and 3B are diagrams illustrating a trajectory and demodulation information of a transducer on an xy plane according to an embodiment. [Figure 3] FIG. 10 is a configuration diagram showing a comparative example. [Figure 4] FIG. 2 is a configuration diagram illustrating a demodulation method in the gyro sensor according to the embodiment. [Figure 5] 10 is a flowchart illustrating a flow of operations including a demodulation method in the gyro sensor according to the embodiment. [Figure 6] FIG. 1 is a diagram plotting the trajectory of a vibrator on the xy plane. [Figure 7]10A and 10B are diagrams illustrating gain characteristics and phase delay amounts in a Hilbert transform filter and an LPF. [Figure 8] 10A and 10B are diagrams showing demodulated information demodulated in the embodiment and the comparative example. [Figure 9] 10A and 10B are diagrams showing demodulated information demodulated in the embodiment and the comparative example. [Figure 10] FIG. 10 is a diagram showing the E value of the output portion where the transient responses of the Hilbert transform filter and the LPF have converged. [Figure 11] FIG. 10 is a diagram illustrating gain ripples in a portion where the gain is flat in a Hilbert transform filter. DETAILED DESCRIPTION OF THE INVENTION
[0009] Hereinafter, embodiments will be described with reference to the drawings. In the following description, components having the same function and configuration will be assigned common reference symbols, and letters or numbers may be added to the end of the common reference symbols. Furthermore, the embodiments shown below are merely examples of devices and methods for embodying the technical ideas of these embodiments, and do not specify the materials, shapes, structures, arrangements, etc. of the components as described below.
[0010] The gyro sensor according to the embodiment will be described below.
[0011] 1. Configuration of the embodiment The configuration of the gyro sensor of the embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the gyro sensor of the embodiment.
[0012] 1, for example, and outputs an electrical signal corresponding to the rotational angular velocity. The gyro sensor 1 includes a MEMS 10, a capacitance-voltage conversion amplifier circuit (hereinafter referred to as a CV amplifier circuit) 20, an analog-to-digital converter (hereinafter referred to as an ADC) 30, a digital signal processor (hereinafter also referred to as a DSP) 40, a central processing unit (CPU) 50, a digital-to-analog converter (hereinafter referred to as a DAC) 60, a RAM 70, a ROM 80, and an interface circuit (I / F) 90.
[0013] The gyro sensor 1 has an analog circuit 1A that processes analog signals and a digital circuit 1D that processes digital signals. The analog circuit 1A includes a MEMS 10, a CV amplifier circuit 20, and an ADC 30. The digital circuit 1D includes a digital signal processing circuit 40, a CPU 50, a DAC 60, a RAM 70, a ROM 80, and an interface circuit 90.
[0014] The MEMS 10 is a device that is installed in the gyro sensor 1 and detects mechanical displacement occurring in the gyro sensor 1. The MEMS 10 includes a vibrator 11, a spring element 12, a sensor 13, and an actuator .
[0015] The oscillator 11 acts as an inertial body. The spring element 12 supports the oscillator 11.
[0016] The sensor 13 includes two sensors 13_1 and 13_2. Hereinafter, when there is no need to distinguish between the sensors 13_1 and 13_2, they may be referred to as the sensor 13. When the sensor 13 is referred to, it refers to either or both of the sensors 13_1 and 13_2.
[0017] The sensors 13_1 and 13_2 each have opposing electrodes, for example, opposing comb-shaped electrodes. The sensor 13_1 detects the displacement (or position) of the vibrator 11 in the x-axis direction on the xy plane as the capacitance of the electrodes. The sensor 13_2 detects the displacement (or position) of the vibrator 11 in the y-axis direction on the xy plane as the capacitance of the electrodes. The sensors 13_1 and 13_2 output signals x1 and y1 corresponding to the detected capacitances to the CV amplifier circuit 20.
[0018] The actuator 14 includes two actuators 14_1 and 14_2. Hereinafter, when there is no need to distinguish between the actuators 14_1 and 14_2, they may be referred to as the actuator 14. When the actuator 14 is referred to, it refers to either or both of the actuators 14_1 and 14_2.
