JP2015108595A - Sensor output correction device, sensor output correction method and sensor output correction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sensor output correction device capable of correcting output of a second inertia sensor having a dynamic range greater than that of a first inertia sensor by using output of the first inertia sensor, and further to provide a sensor output correction method and a sensor output correction program.SOLUTION: A sensor output correction device includes: an operation section including a difference calculation section 110 for calculating a difference ΔSO between two kinds of output synchronized with each other from a first inertia sensor 1 and a second inertia sensor 2 out of three kinds of output, the three kinds of output comprising received output SO1 from the first inertia sensor 1 having a first dynamic range, received output SO2 from the second inertia sensor 2 having a second dynamic range greater than that of the first dynamic range and output of a dynamic range where the first dynamic range and the second dynamic range overlap with each other; and a correction section 120 for correcting the output SO2 from the second inertia sensor 2 on the basis of the difference ΔSO.

Description

  The present invention relates to a sensor output correction device that receives outputs from two sensors having different dynamic ranges, a sensor output correction method, a sensor output correction program, and the like.

  For example, in an acceleration sensor as an inertial sensor, the dynamic range that the sensor can detect, in other words, the maximum acceleration and the detection accuracy are in a trade-off relationship. Therefore, for example, when an acceleration sensor with a narrow dynamic range is used, it is possible to detect a gentle movement with high accuracy, but for a fast movement, the sensor characteristics become saturated and an accurate movement is captured. It becomes impossible to do. On the other hand, when a sensor with a wide dynamic range is used, fast motion can be detected, but there is a problem that fine data cannot be acquired for slow motion.

  Therefore, in the invention described in Patent Document 1, by using an acceleration sensor having a wide dynamic range of ± 24 G and an acceleration sensor having a narrow dynamic range of ± 5, it is possible to detect a high-speed motion from a low speed. .

JP 2011-242323 A

  When two acceleration sensors are used in combination, it is necessary to perform processing such as initial bias setting and temperature correction for each of the two acceleration sensors, which increases costs.

  In some aspects of the present invention, the output of the second inertial sensor having a dynamic range wider than that of the first inertial sensor is corrected using the output of the first inertial sensor that performs high-accuracy measurement by adjusting the bias. It is possible to provide a sensor output correction device, a sensor output correction method, and a sensor output correction program.

  (1) In one aspect of the present invention, an output from a first inertial sensor having a first dynamic range and an output from a second inertial sensor having a second dynamic range wider than the first dynamic range are received. And the difference which calculates the difference of the outputs which are the dynamic ranges which the said 1st dynamic range and the said 2nd dynamic range mutually overlap, and mutually synchronize the outputs from the said 1st inertial sensor and the said 2nd inertial sensor. The present invention relates to a sensor output correction device including a calculation unit including a calculation unit and a correction unit that corrects an output from the second inertial sensor based on a calculation result of the calculation unit.

  In one aspect of the present invention, if the first inertial sensor is bias-adjusted, the output of the first inertial sensor can be used as it is in the first dynamic range. When the second inertial sensor is not adjusted, the output of the second inertial sensor is corrected on the assumption that the outputs of the first and second inertial sensors are equal. That is, the difference between the output of the second inertial sensor and the output of the first inertial sensor can be estimated as the bias value of the second inertial sensor. Here, the bias value is a generic term for errors including zero bias in the initial state where the angular velocity and acceleration are zero, and random drift caused by external factors such as power supply fluctuation and temperature fluctuation. Therefore, the output of the second inertial sensor corrected based on the estimated bias value obtained from the difference is expected to be an appropriate value as a physical quantity in a range exceeding the first dynamic range. That is, the second inertial sensor can be used without adjustment.

  (2) In one aspect of the present invention, the difference calculation unit is configured such that the difference between the minimum value in the output detected by the first inertial sensor and the output of the second inertial sensor synchronized with the minimum value. Can be calculated.

  The difference between the minimum value of the output detected by the first inertial sensor and the output of the second inertial sensor synchronized with the minimum value can be estimated as the zero point bias value of the second inertial sensor, and the zero point estimation is performed. The zero point bias adjustment of the second inertial sensor can be performed using the bias value.

