Inertial sensor device
Patent Information
- Application Number
- JP2021210444
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2021-12-24
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2041-12-24
AI Technical Summary
Inertial sensor devices face a decrease in detection accuracy due to differences in detection timing among multiple sensors, particularly when dealing with digital signal data that is discretized on the time axis, leading to a deterioration in the S/N ratio.
An inertial sensor device with a plurality of inertial measurement modules, utilizing a timer unit and a synthesizing unit that interpolates and synthesizes output data at predetermined times using interpolated data from each module.
The device enhances detection accuracy by aligning time axes and reducing jitter noise, improving the S/N ratio through interpolation processing, thereby achieving high detection accuracy.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an inertial sensor device.
Background Art
[0002] An inertial sensor device including an inertial sensor module having a plurality of inertial sensors such as an acceleration sensor and an angular velocity sensor is known. The inertial sensor device is incorporated into various electronic devices and machines, or mounted on a moving body such as an automobile, and is used for monitoring inertial quantities such as angular velocity and acceleration. It is also applied to a monitoring system for structures such as bridges and elevated railways.
[0003] For example, Patent Document 1 proposes a technique for improving detection accuracy by providing a plurality of sensors typified by an IMU (Inertial Measurement Unit) and synthesizing observation values of the plurality of sensors.
Prior Art Documents
Patent Documents
[0004]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0005] However, in Patent Document 1, when there is a deviation in detection timing by a plurality of sensors, the S / N ratio deteriorates, which may lead to a decrease in detection accuracy. This is a finding based on the verification results of the inventors, and particularly, when the detection data from the sensors is a digital signal, it is significant because it is discretized on the time axis. That is, there has been a demand for an inertial sensor device capable of appropriately synthesizing detection data from a plurality of sensors and having high detection accuracy.
Means for Solving the Problems
[0006] An inertial sensor device according to one embodiment of the present invention is an inertial sensor device having a plurality of inertial measurement modules, including a first inertial measurement module, comprising: a timing unit; a storage unit that stores detection data from each of the plurality of inertial measurement modules in association with the time of the timing unit; and a synthesis processing unit that calculates interpolation data at a predetermined time from the detection data of the first inertial measurement module at least two times, and synthesizes output data at the predetermined time using the interpolation data for each of the plurality of inertial measurement modules, including the interpolation data. [Brief explanation of the drawing]
[0007] [Figure 1] A perspective view of the inertial sensor device according to Embodiment 1. [Figure 2] A transparent plan view of an inertial sensor device. [Figure 3] Exploded perspective view of an inertial sensor device. [Figure 4] Disassembled perspective view of the sensor module. [Figure 5] Front view of the circuit board. [Figure 6] Rear view of the circuit board. [Figure 7] Circuit block diagram of the sensor module. [Figure 8] A timing chart showing the relationship between detection data and output data. [Figure 9] A waveform diagram showing the error due to jitter in the detected data. [Figure 10] Circuit block diagram of the processing circuit in an inertial sensor device. [Figure 11] A perspective view showing different embodiments of an inertial sensor device. [Figure 12] A perspective view showing different embodiments of an inertial sensor device. [Figure 13] A perspective view showing different embodiments of an inertial sensor device. [Modes for carrying out the invention]
[0008] Embodiment 1 ***Overview of the Inertial Sensor Device*** Figure 1 is a perspective view showing an overview of the inertial sensor device. Figure 2 is a transparent plan view of the inertial sensor device. Figure 3 is an exploded perspective view of the internal configuration of the inertial sensor device. Embodiments of the present invention will be described below with reference to the drawings. The embodiments are illustrative examples of devices and methods for realizing the technical concept of the present invention. The technical concept of the present invention does not limit the materials, shapes, structures, arrangements, etc. of the components to those described below. In the drawings, identical or similar elements are denoted by the same or similar reference numerals, and redundant explanations are omitted. The drawings are schematic and may include differences from actual dimensions, relative proportions of dimensions, arrangements, structures, etc.
[0009] As shown in Figures 1 to 3, the inertial sensor device 1 according to the embodiment comprises, for example, a substrate 10, a first sensor module 2A, a second sensor module 2B, and a third sensor module 2C mounted on the substrate 10, a processing circuit 100, and a container 9. The inertial sensor device 1 is a composite sensor module equipped with a plurality of inertial sensors that detect acceleration in three axes and angular velocity around three axes. The inertial sensor device 1 detects the motion state of various objects, such as moving objects like vehicles, robots, and drones, as well as electronic devices like smartphones and tablet terminals. The motion state includes, for example, position, orientation, velocity, acceleration, angular velocity, etc.
[0010] As shown in Figures 1 and 2, the container 9 comprises a base 91 having a recess 911 that opens upward, and a lid 92 fixed to the base 91 so as to close the opening of the recess 911. The container 9 is generally rectangular in shape. The base 91 and the lid 92 define a housing space S inside the recess 911 that is sealed by the lid 92. The housing space S is a space for housing various components such as the substrate 10, the first sensor module 2A, the second sensor module 2B, the third sensor module 2C, and the processing circuit 100. The container 9 protects each component housed in the housing space S from dust, moisture, ultraviolet rays, impact, etc.