[0019] The actuators 14_1 and 14_2 each have opposing electrodes, for example, opposing comb-shaped electrodes. The actuator 14_1 applies a driving force to the vibrator 11 in the x-axis direction. The actuator 14_2 applies a driving force to the vibrator 11 in the y-axis direction.
[0020] The CV amplifier circuit 20 has two CV amplifier circuits 20_1 and 20_2. Hereinafter, when there is no need to distinguish between the CV amplifier circuits 20_1 and 20_2, they may be referred to as the CV amplifier circuit 20. When the CV amplifier circuit 20 is referred to, it refers to either or both of the CV amplifier circuits 20_1 and 20_2.
[0021] The CV amplifier circuit 20 converts a change in capacitance detected by the sensor 13 into a voltage and amplifies the voltage. The CV amplifier circuits 20_1 and 20_2 convert signals x1 and y1 output from the sensors 13_1 and 13_2 into voltages and amplify the voltages. The CV amplifier circuits 20_1 and 20_2 output signals x2 and y2 to the ADC 30 as converted and amplified voltages.
[0022] The ADC 30 has two ADCs, 30_1 and 30_2. Hereinafter, when there is no need to distinguish between the ADCs 30_1 and 30_2, they may be referred to as ADC 30. When ADC 30 is referred to, it refers to either or both of the ADCs 30_1 and 30_2.
[0023] The ADC 30 is a circuit that converts the voltage of the analog signal converted by the CV amplifier circuit 20 into a digital signal. The ADCs 30_1 and 30_2 convert the signals x2 and y2 output from the CV amplifier circuits 20_1 and 20_2 into signals x3 and y3, respectively. The ADCs 30_1 and 30_2 output the signals x3 and y3, respectively, to the digital signal processing circuit 40 (or the CPU 50).
[0024] The digital signal processing circuit 40 performs various processes, i.e., processes related to a demodulation technique, which will be described later, using the signals x3 and y3 output from the ADCs 30_1 and 30_2 and a reference phase φr of a signal output from a CPU 50, which will be described later. The digital signal processing circuit 40 performs the demodulation technique and calculates demodulated information. The digital signal processing circuit 40 outputs the calculated demodulated information to the CPU 50.
[0025] The digital signal processing circuit 40 may be any circuit other than the above-mentioned DSP, such as an FPGA (Field-Programmable Gate Array), as long as it can perform calculations related to digital signals at high speed.
[0026] The CPU (Central Processing Unit) (or processing circuit, control circuit) 50 comprehensively controls the overall operation of the gyro sensor 1. The CPU 50 can also execute various types of arithmetic processing. The CPU 50 and the digital signal processing circuit 40 are interconnected, and demodulation information and signals for controlling the timing of demodulation are transmitted and received. For example, the CPU 50 may execute a demodulation method and calculate demodulation information instead of the digital signal processing circuit 40. The CPU 50 can also include a phase locked loop (PLL) and output a signal (reference phase φr) generated by the PLL.
[0027] The CPU 50 calculates the driving force to be applied to the vibrator 11 based on the demodulated information output from the digital signal processing circuit 40. The CPU 50 outputs to the DAC 60 signals x4 and y4 corresponding to the calculated driving force.
[0028] The DAC 60 has two DACs 60_1 and 60_2. Hereinafter, when there is no need to distinguish between the DACs 60_1 and 60_2, they may be referred to as DAC 60. When DAC 60 is referred to, it refers to either or both of the DACs 60_1 and 60_2.
[0029] The DAC 60 is a circuit that converts a digital signal output by the CPU 50 into an analog signal. The DACs 60_1 and 60_2 convert the signals x4 and y4 output from the CPU 50 into signals x5 and y5, respectively. The DACs 60_1 and 60_2 output the signals x5 and y5 to the actuators 14_1 and 14_2 of the MEMS 10, respectively.
[0030] RAM (Random Access Memory) 70 temporarily stores programs required for processing by CPU 50 and data required for executing the programs. CPU 50 executes the programs in RAM 70 to calculate the data in RAM 70 and stores the calculation results in RAM 70. RAM 70 is used, for example, in calculation processing related to excitation control of vibrator 11, including a demodulation method.
[0031] A ROM (Read Only Memory) 80 non-temporarily stores a startup program required when starting up the gyro sensor 1. The gyro sensor 1 starts up when the CPU 50 executes a program in the ROM 80. The ROM 80 is configured, for example, with an EPROM (Erasable Programmable Read Only Memory), and stores various settings at startup in addition to the startup program.