  (3) In one aspect of the present invention, the difference calculation unit is configured such that the difference between the maximum value among the outputs detected by the first inertial sensor and the output of the second inertial sensor synchronized with the maximum value. Can be calculated.

  The output detected by the first inertia sensor may become a maximum value near the upper limit of the first dynamic range. The difference between this maximum value and the output of the second inertial sensor synchronized with the maximum value may also be maximum. In that case, the continuity between the output of the first inertial sensor in the first dynamic range and the output of the second inertial sensor in the second dynamic range exceeding the upper limit of the first dynamic range is the upper limit of the first dynamic range. This is ensured by using the difference between nearby outputs.

  (4) In one aspect of the present invention, the difference calculation unit includes a difference between outputs at time t1 in outputs detected by the first inertia sensor and the second inertia sensor, and at time t2. An average value with the difference between outputs can be calculated as the difference.

  Since the bias value at time t1 and the estimated bias value at time t2 do not necessarily match, the average value of the difference between outputs at time t1 and the difference between outputs at time t2 is defined as a difference. The output accuracy of the two-inertia sensor may increase.

  (5) In one aspect of the present invention, the output of the first inertial sensor at the time t1 can be a minimum value, and the output of the first inertial sensor at the time t2 can be a maximum value.

  Thus, averaging the estimated bias values of the second inertial sensor may increase the output accuracy of the corrected second inertial sensor.

  (6) In one aspect of the present invention, the difference calculation unit can repeatedly calculate the difference based on a calculation instruction signal and update the difference to the latest.

  Since the bias changes depending on the environment such as temperature, it is not preferable to use the same estimated bias value for a long time, and it is preferable to update the estimated bias value. The calculation instruction signal that is the update timing can be generated, for example, every predetermined time.

  (7) In one aspect of the present invention, the difference calculation unit includes a difference between outputs at time t1 in outputs detected by the first inertia sensor and the second inertia sensor, and at time t2. A difference between outputs is calculated, and the calculation unit further includes a coefficient calculation unit that calculates a coefficient K using the two differences from the difference calculation unit, and a coefficient K as an output of the second inertial sensor. The correction unit can correct the output of the second inertial sensor based on the output from the multiplication unit and the output of the first inertial sensor.

  The difference (estimated bias value) between the output from the first inertia sensor and the output from the second inertia sensor is not constant but changes. As in one aspect of the present invention, by calculating a difference (estimated bias value) between a value obtained by multiplying the output of the second inertial sensor by a coefficient K and the output of the first inertial sensor, a change in the estimated bias value is reduced. can do.

(8) In one aspect of the present invention, the output at time t1 in the output detected by the first inertial sensor is S1 (t1), the output at time t2 is S1 (t2), and the first Among the outputs detected by the two inertial sensors, when the output at time t1 is SO2 (t1) and the output at time t2 is S1 (t2),
K = [S1 (t1) -S1 (t2)] / [SO2 (t1) -SO2 (t2)]
It can be.

  The coefficient K can be a ratio of a change between the times t1 and t2 of the first inertial sensor with respect to a change between the times t1 and t2 of the second inertial sensor. The output SO2 ′ (t3) after correction of the output SO2 (t3) of the second inertial sensor in the range exceeding the overlapping range at time t3 is, for example, output SO2 ′ (t3) = K · SO2 (t3) + [S1 (t1) −K · SO2 (t1)] or output SO2 ′ (t3) = K · SO2 (t3) + [S1 (t2) −K · SO2 (t2)] or the like.

  (9) In one aspect of the present invention, among the outputs detected by the first inertial sensor, the output S1 (t1) at the time t1 is one of a maximum value and a minimum value, and at the time t2. The output S1 (t2) can be the other of the maximum value and the minimum value.

  In this way, the coefficient K can be the ratio of the maximum change between the times t1 and t2 of the first inertial sensor to the maximum change between the times t1 and t2 of the second inertial sensor.

  (10) In one aspect of the present invention, the first inertial sensor and the second inertial sensor can be attached to substantially the same location.