[0011] The base 91 and the lid 92 can be made of aluminum (Al). In addition, as the material of each of the base 91 and the lid 92, for example, metal materials such as Al alloy, zinc (Zn), stainless steel, various ceramics, various resin materials, and composite materials thereof can be adopted.
[0012] The inertial sensor device 1 includes a connector 93 attached to the side wall of the base 91 and a communication substrate 931 disposed in the accommodation space S. The connector 93 is a receptacle for making an electrical connection between the inside and the outside of the container 9. The communication substrate 931 has a circuit for processing communication between the inertial sensor device 1 and an external device.
[0013] The substrate 10 is a circuit board including various elements and wirings. The substrate 10 mounts the first sensor module 2A, the second sensor module 2B, the third sensor module 2C, the processing circuit 100, the internal connector 110, etc. The substrate 10 is fixed relative to the base 91, for example.
[0014] As shown in FIGS. 2 and 3, the first sensor module 2A and the second sensor module 2B are arranged along the X-axis on the lower surface of the substrate 10. The third sensor module 2C is disposed on the upper surface of the substrate 10 so as to overlap the first sensor module 2A when viewed from the direction along the Z-axis. The processing circuit 100 and the internal connector 110 are disposed on the upper surface of the substrate 10 so as to overlap the second sensor module 2B when viewed from the direction along the Z-axis. Thus, by efficiently arranging various components with respect to the area of the substrate 10 and the accommodation space S, miniaturization of the inertial sensor device 1 can be achieved.
[0015] The first sensor module 2A, the second sensor module 2B, and the third sensor module 2C are connected to the processing circuit 100 via the circuit board 10. The processing circuit 100 controls the operation of the first sensor module 2A, the second sensor module 2B, and the third sensor module 2C. The processing circuit 100 is connected to the communication board 931 via an internal connector 110 and wiring (not shown) connected to the internal connector 110.
[0016] The first sensor module 2A, the second sensor module 2B, and the third sensor module 2C have, for example, similar structures. In the following, any of the first sensor module 2A, the second sensor module 2B, and the third sensor module 2C will simply be referred to as "sensor module 2," and redundant explanations will be omitted. A sensor module is also called an inertial measurement module. The number of sensor modules 2 is not limited to 3; it may be 2 or 4 or more.
[0017] ***Sensor Module Configuration*** Figure 4 is an exploded perspective view of the sensor module. Figure 5 is a front view of the circuit board. Figure 6 is a rear view of the circuit board. As shown in Figure 4, the sensor module 2 comprises an outer case 21, an inner case 22, a connecting member 23, and a circuit board 24. The outer case 21 has a recess into which the inner case 22 is inserted. The outer case 21 and the inner case 22 are joined to each other by the connecting member 23 while housing and holding the circuit board 24. The sensor module 2 is square when viewed from above, i.e., along the c-axis as shown in Figure 4. The outer case 21 has, for example, screw holes 211, 212 provided at each of a pair of diagonally opposite corners on the top surface. The sensor module 2 can be fixed to the substrate 10 by screwing it using the screw holes 211, 212.
[0018] As shown in Figures 5 and 6, the circuit board 24 is equipped with a module connector 25, a first angular velocity sensor 26a, a second angular velocity sensor 26b, a third angular velocity sensor 26c, an acceleration sensor 27, a signal processing unit 28, and the like. The module connector 25 connects the sensor module 2 and the circuit board 10. The module connector 25 is exposed to the circuit board 10 through an opening 221 provided in the inner case 22, for example. The first angular velocity sensor 26a detects the angular velocity ωa around the a-axis. The second angular velocity sensor 26b detects the angular velocity ωb around the b-axis. The third angular velocity sensor 26c detects the angular velocity ωc around the c-axis. The acceleration sensor 27 detects acceleration Aa along the a-axis, acceleration Ab along the b-axis, and acceleration Ac along the c-axis, respectively. The three detection axes a, b, and c are defined for each sensor module 2.
[0019] The signal processing unit 28 is configured, for example, by an integrated circuit (IC). The signal processing unit 28 is connected to the first angular velocity sensor 26a, the second angular velocity sensor 26b, the third angular velocity sensor 26c, and the acceleration sensor 27 via the circuit board 24. The signal processing unit 28 is connected to the processing circuit 100 via the circuit board 24, module connector 25, board 10, etc.
[0020] The circuit board 24 is, for example, square in shape when viewed from a direction along the c-axis. When the four quadrants defined around the center O of the circuit board 24 are called the first quadrant Q1, the second quadrant Q2, the third quadrant Q3, and the fourth quadrant Q4, the acceleration sensor 27 is placed in the first quadrant Q1. As shown in Figure 3, the first sensor module 2A, the second sensor module 2B, and the third sensor module 2C are arranged so that their respective first quadrants Q1 are close to each other.