[0032] The interface circuit 90 controls communication between the gyro sensor 1 and the host device 2. The interface circuit 90 transmits, for example, an electrical signal corresponding to the rotational angular velocity detected by the gyro sensor 1 to the host device 2. Alternatively, the interface circuit 90 receives a request from the host device 2.
[0033] The above-mentioned ADC 30, digital signal processing circuit 40, CPU 50, DAC 60, RAM 70, ROM 80, and interface circuit 90 may be configured as a one-chip microcontroller, or may be configured as separate integrated circuits (ICs) and interconnected.
[0034] FIG. 2 is a diagram showing a trajectory of the vibrator 11 on the xy plane when vibrating, and four pieces of demodulated information targeted by the embodiment.
[0035] In Figure 2, oscillator 11 traces an elliptical orbit on the xy plane. The principle of Foucault's pendulum is reproduced in an oscillation system consisting of oscillator 11 and spring element 12, and the rotational angular velocity is detected. In order to detect the rotational angular velocity, oscillator 11 must continue to oscillate in a straight line with a constant amplitude.
[0036] Therefore, the CPU 50 calculates or acquires the major axis amplitude a and minor axis amplitude q of the elliptical orbit described by the oscillator 11 at a given moment, the inclination θ of the elliptical orbit, and the phase of the oscillator 11 on the elliptical orbit from the displacement amounts of the oscillator 11 in the x-axis and y-axis. In this manner, calculating or acquiring the major axis amplitude and minor axis amplitude of the elliptical orbit, the inclination of the elliptical orbit, and the phase on the elliptical orbit from the displacement amounts of the oscillator 11 in the x-axis and y-axis directions is referred to as demodulation in this embodiment, and the obtained information is referred to as demodulation information. The CPU 50 momentarily controls the drive voltages applied to the oscillator 11 in the x-axis and y-axis directions based on the demodulation information, so that the oscillator 11 continues to vibrate in a straight line at a constant amplitude.
[0037] Instead of calculating the major and minor axes of the elliptical orbit, the demodulated information is four items: the energy E (sum of kinetic energy and potential energy) of the oscillator 11, the angular momentum (or quadrature amplitude) Q of the oscillator 11 (angular momentum is generated when an elliptical orbit is drawn), the tilt of the elliptical orbit (or the angle of the main vibration axis) θ, and the phase error δφ between the true phase φ of the oscillator 11 and the reference phase φr. A method based on synchronous detection is known as a method for calculating the above E, Q, θ, and δφ. Hereinafter, this method based on synchronous detection will be referred to as the comparative example.
[0038] A comparative example will be described below: Fig. 3 is a configuration diagram showing the comparative example.
[0039] In the comparative example, first, a PLL is caused to generate a reference signal having a reference phase φr. Then, the reference phase φr calculated by the PLL is caused to track the true phase φ of the elliptical orbit of the oscillator 11. The error in this tracking corresponds to the phase error δφ.
[0040] Next, a sine wave sinφr and a cosine wave cosφr are generated from the reference phase φr by the sine wave generating circuit 31_2 and the cosine wave generating circuit 31_1, respectively.
[0041] Next, multipliers 32_1 and 32_2 multiply a signal x3 corresponding to the displacement amount on the x-axis of vibrator 11 from ADC 30_1 by a cosine wave cosφr and a sine wave sinφr, respectively. Similarly, multipliers 32_3 and 32_4 multiply a signal y3 corresponding to the displacement amount on the y-axis of vibrator 11 from ADC 30_2 by a cosine wave cosφr and a sine wave sinφr, respectively.
[0042] Due to the nature of trigonometric functions, the signal after multiplication is composed of a bias component and a double wave component of the reference phase φr. Therefore, the double wave components are attenuated by low-pass filters (hereinafter referred to as LPF) 33_1 to 33_4, respectively, to extract the bias component. Hereinafter, this is referred to as synchronous detection.
[0043] The bias components extracted from the multiplied signals are respectively expressed as signals c X ,s X,c y ,s y Then, the four pieces of demodulated information E, Q, θ, and Δφ are calculated by the following equations (1) to (4).