  As described above, when the first and second inertial sensors are mounted at substantially the same location, the outputs of the first and second inertial sensors become equal. Thus, since the difference between the output of the second inertia sensor and the output of the first inertia sensor can be estimated as the bias value of the second inertia sensor, the second inertia sensor can be used without adjustment.

  (11) Another aspect of the present invention receives an output from a first inertial sensor having a first dynamic range and an output from a second inertial sensor having a second dynamic range wider than the first dynamic range. And a step of calculating a difference between outputs of the first dynamic range and the second dynamic range that overlap with each other and that are synchronized with each other from the first inertia sensor and the second inertia sensor. And a step of correcting the output from the second inertial sensor based on the difference.

  In another aspect of the present invention, as in one aspect of the present invention, the output of the second inertial sensor is corrected based on the estimated bias value obtained by the difference between the output of the second inertial sensor and the output of the first inertial sensor. Thus, the correction value can be used as a physical quantity in a range exceeding the first dynamic range.

  (12) According to still another aspect of the present invention, an output from a first inertial sensor having a first dynamic range and an output of a second inertial sensor having a second dynamic range wider than the first dynamic range are provided. The computer performs a procedure for calculating a difference between outputs that are output in a dynamic range that overlaps each other and that are synchronized with each other, and a procedure for correcting the output from the second inertial sensor based on the difference. The present invention relates to a sensor output correction program.

  A sensor output correction program according to another aspect of the present invention can cause a computer to execute the operation of the sensor output correction apparatus according to one aspect of the present invention. This program may be stored in the sensor output correction device from the beginning, stored in a storage medium and installed in the sensor output correction device, or downloaded from a server to a communication terminal of the sensor output correction device through a network. May be.

It is a conceptual diagram which shows roughly the structure of the golf swing analyzer which concerns on one Embodiment of this invention. It is a schematic block diagram which shows an example of a sensor output correction apparatus. It is the characteristic view which plotted output SO1 of the 1st inertial sensor, output SO2 of the 2nd inertial sensor, and those difference (SO1-SO2). It is a figure for demonstrating the operation | movement which calculates a difference repeatedly and makes the newest difference an estimated bias value. It is a schematic block diagram which shows another example of a sensor output correction apparatus. It is a characteristic view which shows the example which multiplied the coefficient K to the output of the 2nd inertial sensor, and optimized the difference OS as an estimation bias device. It is the characteristic view which plotted the output of the 2nd inertial sensor corrected using the coefficient K for the output of the 2nd inertial sensor, and the output of the 1st inertial sensor.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. The present embodiment described below does not unduly limit the contents of the present invention described in the claims, and all the configurations described in the present embodiment are essential as means for solving the present invention. Not necessarily.

(1) Configuration of Motion Analysis Device FIG. 1 schematically shows a configuration of a motion analysis device (for example, a golf swing analysis device) 11 according to an embodiment of the present invention. A first inertial sensor 1 and a second inertial sensor 2 are connected to the motion analysis device 11. The first inertial sensor 1 is mounted with an acceleration sensor having a first dynamic range of ± 24G, for example. The second inertial sensor 2 is mounted with an acceleration sensor having a second dynamic range that is wider than the first dynamic range, for example, ± 70 G. In addition to the acceleration sensor, for example, a gyro sensor is incorporated in each of the first inertial sensor 1 and the second inertial sensor 2. The acceleration sensor can individually detect acceleration in the three axes x, y, and z directions orthogonal to each other. The gyro sensor can individually detect angular velocities around the three axes x, y, and z orthogonal to each other. The first inertial sensor 1 and the second inertial sensor 2 each output a detection signal. In each detection signal, acceleration and angular velocity are specified for each individual axis. The acceleration sensor and the gyro sensor detect acceleration and angular velocity information with relatively high accuracy.