[0021] Specifically, in the example shown in Figure 3, the acceleration sensor 27A of the first sensor module 2A and the acceleration sensor 27C of the third sensor module 2C are arranged to overlap each other when viewed from the direction along the Z axis. The acceleration sensor 27A of the first sensor module 2A and the acceleration sensor 27B of the second sensor module 2B are arranged to overlap each other when viewed from the direction along the X axis. This makes it possible to minimize the difference in acceleration received by each of the acceleration sensors 27A, 27B, and 27C.
[0022] The module connector 25 is located on the upper surface 241 of the circuit board 24 in the second quadrant Q2 and the third quadrant Q3. The first angular velocity sensor 26a is located on the side surface of the circuit board 24 in the fourth quadrant Q4. The second angular velocity sensor 26b is located on the side surface of the circuit board 24 in the first quadrant Q1. The third angular velocity sensor 26c is located on the upper surface 241 of the circuit board 24 in the fourth quadrant Q4. The acceleration sensor 27 is located on the upper surface 241 of the circuit board 24 in the first quadrant Q1. The signal processing unit 28 is located on the lower surface 242 of the circuit board 24 in the third quadrant Q3. The screw hole 211 is located in the second quadrant Q2, and the screw hole 212 is located in the fourth quadrant Q4.
[0023] ***Sensor Module Configuration*** Figure 7 is a block diagram showing the circuit configuration of the sensor module. As shown in Figure 7, the sensor module 2 comprises an inertial sensor 20 including at least one of a first angular velocity sensor 26a, a second angular velocity sensor 26b, a third angular velocity sensor 26c, and an acceleration sensor 27, a signal processing unit 28, a timing unit 32, a communication unit 31, and a storage unit 30. The communication unit 31 includes a module connector 25 (Figure 5). The storage unit 30 stores, for example, various parameters used for correction in the signal processing unit 28.
[0024] The inertial sensor 20 outputs signals related to multiple detection axes to the signal processing unit 28. The signal processing unit 28 corrects the signals output from the inertial sensor 20 so that the multiple detection axes are orthogonal to each other. For example, multiple detection axes forming a three-dimensional orthogonal coordinate system are set for each sensor module 2. In addition, the signal processing unit 28 corrects the offset error and scale factor error included in the signals input from the inertial sensor 20. The corrected inertial data is then read out to the processing circuit 100 (Figure 10) of the inertial sensor device 1 at a timing synchronized with the output synchronization signal described later.
[0025] The timing unit 32 is a timer circuit equipped with an oscillator such as a crystal oscillator, and supplies a clock signal to the signal processing unit 28. The communication unit 31 is a digital interface circuit that enables bidirectional communication with the processing circuit 100 (Figure 3) of the inertial sensor device 1 via the module connector 25 (Figure 5).
[0026] ***Challenges of Conventional Technology*** Figure 8 is a timing chart showing the relationship between detection data and output data. The top of Figure 8 shows the timing of inertial data readout in the first sensor module 2A. First, in the first sensor module 2A, inertial data is acquired at the timing of the clock in its own timing unit 32. The sampling period is, for example, 2kHz, but is not limited to this; for example, 20kHz may also be used. Next, when the acquisition of inertial data for one sample is completed, an output synchronization signal (Data Ready) is output. The data ready signal is synchronized with the sampling period. Here, since each sensor module 2 is equipped with its own timing unit 32, the output synchronization signal is transmitted based on the clock of each sensor module 2. Specifically, in the first sensor module 2A, the output synchronization signal is output at the timing of the clock in its own timing unit 32. The same applies to the second sensor module 2B and the third sensor module 2C.
[0027] When the processing circuit 100 of the inertial sensor device 1 receives an output synchronization signal, it reads inertial data from the first sensor module 2A in synchronization with the output synchronization signal and stores it in the first storage unit 41 (Figure 10) as detected data Da[1]. At this time, the detected data Da[1] is stored in conjunction with the reading time Ta[1] of the data. The time Ta[1] is the reading time of the detected data Da[1] based on the clock of the timing unit 40 (Figure 10) of the processing circuit 100.
[0028] The same applies to the second sensor module 2B, where an output synchronization signal (Data Ready) is output at the timing of the clock in its own timing unit 32. Then, synchronized with the output synchronization signal, the inertial data is read out, and the detected data Db[1] and the time Tb[1] are stored as a pair. The same applies to the third sensor module 2C, where an output synchronization signal (Data Ready) is output at the timing of the clock in its own timing unit 32. Then, synchronized with the output synchronization signal, the inertial data is read out, and the detected data Dc[1] and the time Tc[1] are stored as a pair.