[0044]
number
[0045] In the above comparative example, the reference phase φr is made to follow the true phase φ by a PLL feedback control mechanism, and the phase error δφ is converged to 0. Therefore, due to this mechanism, the phase error δφ is not 0 from the start of the PLL feedback control until it converges. Also, the phase error δφ may not become 0 if the trajectory of the vibrator 11 deviates from a constant state due to some disturbance factor. In this case, errors occur in the demodulation information E, Q, θ, and δφ in equations (1) to (4), which are calculated assuming that δφ=0.
[0046] Furthermore, if the cutoff frequencies (frequency at the boundary between the pass band and the attenuation band) of the LPFs 33_1 to 33_4 are set low in order to increase the attenuation of the double wave in the multiplied signal, the demodulation accuracy of the demodulation information E, Q, θ, and δφ improves. However, the phase delay caused by the LPFs 33_1 to 33_4 increases, which limits the frequency band of the feedback control that determines the drive voltage of the PLL and the oscillator 11. In this case, when the oscillator 11 moves in an elliptical orbit or the amplitude decreases, it becomes difficult to improve the function of quickly returning the oscillator 11 to a linear orbit with a constant amplitude.
[0047] 2. Operation of the embodiment In this embodiment, unlike the comparative example, synchronous detection is not used, but a demodulation method using a Hilbert transform is executed instead. Fig. 4 is a configuration diagram showing the demodulation method in the gyro sensor 1 of the embodiment. Fig. 5 is a flowchart showing the flow of operations including the demodulation method in the gyro sensor 1. The demodulation method shown in the flowchart is controlled by a CPU 50.
[0048] First, when an input that generates a rotational angular velocity is applied to the gyro sensor 1, the trajectory of the vibrator 11 in the MEMS 10 is displaced. Then, the sensors 13_1 and 13_2 in the MEMS 10 detect the displacement (or position) of the vibrator 11 on the xy plane and output signals x1 and y1 corresponding to the displacement (or position) of the vibrator 11 on the x-axis and y-axis, respectively (S1). The signals x1 and y1 are, for example, the amount of change in the capacitance of the electrodes of the sensors 13_1 and 13_2, respectively.
[0049] Next, the signals x1 and y1 are input to CV amplifier circuits 20_1 and 20_2, respectively. The CV amplifier circuits 20_1 and 20_2 convert the signals x1 and y1 into voltages and amplify them. The CV amplifier circuits 20_1 and 20_2 output the converted and amplified signals x2 and y2 to the ADCs 30_1 and 30_2, respectively (S2).
[0050] Next, signals x2 and y2 are input to ADCs 30_1 and 30_2, respectively. ADCs 30_1 and 30_2 convert analog signals x2 and y2 into digital signals x3 and y3. ADCs 30_1 and 30_2 output signals x3 and y3 to digital signal processing circuit 40 (or CPU 50) (S3). Signal x3 is a signal corresponding to the displacement (or position) of vibrator 11 on the x-axis. Signal y3 is a signal corresponding to the displacement (or position) of vibrator 11 on the y-axis.
[0051] Next, the signal x3 output from the ADC 30_1 is input to a delay circuit 41 and a Hilbert transform filter 42. The signal y3 output from the ADC 30_2 is input to a delay circuit 43 and a Hilbert transform filter 44. In addition, the reference phase φr calculated by the PLL is input to a delay circuit 45.
[0052]
number
[0053] Here, the Hilbert transform filters 42 and 44 are assumed to be FIR (Finite Impulse Response) filters with, for example, 2d taps (d is an integer greater than 0). An FIR filter is a filter whose impulse response has a finite duration, and the number of taps is the number of coefficients used for multiplication in the FIR filter. The delay number of each of the delay circuits 41, 43, and 45 is assumed to be d samples. Each of the delay circuits 41, 43, and 45 delays a signal by d samples.
[0054] The FIR filters output the complex parts of the signals obtained by Hilbert transforming signals x3 and y3, respectively. Delay circuits 41 and 43 delay signals x3 and y3 by the phase delay that occurs in the complex parts of the signals that have been Hilbert transformed by the FIR filters. Delay circuit 45 delays reference phase φr by the phase delay that occurs in the complex parts of the signals that have been Hilbert transformed by the FIR filters.