  The first inertial sensor 1 is attached to a golf club (exercise tool) 13. The golf club 13 includes a shaft 13a and a grip 13b. The grip 13b is gripped by hand. The grip 13b is formed coaxially with the axis of the shaft 13a. A club head 13c is coupled to the tip of the shaft 13a. Desirably, the first inertial sensor 1 is attached to the shaft 13 a or the grip 13 b of the golf club 13. The first inertial sensor 1 may be fixed to the golf club 13 so as not to be relatively movable. The second inertial sensor 2 is fixed on substantially the same location as the first inertial sensor 1, for example, on the first inertial sensor 1. The second inertia sensor 2 may be fixed to the first inertia sensor 1 so as not to be relatively movable.

  Here, when the first inertial sensor 1 is attached, assuming that the local coordinate system of the first inertial sensor 1 is x1, y1, z1, for example, the y1 axis is set parallel to the axis of the shaft 13a. For example, the x1 axis, which is another detection axis of the first inertial sensor 1, is set parallel to the target direction A that intersects the face surface of the club head 13c. For example, the z1 axis, which is another detection axis of the first inertial sensor 1, is set to be orthogonal to the x1 axis and the y1 axis.

  On the other hand, when attaching the second inertial sensor 2 to the local coordinate system of the second inertial sensor 2 as x2, y2, and z2, the y2 axis is parallel to the y1 axis, the x2 axis is parallel to the x1 axis, and the z2 axis is z1. Set parallel to the axis.

  The golf swing analysis device 11 includes an arithmetic processing circuit 14. A first inertial sensor 1 and a second inertial sensor 2 are connected to the arithmetic processing circuit 14. In connection, a predetermined interface circuit 15 is connected to the arithmetic processing circuit 14. The interface circuit 15 may be connected to the first inertial sensor 1 and the second inertial sensor 2 by wire or may be connected to the first inertial sensor 1 and the second inertial sensor 2 by wireless. Detection signals are supplied to the arithmetic processing circuit 14 from the first inertial sensor 1 and the second inertial sensor 2. The arithmetic processing circuit 14 is provided with a sensor output correction device 100 that corrects the output of the second inertial sensor 2. The sensor output correction device 100 will be described later.

  A storage device 16 is connected to the arithmetic processing circuit 14. The storage device 16 can store, for example, a golf swing analysis software program (motion analysis program) 17 and related data. The arithmetic processing circuit 14 executes a golf swing analysis software program 17 including a sensor output correction program to realize a golf swing analysis method. The storage device 16 can include a DRAM (Dynamic Random Access Memory), a mass storage device unit, a non-volatile memory, and the like. For example, in the DRAM, the golf swing analysis software program 17 is temporarily held when the golf swing analysis method is executed. A golf swing analysis software program 17 and data are stored in a mass storage unit such as a hard disk drive (HDD). The nonvolatile memory stores a relatively small capacity program such as BIOS (basic input / output system) and data.

  An image processing circuit 18 is connected to the arithmetic processing circuit 14. The arithmetic processing circuit 14 sends predetermined image data to the image processing circuit 18. A display device 19 is connected to the image processing circuit 18. In connection, a predetermined interface circuit (not shown) is connected to the image processing circuit 18. The image processing circuit 18 sends an image signal to the display device 19 according to the input image data. An image specified by the image signal is displayed on the screen of the display device 19. The display device 19 is a liquid crystal display or other flat panel display. Here, the arithmetic processing circuit 14, the storage device 16, and the image processing circuit 18 are provided as a computer device, for example.

  An input device 21 is connected to the arithmetic processing circuit 14. The input device 21 includes at least alphabet keys and numeric keys. Character information and numerical information are input from the input device 21 to the arithmetic processing circuit 14. The input device 21 may be composed of a keyboard, for example. The combination of the computer device and the keyboard may be replaced with, for example, a smartphone, a mobile phone terminal, a tablet PC (personal computer), or the like.

(2) Sensor Output Correction Device A sensor output correction device 100 provided in the arithmetic processing circuit 14 shown in FIG. 1 will be described with reference to FIG. The output of the first inertial sensor 1 is SO1, and the output of the second inertial sensor 2 is SO2. The sensor output correction device 100 includes a calculation unit 110 and a correction unit 120. In the present embodiment, the calculation unit 110 is the outputs SO1 and SO2 of the overlapping range (that is, ± 24G) of the first dynamic range and the second dynamic range, and is received from the first inertial sensor 1 and the second inertial sensor 2. The difference calculation unit 110A that calculates the difference ΔSO = SO1-SO2 between outputs synchronized with each other. The correcting unit 120 corrects the output SO2 from the second inertial sensor 2 to the corrected output SO2 ′ based on the difference ΔSO.