[0029] Here, as shown in Figure 8, there is a slight discrepancy between the read time Ta[1] of the detection data Da[1] from the first sensor module 2A and the read time Tb[1] of the detection data Db[1] from the second sensor module 2B. In other words, there is a time axis discrepancy between the read time Ta[1] and the read time Tb[1]. Similarly, jitter occurs between the read time Ta[1] and the read time Tc[1], and between the read time Tb[1] and the read time Tc[1]. This is because each sensor module 2 is equipped with its own timing unit 32, which outputs an output synchronization signal based on its own time. If we simply generate the output data at Ts[1] from Da[1], Db[1], and Dc[1], we will be using data that is shifted in the time axis from time Ts[1], in other words, data with jitter error, and as a result, errors will occur in the output data.
[0030] As an example, we consider the difference between the detection data Da[1] from the first sensor module 2A and the detection data Db[1] from the second sensor module 2B, i.e., the jitter error. Figure 9 shows the detection data 70 for the first sensor module 2A and the detection data 71 for the second sensor module 2B, as well as the difference between them (jitter error waveform 72), when the same 100Hz sine wave is input to both the first sensor module 2A and the second sensor module 2B, where there is a frequency difference between the clocks of the first sensor module 2A and the second sensor module 2B. The vertical axis represents time (sec), and the vertical axis represents the signal level. The waveform diagram in Figure 9 is based on simulation results, which were performed using a program on a personal computer (PC).
[0031] Error waveform 72 shows a beat waveform (beat pattern) that oscillates and gradually increases over time, then returns to zero, and then oscillates and increases again over time, reaching a maximum value of approximately 25%. If there is a frequency difference between the clock of the second sensor module 2B and the clock of the third sensor module 2C, a jitter error will similarly occur between the detection data of the second sensor module 2B and the detection data of the third sensor module 2C when the same 100Hz sine wave is input to both the second sensor module 2B and the third sensor module 2C. Therefore, it is clear that errors will occur if the detection data Da[1] from the first sensor module 2A, the detection data Db[1] from the second sensor module 2B, and the detection data Dc[1] from the third sensor module 2C are simply combined.
[0032] ***Circuit block configuration of the processing circuit in an inertial sensor device*** Figure 10 is a circuit block diagram of the processing circuit in an inertial sensor device. As shown in Figure 10, the processing circuit 100 of the inertial sensor device 1 consists of a timing unit 40, a first storage unit 41, a second storage unit 42, a synthesis processing unit 43, a communication unit 44, and the like.
[0033] The timing unit 40 is a timer circuit equipped with an oscillator such as a crystal oscillator, and supplies clock signals and time data to the synthesis processing unit 43, the first storage unit 41, and the like. The first storage unit 41 is a buffer memory configured to include RAM (Random Access Memory) and has three storage areas 41a, 41b, and 41c for each sensor module 2. Storage area 41a is a storage area for the detection data Da[n] of the first sensor module 2A and the read time Ta[n], and stores them as a pair of linked detection data (Ta[1],Da[1]). In addition, storage area 41a is reserved for storing multiple sets of detection data pairs in chronological order. For example, it is provided to store 5 sets of detection data pairs. However, it is not limited to 5 sets, and any number necessary for the interpolation process described later is acceptable. The same applies to storage areas 41b and 41c. In other words, the first storage unit 41 stores the detection data of each of the multiple sensor modules 2 in association with the time of the timing unit 40.
[0034] The second memory unit 42 is a main memory that includes ROM (Read Only Memory) and RAM, and stores programs executed by the synthesis processing unit 43, as well as related data. These programs include interpolation processing programs, which will be described later. The first memory unit 41 may also be provided as part of the second memory unit 42. The synthesis processing unit 43 is the control unit of the processing circuit 100 and is configured to include one or more processors. The synthesis processing unit 43 performs interpolation processing on the detection data from each sensor module 2 to synchronize it, and synthesizes the multiple interpolated detection data to generate output data. Details of the interpolation processing will be described later. The communication unit 44 is an interface circuit that outputs the output data from the synthesis processing unit 43 to an external device via the internal connector 110 (Figure 2) and the communication board 931 through the connector 93.
[0035] ***Interpolation processing of detected data - 1 (Past linear interpolation)*** Here, the interpolation processing of detected data by the synthesis processing unit 43 and the method of generating output data will be explained using Figures 8 and 10. Note that Main sync in Figure 8 represents the reference signal Ts[n] at the reference clock (2kHz) in the timing unit 40 of the processing circuit 100.
[0036] In this embodiment, the synthesis processing unit 43 uses the two most recent pairs of detection data prior to the timing of the reference signal Ts[3] to obtain interpolated data for the timing of the previous reference signal Ts[2] for the detection data from each sensor module 2 by linear interpolation. The timing of the reference signal Ts[3] corresponds to a predetermined time. Specifically, for the detection data pair (Ta[n], Da[n]) of the first sensor module 2A, linear interpolation is performed using equation (1) to obtain the interpolated data DaTs[2] at the timing of the reference signal Ts[2]. Similarly, for the second sensor module 2B, the interpolated data DbTs[2] is obtained using equation (2), and for the third sensor module 2C, the interpolated data DcTs[2] is obtained using equation (3).