[0055] The delay circuits 41, 43 and 45 and the Hilbert transform filters 42 and 44 are implemented in the digital signal processing circuit 40 as software or hardware.
[0056] The calculation of the demodulation information E, Q, θ, and Δφ according to the above equations (5) to (8) is executed by the digital signal processing circuit 40. For example, when executed by the digital signal processing circuit 40, each process is executed at a higher speed than when executed by the CPU 50.
[0057] 4 and equations (5) to (8), in the demodulation method of the embodiment, the reference phase φr is not involved in the demodulation of E, Q, and θ, i.e., the reference phase φr is not included in equations (5) to (7). Therefore, in the demodulation method of the embodiment, unlike the comparative example, demodulation errors due to PLL feedback control do not occur in E, Q, and θ.
[0058] In the comparative example, the setting of the cutoff frequency of the LPF for synchronous detection determines the trade-off relationship between demodulation accuracy and the band limit of the feedback control that determines the drive voltage of the PLL and the vibrator 11. On the other hand, in the embodiment, the number of taps of the Hilbert transform filter and the design of its filter coefficients determine the trade-off relationship between demodulation accuracy and the band limit of the feedback control. This will be explained in the effect of the embodiment below.
[0059] Next, the calculated demodulation information E, Q, θ, and δφ is input to the CPU 50. Based on the demodulation information E, Q, θ, and δφ, the CPU 50 calculates the driving force to be applied to the vibrator 11. The CPU 50 outputs signals x4 and y4 corresponding to the calculated driving forces to the DACs 60_1 and 60_2, respectively (S6).
[0060] Next, signals x4 and y4 are input to DACs 60_1 and 60_2, respectively. DACs 60_1 and 60_2 convert the digital signals x4 and y4, respectively, into analog voltage signals. DACs 60_1 and 60_2 output the converted voltage signals x5 and y5 to actuators 14_1 and 14_2 in MEMS 10, respectively (S7). Signal x5 is a voltage corresponding to the driving force calculated by CPU 50, i.e., the driving force applied to vibrator 11 in the x-axis direction. Signal y5 is a voltage corresponding to the driving force calculated by CPU 50, i.e., the driving force applied to vibrator 11 in the y-axis direction.
[0061] Next, the signals x5 and y5 are input to the actuators 14_1 and 14_2 in the MEMS 10, respectively. The actuators 14 in the MEMS 10 apply driving forces to the vibrator 11 in accordance with the signals x5 and y5 (S8). That is, the actuator 14_1 applies a driving force in the x-axis direction to the vibrator 11 in accordance with the signal x5. The actuator 14_2 applies a driving force in the y-axis direction to the vibrator 11 in accordance with the signal y5. This performs excitation control on the vibrator 11.
[0062] 3. Effects of the embodiment According to the embodiment, it is possible to provide a gyro sensor and a demodulation circuit that can improve performance.
[0063] The results of a numerical simulation performed on the demodulation method of the embodiment are shown below.
[0064] First, we assume an ideal two-degree-of-freedom vibration system with a resonant frequency of 10 kHz in the xy plane direction and no damping as a model of the vibration system created with MEMS 10. We assign appropriate values to the initial position and initial velocity of the vibrator 11 of this model, and simulate the state in which free vibration continues.
[0065] Figure 6 is a diagram plotting the trajectory of the vibrator 11 on the xy plane. The initial simulation time is set to 0.01 seconds, and 0.003 seconds after the start of the simulation, a step change is made to the reference phase φr of the PLL signal. 0.006 seconds after the start of the simulation, a constant rotational angular velocity is generated in the gyro sensor 1.
[0066] Under the above conditions, signals x3 and y3 corresponding to the displacement of the vibrator 11 in the x-axis and y-axis directions on the xy plane were acquired as time-series data, and calculations were performed using the demodulation method of the embodiment and the comparative example. The demodulated information E, Q, θ, and δφ were then compared between the demodulation method of the embodiment and the comparative example. The sampling frequency of the digital signal processing in the ADC was set to 100 kHz, and the LPF used in the comparative example was a third-order Butterworth filter with a cutoff frequency of 1 kHz. Meanwhile, the Hilbert transform filter used in the demodulation method of the embodiment had 32 taps, and its coefficients were appropriately adjusted so that the gain characteristics of the Hilbert transform filter were flat at a resonant frequency of 10 kHz. Because the Hilbert transform filter had 32 taps, the delay number d of the delay circuit was 16.