  Here, the first inertial sensor 1 uses a high-performance inertial sensor in which calibration is performed and the bias value is adjusted. The bias value is a generic term for errors including zero bias in an initial state where the angular velocity and acceleration are zero, and random drift caused by external factors such as power supply fluctuation and temperature fluctuation. Therefore, it is not necessary to correct the output SO1 of the first inertial sensor 1 by the sensor output correction device 100. That is, since the first inertial sensor 1 is bias-adjusted, the output SO1 of the first inertial sensor 1 can be used as it is in the first dynamic range (± 24G). However, in the range exceeding the first dynamic range (± 24G), the output SO1 of the first inertial sensor 1 is saturated, so that the output SO2 of the second inertial sensor 2 needs to be used instead.

  On the other hand, the second inertial sensor 2 is an inertial sensor without a bias, although the dynamic range is wider than that of the first inertial sensor 1. Therefore, in this embodiment, the correction target of the sensor output correction device 100 is the output SO2 of the second inertial sensor 2.

  As a premise of the bias adjustment principle of the second inertial sensor 2, the outputs SO1 and SO2 of the first and second inertial sensors attached at substantially the same location are in the overlapping range (± 24G) of the first and second dynamic ranges. Should be equal. The sensor output correction device 100 corrects the output SO2 of the second inertial sensor 2 on the precondition of this. The difference calculation unit 110A calculates a difference ΔSO = SO1−SO2 between outputs received from the first inertial sensor 1 and the second inertial sensor 2 in the overlap range (± 24G). That is, the difference ΔSO between the output SO2 of the second inertial sensor 2 and the output SO1 of the first inertial sensor 1 can be estimated as the bias value of the second inertial sensor 2 from the above-described preconditions. This is because the output SO2 'of the overlapping range (± 24G) of the second inertial sensor 2 corrected by the difference ΔSO can be regarded as being equal to the output SO1 of the first inertial sensor 1. Therefore, the output SO2 'of the second inertial sensor 2 corrected based on the estimated bias value obtained by the difference ΔSO is expected as a physical quantity that is bias-adjusted even in a range that exceeds the first dynamic range (± 24G). That is, the second inertial sensor 2 can be used without adjustment.

(2-1) Sensor output correction example 1
FIG. 3 plots the output SO1 of the first inertial sensor 1 and the output SO2 of the second inertial sensor 2 with the horizontal axis representing time (msec) and the vertical axis representing acceleration (G). FIG. 3 shows an output example when the golf club 13 is swung so that both the outputs SO1 and SO2 increase in acceleration over time. However, the output SO1 of the first inertial sensor 1 is saturated when it exceeds the upper limit (+ 24G in FIG. 2) of the first dynamic range (± 24G).

  The output SO2 of the second inertial sensor 2 rises with time without being saturated even if it exceeds the upper limit (+ 24G in FIG. 3) of the first dynamic range (± 24G). However, even in the first dynamic range (± 24G), the output SO2 of the second inertial sensor 2 does not coincide with the output SO1 of the first inertial sensor 1 (the output SO2 is lower in FIG. 2).

  FIG. 3 further plots a difference ΔSO (= SO1−SO2) that is an output from the difference calculation unit 110A shown in FIG. In FIG. 2, the difference ΔSO is not constant, but the difference at any one time within the first dynamic range (± 24G) or the average value of the differences at a plurality of times can be used as the estimated bias adjustment value. This is because the target for correcting the output of the second inertial sensor 2 is a range exceeding the first dynamic range (± 24 G), and the output SO1 of the first inertial sensor 1 is saturated in that range.