[0037] DaTs[2]=Da[2]+{(Da[3]-Da[2]) / (Ta[3]-Ta[2])}*(Ts[2]-Ta[2]) ...Equation (1) DbTs[2]=Db[2]+{(Db[3]-Db[2]) / (Tb[3]-Tb[2])}*(Ts[2]-Tb[2]) ...Equation (2) DcTs[2]=Dc[2]+{(Dc[3]-Dc[2]) / (Tc[3]-Tc[2])}*(Ts[2]-Tc[2]) ...Equation (3)
[0038] In formula (1), the change is obtained from the difference between the detected data Da[3] and the previous detected data Da[2], and the slope is calculated by dividing this change by the difference between the readout time Ta[3] and the readout time Ta[2]. The amount of error change on the time axis is derived by multiplying this slope by the difference between the time of the reference signal Ts[2] and the readout time Ta[2]. Then, the base value of the detected data, Da[2], is added. This allows us to derive the interpolated data DaTs[2] at the timing of the reference signal Ts[2]. The same applies to equations (2) and (3). In other words, the synthesis processing unit 43 calculates the interpolated data by interpolating the detected data at two time points using the linear function equations (1) to (3). Furthermore, the timing of the reference signal Ts[3] as a predetermined time is later than the readout times Ta[2] and Ta[3] as two time points.
[0039] Next, we will explain how to combine the interpolated data DaTs[2], DbTs[2], and DcTs[2] obtained by equations (1) to (3). The synthesis processing unit 43 synthesizes three interpolated data using formula (4) to generate the output data DsTs[3].
[0040] DsTs[3]=(DaTs[2]+DbTs[2]+DcTs[2]) / 3...Equation (4)
[0041] As shown in equation (4), the synthesis processing unit 43 averages the interpolated data DaTs[2], interpolated data DbTs[2], and interpolated data DcTs[2] to generate the output data DsTs[3]. In other words, the synthesis processing unit 43 calculates the interpolated data at a predetermined time using the detection data of the first sensor module 2A at at least two time points, and synthesizes the output data at a predetermined time point using the interpolated data for each of the multiple sensor modules 2, including the interpolated data.
[0042] As described above, the inertial sensor device 1 of this embodiment provides the following advantages. The inertial sensor device 1 is an inertial sensor device having a plurality of sensor modules, including a first sensor module 2A as a first inertial measurement module, and comprises a timing unit 40, a first storage unit 41 as a storage unit that stores the detection data of each of the plurality of sensor modules 2 in association with the time of the timing unit 40, and a synthesis processing unit 43 that calculates interpolation data at a predetermined time from the detection data of the first sensor module 2A at at least two times, and synthesizes output data at a predetermined time using the interpolation data for each of the plurality of sensor modules 2, including the interpolation data.
[0043] According to this, for each inertial measurement module, synchronized interpolated data is calculated at a predetermined time. Then, the synthesis processing unit 43 takes the average value of the multiple interpolated data with aligned time axes and synthesizes them to generate output data. Therefore, by applying interpolation processing to multiple detection data that were discrete due to digital output, alignment in the time axis direction can be achieved, thereby reducing jitter noise that changes with time. Furthermore, when N detection data are synthesized, random noise can be reduced to 1 / √N, thereby improving detection accuracy. Therefore, it is possible to provide an inertial sensor device 1 that can appropriately synthesize detection data from multiple inertial measurement modules and has high detection accuracy.
[0044] Furthermore, the synthesis processing unit 43 calculates interpolated data by interpolating the detected data at two time points using equations (1) to (3), which are linear functions. According to this method, interpolated data can be calculated using a simple linear function, which reduces the storage capacity of the second storage unit 42 involved in the interpolation process and also reduces the computational load on the synthesis processing unit 43.
[0045] Furthermore, the timing of the reference signal Ts[3] as a predetermined time is later than the two time points, readout time Ta[2] and readout time Ta[3]. According to this method, interpolated data can be obtained by linear interpolation using two pairs of detection data from a time prior to a predetermined time.
[0046] Embodiment 2 ***Interpolation processing of detected data - 2 (Past quadratic interpolation)*** In the above embodiment, the interpolated data was calculated by linearly interpolating the detection data at two time points using a linear function. However, the method is not limited to this method; any method that calculates the interpolated data by interpolating the detection data at n+1 or more time points using an nth-degree function is acceptable. Hereafter, the same parts as in the above embodiment will be numbered, and redundant explanations will be omitted.
[0047] In this embodiment, the synthesis processing unit 43 uses the three most recent pairs of detection data prior to the timing of the reference signal Ts[3] to obtain interpolated data for the timing of the previous reference signal Ts[2] for each detection data from the sensor module 2 using quadratic interpolation. First, the basic formula for quadratic interpolation is equation (5).