[0067] Figure 7(a) shows a comparison of the gain characteristics of the LPF and the Hilbert transform filter, and Figure 7(b) shows a comparison of the phase delay of the LPF and the phase delay of the Hilbert transform filter (delay number d = 16). Figures 8 and 9 show a comparison of the demodulated information E, Q, θ, and δφ demodulated by each method.
[0068] As shown in Fig. 8, in the comparative example, interference occurs in E and Q due to a step change in the reference phase φr of the signal output from the PLL at 0.003 seconds. On the other hand, this phenomenon does not occur in the demodulation method of the embodiment. Therefore, it can be seen that in the embodiment, the demodulation accuracy of E and Q is not affected by the feedback control operation by the PLL. In other words, it can be seen that no error due to demodulation occurs in E and Q in the embodiment.
[0069] 9, it can be seen that the demodulation method of the embodiment can calculate θ slightly faster than the comparative example, and that the demodulation method of the embodiment has a shorter transient period and can calculate δφ faster than the comparative example.
[0070] Fig. 10 is a graph (enlarged view) comparing the values of E from 0.009 to 0.01 seconds, when the transient responses of the LPF and Hilbert transform filter have sufficiently converged. Fig. 11 is a graph showing the gain ripple in the part of the Hilbert transform filter where the gain is flat.
[0071] As shown in Figure 10, in the comparative example, the second harmonic component that could not be completely removed by the LPF appears as an oscillatory error. On the other hand, in the demodulation method of the embodiment, a similar oscillatory error appears due to the Hilbert transform using the Hilbert transform filter. However, it can be seen that the demodulation method of the embodiment can reduce the error compared to the comparative example.
[0072] As shown in Figure 11, errors in the Hilbert transform are caused by gain ripple in the flat gain area. Reducing this ripple improves the accuracy of the Hilbert transform, but this requires increasing the number of taps in the Hilbert transform filter. However, increasing the number of taps increases the phase delay, which limits the widening of the frequency band of each feedback control system, just like in the comparative example.
[0073] As a result, the demodulation accuracy in the steady state is better in the demodulation method of the embodiment than in the comparative example. Furthermore, in the response comparison shown in Figures 8 and 9, the demodulation method of the embodiment has a faster signal response for each demodulation information, which is advantageous for widening the frequency band of each feedback control.
[0074] As described above, according to the embodiment, it is possible to provide a gyro sensor that can improve the accuracy of demodulation and improve performance. Furthermore, by focusing on the demodulation circuit that executes the demodulation method of the embodiment, it is possible to provide a demodulation circuit that can improve the accuracy of demodulation. The demodulation circuit includes a Hilbert transform filter, a delay circuit, and a circuit that performs the arithmetic processing of equations (4) to (8) implemented in the digital signal processing circuit 40 (or the CPU 50).
[0075] Furthermore, although the above embodiment has been described using a rate integration type gyro sensor as an example, the demodulation circuit of the embodiment can also be applied to other gyro sensors. Furthermore, the functional blocks can be realized as either hardware or computer software, or a combination of both. It is not necessary for the functional blocks to be distinguished as in the above embodiment. For example, some functions may be performed by functional blocks other than the illustrated functional blocks. Furthermore, the illustrated functional blocks may be further divided into smaller functional sub-blocks. Furthermore, the order of the processes in the flowcharts described in the above embodiment can be changed as much as possible.
[0076] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]
[0077] 1...gyro sensor, 1A...analog circuit, 1D...digital circuit, 2...host device, 11...vibrator, 12...spring element, 13...sensor, 14...actuator, 20...capacitance-voltage conversion amplifier circuit, 30...analog-to-digital conversion circuit (ADC), 31_1...cosine wave generation circuit, 31_2...sine wave generation circuit, 32...multiplier, 33...low-pass filter, 40...digital signal processing circuit, 41...delay circuit, 42...Hilbert transform filter, 43...delay circuit, 44...Hilbert transform filter, 45...delay circuit, 50...CPU, 60...digital-to-analog conversion circuit (DAC), 70...RAM, 80...ROM, 90...interface circuit
Claims
1. a MEMS including a vibrator, a spring element supporting the vibrator, a sensor detecting a position of the vibrator on a two-dimensional plane, and an actuator applying a driving force to the vibrator; a processing circuit including a phase locked loop circuit for tracking a phase φ of the vibrator on a two-dimensional plane, the processing circuit calculating a driving force to be applied to the vibrator based on position information corresponding to the position detected by the sensor; Equipped with The processing circuitry [Equation 1] A gyro sensor that calculates the driving force based on the energy E, the quadrature amplitude Q, the vibration main axis angle θ, and the error δφ.