  Therefore, the difference calculation unit 110A, based on the calculation instruction signal synchronized with the time t1, the minimum value SO1 (t1) at the time t1 in the output detected by the first inertial sensor 1, and the minimum value The difference [SO1 (t1) −SO2 (t1)] from the output SO2 (t1) of the synchronized second inertial sensor 2 can be calculated. In this case, the correction unit 120 uses the difference [SO1 (t1) −SO2 (t1)] as the output SO2 (t3) of the second inertial sensor 2 at time t3 when the range exceeds the first dynamic range (± 24G). As the corrected output SO2 ′ (t3) corrected in step S2, “SO2 ′ (t3) = output SO2 (t3) + [SO1 (t1) −SO2 (t1)]” is calculated.

  Difference [SO1 (t1) −SO2 (t1) between the minimum value SO1 (t1) in the output detected by the first inertial sensor 1 and the output SO2 (t1) of the second inertial sensor synchronized with the minimum value ]] Can be estimated as the zero point bias value of the second inertial sensor 2, and the zero point bias adjustment of the second inertial sensor 2 can be performed using the zero point estimated bias value.

  However, in the case of the characteristics shown in FIG. 3, if the difference [SO1 (t1) −SO2 (t1)] at time t1 is the estimated bias value, the first inertial sensor at time t2 close to the upper limit of the first dynamic range. 1 and the corrected output SO2 ′ (t3) = output SO2 (t3) + [SO1 (t1) −SO2 (t1)] of the second inertial sensor 2 are deteriorated.

  In such a case, the difference calculation unit 110A, based on the calculation instruction signal synchronized with the time t2, the maximum value SO1 (t2) at the time t2 in the output detected by the first inertial sensor, and its maximum The difference [SO1 (t2) −SO2 (t2)] from the output SO2 (t2) of the second inertial sensor 2 synchronized with the value can be calculated as a bias estimated value. In this case, the correction unit 120 uses the difference [SO1 (t2) −SO2 (t2)] as the output SO2 (t3) of the second inertial sensor 2 at time t3 when the range exceeds the first dynamic range (± 24G). As the corrected output SO2 ′ (t3) corrected in step S2, SO2 ′ (t3) = output SO2 (t3) + [SO1 (t2) −SO2 (t2)] is calculated and obtained.

  In the example of FIG. 3, the output SO1 (t2) of the first inertial sensor 1 in the first dynamic range and the output of the second inertial sensor 2 in the second dynamic range exceeding the upper limit of the first dynamic range are continuous. This is ensured by using the difference between outputs at time t2 near the upper limit of the first dynamic range.

  As a modification, the difference calculation unit 110A selects outputs at different times from the outputs detected by the first inertial sensor 1 and the second inertial sensor 2, and for example, a time based on a calculation instruction signal synchronized with the time t1. Average value [SO1 (t1) −SO2 (t1)] / 2+ [SO1 (t2) of the difference between the outputs at t1 and the difference between the outputs at time t2 based on the calculation instruction signal synchronized with time t2, for example. -SO2 (t2)] / 2 can be calculated as a difference. In this case, the calculation unit 110 can incorporate a storage unit, a divider, and the like in addition to the subtractor.

  Since the bias value at time t1 and the estimated bias value at time t2 do not necessarily match, the average value of the difference between outputs at time t1 and the difference between outputs at time t2 is defined as a difference. The output accuracy of the two-inertia sensor 2 may increase. As in the example of FIG. 3, a calculation instruction synchronized with times t1 and t2, with the output of the first inertial sensor 1 at time t1 being the minimum value and the output of the first inertial sensor 1 at time t2 being the maximum value A signal may be selected.

  The difference calculation unit 110A can repeatedly calculate the difference ΔSO based on a calculation instruction signal synchronized with the times T1, T2, T3,... Shown in FIG.

  Since the bias changes depending on the environment such as temperature, it is not preferable to use the same estimated bias value for a long time, and it is preferable to update the estimated bias value. The calculation instruction signal that is the update timing can be generated, for example, every predetermined time.

(2-2) Sensor output correction example 2
The calculation unit 110 illustrated in FIG. 5 includes a difference calculation unit 111, a coefficient calculation unit 112, and a multiplication unit 113. Among the outputs detected by the difference calculation unit 111, the first inertial sensor 1 and the second inertial sensor 2, the difference ΔSO (t1) between outputs at time t1 and the difference ΔSO (t2) between outputs at time t2 ) And. Here, ΔSO (t1) = SO1 (t1) −SO2 (t1), and ΔSO (t2) = SO1 (t2) −SO2 (t2).