[0048] Dn[y]=a1*(Tn[x]) 2 +b1*(Tn[x])+c1 …Equation (5)
[0049] Specifically, in the first sensor module 2A, the three most recent detection data pairs from the past (Ta[1],Da[1]), (Ta[2],Da[2]), and (Ta[3],Da[3]) are substituted into equation (5), resulting in equations (6) and (8).
[0050] Da[1]=a1*(Ta[1])*(Ta[1])+b1*(Ta[1])+c1...Equation (6) Da[2]=a1*(Ta[2])*(Ta[2])+b1*(Ta[2])+c1...Equation (7) Da[3]=a1*(Ta[3])*(Ta[3])+b1*(Ta[3])+c1...Formula (8) Then, solve the system of equations (6) and (8) to find the coefficients a1, b1, and c1.
[0051] Similarly, for the second sensor module 2B and the third sensor module 2C, substitute the three most recent detection data pairs from the past into equation (5) to obtain equations (9) through (11) and equations (12) through (14).
[0052] Db[1]=a2*(Tb[1])*(Tb[1])+b2*(Tb[1])+c2...Formula (9) Db[2]=a2*(Tb[2])*(Tb[2])+b2*(Tb[2])+c2...Formula (10) Db[3]=a2*(Tb[3])*(Tb[3])+b2*(Tb[3])+c2...Equation (11) Then, solve the system of equations (9) and (11) to find the coefficients a², b², and c².
[0053] Dc[1]=a3*(Tc[1])*(Tc[1])+b3*(Tc[1])+c3...Equation (12) Dc[2]=a3*(Tc[2])*(Tc[2])+b3*(Tc[2])+c3...Equation (13) Dc[3]=a3*(Tc[3])*(Tc[3])+b3*(Tc[3])+c3...Equation (14) Then, solve the system of equations (14) from equation (12) to find the coefficients a3, b3, and c3.
[0054] Next, the interpolated data DaTs[2], DbTs[2], and DcTs[2] at the timing of the reference signal Ts[2] are obtained by quadratic interpolation using equation (16) from equation (15). In other words, the synthesis processing unit 43 calculates the interpolated data by interpolating the detected data at the three time points using a quadratic function.
[0055] DaTs[2]=a1*Ts[2] 2 +b1*Ts[2]+c1 …Equation (15) DbTs[2]=a2*Ts[2] 2 +b2*Ts[2]+c2 …Equation (16) DcTs[2]=a3*Ts[2] 2 +b3*Ts[2]+c3 …Equation (17)
[0056] Next, we will explain how to combine the interpolated data DaTs[2], DbTs[2], and DcTs[2] obtained from equations (15) to (17). The synthesis processing unit 43 synthesizes three interpolated data using formula (18) and generates output data.
[0057] DsTs[3]=(DaTs[2]+DbTs[2]+DcTs[2]) / 3...Equation (18)
[0058] As shown in equation (18), the synthesis processing unit 43 averages the interpolated data DaTs[2], interpolated data DbTs[2], and interpolated data DcTs[2] to generate the output data DsTs[3].
[0059] As described above, the inertial sensor device 1 of this embodiment provides the following effects in addition to those of the above embodiment. The synthesis processing unit 43 calculates interpolated data by interpolating the detection data at three time points using a quadratic function. Specifically, the three most recent pairs of detection data from the past are substituted into equation (5), three simultaneous equations are set up, the coefficients are found by solving these simultaneous equations, and then the interpolated data at the timing of the reference signal one time prior to a predetermined time is calculated. According to this method, interpolated data is calculated using quadratic interpolation with a quadratic function, thus improving interpolation accuracy. Therefore, jitter noise can be reduced further. Therefore, it is possible to provide an inertial sensor device 1 that can appropriately synthesize detection data from multiple inertial measurement modules and has high detection accuracy.
[0060] Furthermore, the synthesis processing unit 43 calculates interpolated data by interpolating the detected data at n+1 or more time points using an n-th degree function. According to this method, interpolated data is calculated using n-th degree interpolation with an n-th degree function, thus improving interpolation accuracy. Therefore, jitter noise can be reduced further.
[0061] Embodiment 3 ***Interpolation Processing of Detected Data - 3 (Future Linear Interpolation)*** In Embodiment 1, the interpolation data at the timing of the reference signal Ts[2] one step prior to a predetermined time was described as being obtained by linear interpolation, but it is also possible to obtain the interpolation data at the timing of a predetermined time. Hereafter, the same parts as in the above embodiment will be numbered, and redundant explanations will be omitted.
[0062] In this embodiment, the synthesis processing unit 43 uses the two most recent past data points at the predetermined time Ts[3] timing to obtain the interpolated data at Ts[3] timing by linear interpolation. Specifically, for the detection data pair (Ta[n], Da[n]) of the first sensor module 2A, linear interpolation is performed using equation (19) to obtain the interpolated data DaTs[3] at the timing of the reference signal Ts[3]. Similarly, for the second sensor module 2B, the interpolated data DbTs[3] is obtained using equation (20), and for the third sensor module 2C, the interpolated data DcTs[3] is obtained using equation (21).