2. the processing circuit includes a finite impulse response (FIR) filter and a delay circuit; the FIR filter calculates the complex part of a signal obtained by subjecting the position information to a Hilbert transform; the delay circuit delays the position information and the reference phase by a phase delay that occurs in the complex part of the signal that has been Hilbert transformed by the FIR filter. The gyro sensor according to claim 1 .
3. a first conversion circuit that converts a first signal corresponding to the position detected by the sensor into a first voltage; a second conversion circuit that converts the first voltage converted by the first conversion circuit into the position information of a digital signal; Further provided with The gyro sensor according to claim 1 .
4. Further comprising a third conversion circuit; the processing circuit outputs a second signal corresponding to the calculated driving force; the third conversion circuit converts the second signal into a second voltage of an analog signal; the actuator applies the driving force to the vibrator in response to the second voltage; The gyro sensor according to claim 1 .
5. the processing circuit calculates the driving force based on the position information of the vibrator, and applies the driving force to the vibrator, thereby performing control to vibrate the vibrator linearly with a first amplitude. The gyro sensor according to claim 1 .
6. The sensor having opposing electrodes, detecting a capacitance of the electrode corresponding to the position of the vibrator, and outputting a signal corresponding to the capacitance; The gyro sensor according to claim 1 .
7. the processing circuit includes a DSP (Digital Signal Processor), The DSP executes the equations (A), (B), (C), and (D). The gyro sensor according to claim 1 .
8. the processing circuit includes an FPGA (Field-Programmable Gate Array); The FPGA implements the equations (A), (B), (C), and (D). The gyro sensor according to claim 1 .
9. the processing circuit includes a CPU (Central Processing Unit), The CPU executes the formulas (A), (B), (C), and (D). The gyro sensor according to claim 1 .
10. the gyro sensor includes a rate integrating gyroscope; The gyro sensor according to claim 1 .
11. a sensor for detecting the position of the vibrator on a two-dimensional plane; a processing circuit including a phase locked loop for tracking the phase φ of the vibrator on a two-dimensional plane, and performing arithmetic processing using position information according to the position; Equipped with The processing circuitry [Equation 2]
12. the processing circuit includes a finite impulse response (FIR) filter and a delay circuit; the FIR filter outputs the complex part of a signal obtained by subjecting the position information to a Hilbert transform; the delay circuit delays the position information and the reference phase by a phase delay that occurs in the complex part of the signal that has been Hilbert transformed by the FIR filter. The demodulation circuit according to claim 11.
13. a first conversion circuit that converts a first signal corresponding to the position detected by the sensor into a first voltage; a second conversion circuit that converts the first voltage converted by the first conversion circuit into the position information of a digital signal; Further provided with The demodulation circuit according to claim 11.
14. Further comprising a third conversion circuit; the processing circuit outputs a second signal corresponding to the calculated driving force; the third conversion circuit converts the second signal into a second voltage analog signal; The demodulation circuit according to claim 11.
15. The sensor having opposing electrodes, detecting a capacitance of the electrode corresponding to the position of the vibrator, and outputting a signal corresponding to the capacitance; The demodulation circuit according to claim 11.
16. the processing circuit includes a DSP (Digital Signal Processor), The DSP executes the equations (A), (B), (C), and (D). The demodulation circuit according to claim 11.
17. the processing circuit includes an FPGA (Field-Programmable Gate Array); The FPGA implements the equations (A), (B), (C), and (D). The demodulation circuit according to claim 11.
18. the processing circuit includes a CPU (Central Processing Unit), The CPU executes the formulas (A), (B), (C), and (D). The demodulation circuit according to claim 11.