  The coefficient calculation unit 112 calculates a coefficient K by which the output of the second inertial sensor 2 is multiplied using the two differences ΔSO (t1) and ΔSO (t2) from the difference calculation unit 111.

  Here, the difference (estimated bias value) between the output from the first inertia sensor and the output from the second inertia sensor is not constant but changes. By calculating the difference (estimated bias value) between the value obtained by multiplying the output of the second inertial sensor by the coefficient K and the output of the first inertial sensor, the change in the estimated bias value can be reduced.

Of the outputs detected by the first inertial sensor 1, the output at time t1 is SO1 (t1), the output at time t2 is SO1 (t2), and the output detected by the second inertial sensor 2 is When the output at time t1 is S2 (t1) and the output at time t2 is SO1 (t2),
K = [SO1 (t1) -SO1 (t2)] / [SO2 (t1) -SO2 (t2)]
It can be.

  Thus, the coefficient K can be the ratio of the change between the times t1 and t2 of the first inertial sensor 1 with respect to the change between the times t1 and t2 of the second inertial sensor 2. The multiplier 113 multiplies the output SO2 of the second inertial sensor 2 by a coefficient K as shown in FIG.

  6, the horizontal axis is time (msec), the vertical axis is acceleration (G), and the output SO1 of the first inertial sensor 1 and the output SO2 of the second inertial sensor 2 are multiplied by a coefficient K, as in FIG. K · SO 2 is plotted. The difference ΔSO = SO1−K · SO2 between the output SO1 of the first inertial sensor 1 and K · SO2 plotted at the same time is substantially constant. Therefore, the difference ΔSO = SO1−K · SO2 shown in FIG. 6 is more excellent as the estimated bias value than the case where the difference ΔSO = SO1−SO2 shown in FIG. 3 is used.

  As shown in FIG. 5, a correction unit 121 to which the output of the calculation unit 110 is input is provided. Here, the corrected output SO2 ′ (t3) of the correction unit 121 when the output SO2 (t3) of the second inertial sensor in the range exceeding the overlapping range at time t3 is received is, for example, the output S2 '(T3) = K · SO2 (t3) + [SO1 (t1) −K · SO2 (t1)] or output S2 ′ (t3) = K · SO2 (t3) + [SO1 (t2) −K · SO2 ( t2)] and the like. When the corrected output SO2 ′ of the correction unit 121 is plotted from time t1, as shown in FIG. 7, it overlaps with the output OS1 of the first inertial sensor 1 in the first dynamic range of + 24G or less, and the first dynamic range in the second dynamic range. The value increases continuously from the upper limit of the output OS1 of the inertial sensor 1.

  As shown in FIG. 3, among the outputs detected by the first inertial sensor, the output SO1 (t1) at time t1 is one of the maximum value and the minimum value, and the output SO1 (t2) at time t2 is It can be selected to be the other of the maximum value and the minimum value. In this way, the coefficient K can be the ratio of the maximum change between the times t1 and t2 of the first inertial sensor 1 to the maximum change between the times t1 and t2 of the second inertial sensor 2.

  Although the present embodiment has been described in detail, those skilled in the art will readily understand that many modifications are possible without substantially departing from the novel matters and effects of the present invention. Therefore, all such modifications are included in the scope of the present invention. For example, a term described with a different term having a broader meaning or the same meaning at least once in the specification or the drawings can be replaced with the different term in any part of the specification or the drawings. Further, the configurations and operations of the first and second inertial sensors 1 and 2, the arithmetic processing circuit 14, the sensor output correction device 100, and the like are not limited to those described in the present embodiment, and various modifications can be made. The example of motion analysis to which the present invention is applied is not limited to golf. The present invention is not necessarily limited to motion analysis, and can be applied to a wide range of applications using the first and second inertial sensors 1 and 2 having different dynamic ranges.