[0063] DaTs[3]=Da[3]+{(Da[3]-Da[2]) / (Ta[3]-Ta[2])}*(Ts[3]-Ta[3]) ...Equation (19) DbTs[3]=Db[3]+{(Db[3]-Db[2]) / (Tb[3]-Tb[2])}*(Ts[3]-Tb[3]) ...Equation (20) DcTs[3]=Dc[3]+{(Dc[3]-Dc[2]) / (Tc[3]-Tc[2])}*(Ts[3]-Tc[3]) ...Equation (21)
[0064] In formula (19), the change is obtained from the difference between the detected data Da[3] and the previous detected data Da[2], and the slope is calculated by dividing this change by the difference between the readout time Ta[3] and the readout time Ta[2]. The error change in the time axis is derived by multiplying this slope by the difference between the time of the reference signal Ts[3] and the readout time Ta[3]. Then, the base value of the detected data, Da[3], is added. This leads to the deriving of the interpolated data DaTs[3] at the timing of the reference signal Ts[3]. The same applies to equations (20) and (21).
[0065] Then, the synthesis of the three interpolated data obtained from equation (19) to equation (21) is performed using equation (22). Specifically, the synthesis processing unit 43 averages the three interpolated data to generate the output data DsTs[3].
[0066] DsTs[3]=(DaTs[3]+DbTs[3]+DcTs[3]) / 3...Equation (22)
[0067] As described above, the inertial sensor device 1 of this embodiment provides the following effects in addition to the effects of Embodiment 1. The synthesis processing unit 43 uses the two most recent past data points at the predetermined time Ts[3] timing to obtain the interpolated data at Ts[3] timing by linear interpolation. According to this method, interpolated data at the predetermined time Ts[3] timing can be derived without delay. Therefore, jitter noise can be reduced without delay.
[0068] Embodiment 4 ***Interpolation Processing of Detected Data - 4 (Future Quadratic Interpolation)*** In Embodiment 2, the interpolation data at the timing of the reference signal Ts[2] one step prior to a predetermined time was described as being obtained by quadratic interpolation, but it is also possible to obtain the interpolation data at the timing of a predetermined time. Hereafter, the same parts as in the above embodiment will be numbered, and redundant explanations will be omitted.
[0069] In this embodiment, the synthesis processing unit 43 uses the three most recent pairs of detection data prior to the timing of the reference signal Ts[3] to obtain interpolated data at a predetermined time, which is the timing of the reference signal Ts[3], using quadratic interpolation. The basic formula for quadratic interpolation is the equation (5) described above.
[0070] First, the process from equation (6) to equation (14) is the same as described in Embodiment 2. Specifically, for each sensor module 2, the three most recent pairs of detection data from the past are substituted into equation (5) to set up a system of three equations, and these equations are solved to obtain three coefficients for each sensor module 2.
[0071] Furthermore, this embodiment differs from Embodiment 2 in that the interpolated data DaTs[3], DbTs[3], and DcTs[3] at the timing of the reference signal Ts[3] are obtained by quadratic interpolation from equation (23) to equation (25).
[0072] DaTs[3]=a1*Ts[3] 2 +b1*Ts[3]+c1…Formula (23) DbTs[3]=a2*Ts[3] 2 +b2*Ts[3]+c2…Equation (24) DcTs[3]=a3*Ts[3] 2 +b3*Ts[3]+c3 …Equation (25)
[0073] Next, the interpolated data DaTs[3], DbTs[3], and DcTs[3] obtained from equation (23) using equation (25) are combined using equation (26) to generate the output data DsTs[3].
[0074] DsTs[3]=(DaTs[3]+DbTs[3]+DcTs[3]) / 3...Equation (26)
[0075] As shown in equation (26), the synthesis processing unit 43 averages the interpolated data DaTs[3], interpolated data DbTs[3], and interpolated data DcTs[3] to generate the output data DsTs[3].
[0076] As described above, the inertial sensor device 1 of this embodiment provides the following effects in addition to the effects of Embodiment 2. The synthesis processing unit 43 uses the three most recent past data points at the predetermined time Ts[3] timing to obtain the interpolated data at Ts[3] timing by quadratic interpolation. According to this method, interpolated data at the predetermined time Ts[3] timing can be derived without delay. Therefore, jitter noise can be reduced without delay.
[0077] Embodiment 5 ***Different forms of inertial sensor devices*** Figure 11 is a perspective view showing a different embodiment of the inertial sensor device and corresponds to Figure 3. Figure 12 is a perspective view showing a different embodiment of the inertial sensor device and corresponds to Figure 3. Figure 13 is a perspective view showing a different embodiment of the inertial sensor device and corresponds to Figure 3.
[0078] The configuration of the inertial sensor device 1 is not limited to the configurations described in Figures 1 to 3, and may be of a different configuration. For example, as shown in Figure 11, the inertial sensor device 1A of this embodiment includes a first sensor module 2A, a second sensor module 2B, and a third sensor module 2C stacked in a direction toward each other in one direction, i.e., along the Z-axis. The inertial sensor device 1A further includes four substrates 10A, 10B, 10C, and 10D, and a processing circuit 100 and an internal connector 110 mounted on each substrate 10D. The substrates 10A to 10D are fixed relative to each other. The first sensor module 2A is mounted on substrate 10A. The second sensor module 2B is mounted on substrate 10B. The third sensor module 2C is mounted on substrate 10C.
[0079] The first sensor module 2A, the second sensor module 2B, and the third sensor module 2C are connected to the processing circuit 100 in a daisy-chain configuration using multiple cables 4a, 4b, and 4c. Cables 4a, 4b, and 4c connect the first sensor module 2A, the second sensor module 2B, the third sensor module 2C, and the processing circuit 100, for example, via connectors mounted on circuit boards 10A, 10B, and 10C. This series connection of each sensor module increases design flexibility and allows for easy increase in the number of sensor modules. This further improves the signal-to-noise ratio of the output signal.
[0080] Furthermore, as shown in Figure 12, the inertial sensor device 1B of this embodiment includes a first sensor module 2A, a second sensor module 2B, and a third sensor module 2C arranged on the same plane. The inertial sensor device 1B includes substrates 10A to 10D arranged on a single plane along the XY plane. Similar to the example shown in Figure 11, the first sensor module 2A, the second sensor module 2B, and the third sensor module 2C are connected to the processing circuit 100 in a daisy-chain configuration by multiple cables 4a, 4b, and 4c. Therefore, design flexibility is improved, and the number of sensor modules can be easily increased.
[0081] Furthermore, as shown in Figure 13, the inertial sensor device 1C of this embodiment includes a single substrate 10E instead of multiple substrates 10A to 10D arranged on the same plane. The first sensor module 2A, the second sensor module 2B, the third sensor module 2C, the processing circuit 100, and the internal connector 110 are mounted on substrate 10E. Substrate 10E has wiring connecting the first sensor module 2A, the second sensor module 2B, and the third sensor module 2C to the processing circuit 100. The processing circuit 100 is connected in parallel to each sensor module 2, for example. This allows for more efficient use of communication capacity compared to series wiring.
[0082] These inertial sensor devices 1A, 1B, and 1C can also obtain the same effects and advantages as those of the embodiments described above. [Explanation of Symbols]
[0083] 1…Inertial sensor device, 1A~1C…Inertial sensor device, 2…Sensor module, 2A…First sensor module, 2B…Second sensor module, 2C…Third sensor module, 4a~4c…Cable, 9…Container, 10…Circuit board, 10A~10E…Circuit board, 20…Inertial sensor, 21…Outer case, 22…Inner case, 23…Bonding member, 24…Circuit board, 25…Module connector, 26a…First angular velocity sensor, 26b…Second angular velocity sensor, 26c…Third angular velocity sensor, 27…Accelerometer -, 27A~27C...accelerometer, 28...signal processing unit, 30...storage unit, 31...communication unit, 32...timing unit, 40...timing unit, 41...first storage unit, 41a...storage area, 41b...storage area, 42...second storage unit, 43...combination processing unit, 44...communication unit, 70...reference data waveform, 71...detection data waveform, 72...error waveform, 91...base, 92...lid, 93...connector, 100...processing circuit, 110...internal connector, 211,212...screw holes, 221...opening, 241...top surface, 242...bottom surface, 911...recess, 931...communication board.
Claims
1. An inertial sensor device having a plurality of inertial measurement modules including a first inertial measurement module. There was, A timing section; The detection data of each of the plurality of inertial measurement modules is associated with the time of the clock unit. A storage unit for storing the The detection data of the first inertial measurement module at at least two times is used to determine a predetermined Calculating interpolated data at a time, and storing the interpolated data in the plurality of inertial measurement modules. A synthesis process of synthesizing output data at the predetermined time using the interpolation data for each A part, Inertial sensor device.
2. The synthesis processing unit interpolates the detection data at n+1 or more times using an n-th order function. The interpolation data is calculated by The inertial sensor device according to claim 1 .
3. The synthesis processing unit interpolates the detection data at the two times using a linear function. The interpolation data is calculated by the above. The inertial sensor device according to claim 1 .
4. The predetermined time is a time later than the two times. The inertial sensor device according to claim 3 .
5. The synthesis processing unit performs interpolation of the detection data at three times using a quadratic function. The interpolation data is calculated by the above. The inertial sensor device according to claim 1 .
6. An inertial sensor device having a plurality of inertial measurement modules including a first inertial measurement module. There was, A timing section; The detection data of each of the plurality of inertial measurement modules is associated with the time of the clock unit. A storage unit for storing the The detection data of each of the plurality of inertial measurement modules is subjected to an interpolation process to synchronize the data. a synthesis processing unit that synthesizes the plurality of interpolated detection data to generate output data; , comprising Inertial sensor device.