  DESCRIPTION OF SYMBOLS 1 1st inertial sensor, 2nd 2nd inertial sensor, 11 motion analyzer (golf swing analyzer), 13 exercise tools, golf club, 13a shaft, 13b grip, 13c club head, 14 computer (arithmetic processing circuit), 17 motion Analysis program (golf swing analysis software program), 100 sensor output correction device, 110 calculation unit, 110A, 111 difference calculation unit, 120 correction unit, 112 coefficient calculation unit, 113 multiplication unit, 121 correction unit

Claims (12)

An output from a first inertial sensor having a first dynamic range and an output from a second inertial sensor having a second dynamic range wider than the first dynamic range are received, and the first dynamic range and the A calculation unit including a difference calculation unit that calculates a difference between outputs synchronized with each other from the first inertial sensor and the second inertial sensor, which are outputs of dynamic ranges overlapping each other in the second dynamic range;
A correction unit that corrects an output from the second inertial sensor based on a calculation result of the calculation unit;
A sensor output correction device. The sensor output correction device according to claim 1,
The difference calculating unit calculates a difference between a minimum value of outputs detected by the first inertial sensor and an output of the second inertial sensor synchronized with the minimum value. Correction device. The sensor output correction device according to claim 1,
The difference calculating unit calculates a difference between a maximum value of outputs detected by the first inertial sensor and an output of the second inertial sensor synchronized with the maximum value. Correction device. The sensor output correction device according to claim 1,
The difference calculation unit calculates an average value of a difference between outputs at a time t1 in outputs detected by the first inertia sensor and the second inertia sensor and a difference between outputs at a time t2. A sensor output correction device characterized by calculating as a difference. In the sensor output correction device according to claim 4,
The sensor output correction apparatus characterized in that the output of the first inertial sensor at the time t1 is set to a minimum value, and the output of the first inertial sensor at the time t2 is set to a maximum value. In the sensor output correction device according to any one of claims 1 to 5,
The difference calculation unit repeatedly calculates the difference based on a calculation instruction signal, and updates the difference to the latest. In the sensor output correction according to claim 1,
The difference calculation unit calculates a difference between outputs at time t1 and a difference between outputs at time t2 among outputs detected by the first inertia sensor and the second inertia sensor,
The calculation unit further includes a coefficient calculation unit that calculates a coefficient K using the two differences from the difference calculation unit;
A multiplier for multiplying the output of the second inertial sensor by a coefficient K,
The sensor output correction device, wherein the correction unit corrects an output of the second inertial sensor based on an output from the multiplication unit and an output of the first inertial sensor. In the sensor output correction device according to claim 7,
Of the outputs detected by the first inertial sensor, the output at time t1 is S1 (t1), the output at time t2 is S1 (t2), and the output detected by the second inertial sensor is When the output at time t1 is SO2 (t1) and the output at time t2 is S1 (t2),
K = [S1 (t1) -S1 (t2)] / [SO2 (t1) -SO2 (t2)]
A sensor output correction device characterized by that. In the sensor output correction device according to claim 8,
Among the outputs detected by the first inertial sensor, the output S1 (t1) at the time t1 is one of the maximum value and the minimum value, and the output S1 (t2) at the time t2 is the maximum value and A sensor output correction device characterized by being the other of the minimum values. In the sensor output correction device according to any one of claims 1 to 9,
The sensor output correction device, wherein the first inertia sensor and the second inertia sensor are attached to substantially the same location. Receiving an output from a first inertial sensor having a first dynamic range and an output from a second inertial sensor having a second dynamic range wider than the first dynamic range;
A step of calculating a difference between outputs of the first dynamic range and the second dynamic range that overlap each other, the outputs being synchronized with each other from the first inertia sensor and the second inertia sensor;
Correcting the output from the second inertial sensor based on the difference;
A sensor output correction method comprising: Out of the output from the first inertial sensor having the first dynamic range and the output of the second inertial sensor having the second dynamic range wider than the first dynamic range, output data having a dynamic range overlapping each other, and A procedure for calculating a difference between outputs synchronized with each other;
Correcting the output from the second inertial sensor based on the difference;
A sensor output correction program characterized by causing a computer to execute.

Images