JP2016144733A - Swing analysis system using motion sensor, swing analysis method, and swing analysis program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a swing analysis system which can obtain more accurate results by considering individual differences in individuals and sensors, a swing analysis method, and a swing analysis program.SOLUTION: A swing analysis system comprises a motion sensor attached to a subject of swing analysis, and a terminal device which can perform radio communication with the motion sensor. The terminal device comprises a swing data analysis part for performing swing analysis based on sensor data output from the motion sensor, and an error correction part for correcting output error of the motion sensor. The error correction part estimates the output error of the motion sensor by obtaining a difference between a coordinate point at the time of address of the subject and a coordinate point at the time of impact on the basis of the sensor data output from the motion sensor, and performs recalculation of the sensor data output from the motion sensor by using the estimated error.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a swing analysis system using a golf club or the like, a swing analysis method, and a swing analysis program, and more particularly to a swing analysis system, a swing analysis method, and a swing analysis program using a motion sensor.

  2. Description of the Related Art Conventionally, various devices for analyzing human movements such as golf clubs and baseball bat swings have been proposed. For example, a device that attaches a marker to a golf club or the like, continuously shoots with a camera until a ball is hit, and analyzes a swing based on the captured image is generally used. However, such an apparatus cannot visualize the entire swing, and therefore cannot perform a detailed swing analysis. Further, since the apparatus itself becomes a large scale, it is not an environment in which the user can easily measure the swing at any time.

  In order to solve the problem in the swing analysis apparatus using such a camera, Non-Patent Document 1 discloses a technique for performing a swing analysis using a compact motion sensor. Specifically, in the swing analysis device of Non-Patent Document 1, a motion sensor, a battery, and a communication module arranged in series are embedded in a grip of a golf club. And it is the structure which performs the swing of a golf club with a swing machine and analyzes the data output from a motion sensor.

Kevin W. King, “THE DESIGN AND APPLICATION OFWIRELESS MEMS INERTIAL MEASUREMENT UNITS FOR THE MEASUREMENT AND ANALYSIS OFGOLF SWINGS”, [online], 2008, Internet <URL: http://deepblue.lib.umich.edu/bitstream/2027.42 /58460/1/kwking_1.pdf>

  However, the swing analysis technique in Non-Patent Document 1 is to perform a swing analysis based on data obtained by measuring a swing by a swing machine, and is based on a person or a sensor when performing a swing analysis by a person. Individual differences and blurring are not considered. Specifically, the following problems can be considered when actually performing a swing analysis by a person.

  First, in Non-Patent Document 1, the direction perpendicular to the face surface at the time of addressing is set as a target line that is a reference for angle measurement of the swing locus. Since the swing machine can always take a fixed address posture, the direction perpendicular to the face surface can always be set as the target line. However, it is difficult for a general user to always take a fixed address posture for each swing and to recognize whether the face surface is facing the target direction. Therefore, the actual target line changes depending on the address posture, and there is a possibility that correct angle measurement cannot be performed.

  Further, since the motion sensor is a machine, there are output errors due to physical individual differences such as bias of electrodes and springs, and events in which errors increase due to heat generation over time. Although the measurement results differ depending on the output errors of these sensors, the sensor error is not considered in the swing analysis in Non-Patent Document 1.

  Further, Non-Patent Document 1 assumes a swing operation by a swing machine as described above, and the swing is performed immediately after the address state. However, in general, a general user does not swing immediately from an address state, but often starts a swing after performing pre-routine or swing preparatory movement (hereinafter referred to as “wuggle”). . When measurement data corresponding to such a waggle operation is used for swing analysis, a large error may occur in the subsequent swing measurement result.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a swing analysis system and swing analysis capable of obtaining a more accurate analysis result in consideration of individual differences between people and sensors. A method and a swing analysis program are provided.

  A swing analysis system according to an aspect of the present invention that solves the above-described problem is a swing analysis system that includes a motion sensor attached to an object of swing analysis and a terminal device capable of wireless communication with the motion sensor. Based on the sensor data output from the motion sensor, the terminal device obtains the posture information of the object from the swing analysis unit that performs the swing analysis and the sensor data when the motion sensor is stationary. And a target line setting unit that sets a target line as a reference for swing analysis in the swing analysis means.

  The motion sensor may include an acceleration sensor capable of detecting acceleration in three axis directions, and the target line setting unit may obtain posture information from sensor data output from the acceleration sensor. Further, the posture information is a vector in the gravitational acceleration direction, and the target line setting unit may obtain the target line in the first coordinate system based on the outer product of the vector in the gravitational acceleration direction and the vector in the Z-axis direction. good.

  Further, the target line setting unit may convert the target line in the first coordinate system into the second coordinate system and set it as the target line used in the swing analysis means. Further, the target line setting unit may process the set target line so as to be parallel to one axis in the second coordinate system.

  The swing analysis unit may use the target line when determining the swing trajectory obtained from the sensor data.

  The swing analysis system further includes a stationary state detection unit that detects a state in which the motion sensor is stationary. The stationary state detection unit is configured to obtain an average value and a standard value of a predetermined number of sensor data output from the motion sensor. A deviation value may be calculated, and a case where the standard deviation value is lower than a predetermined threshold value may be detected as a stationary state. Furthermore, the predetermined number and the predetermined threshold value are variable according to the use conditions, and the use conditions may include the sensitivity of the motion sensor, the age of the user, and the type of the object of the swing analysis.

  The swing analysis system may further include a posture determination unit that determines whether or not the motion sensor is correctly attached when the motion sensor is stationary. Further, the posture determination unit may determine whether or not the motion sensor is correctly attached based on the sign and average value of the sensor data output from the motion sensor.

  A swing analysis method according to another aspect of the present invention is a swing analysis method executed in a terminal device connected to a motion sensor attached to an object of swing analysis so as to be capable of wireless communication. A step of obtaining posture information of an object from sensor data in a stationary state, a step of setting a target line based on posture information, and a step of performing a swing analysis using sensor data output from a motion sensor And using a target line as a reference in swing analysis.

  A swing analysis program according to another aspect of the present invention causes a CPU of a terminal device connected to a motion sensor attached to an object of swing analysis to perform wireless communication so that each step in the swing analysis method is executed. It is characterized by.

  A swing analysis system according to another aspect of the present invention is a swing analysis system including a motion sensor attached to an object of swing analysis and a terminal device capable of wireless communication with the motion sensor. Includes a swing data analysis unit that performs a swing analysis based on sensor data output from the motion sensor, and an error correction unit that corrects an output error of the motion sensor. The error correction unit is output from the motion sensor. Based on the sensor data, the difference between the coordinate point at the address of the target object and the coordinate point at the time of impact is obtained and estimated as the output error of the motion sensor, and output from the motion sensor using the estimated error. The sensor data is recalculated. The motion sensor may include an acceleration sensor, and the error correction unit may correct an output error in the acceleration sensor.

  A swing analysis system according to another aspect of the present invention is a swing analysis system including a motion sensor attached to an object of swing analysis and a terminal device capable of wireless communication with the motion sensor. Includes a swing state determination unit that determines a swing state from sensor data output from the motion sensor, and a swing data analysis unit that performs a swing analysis based on the sensor data output from the motion sensor. Is provided with an angular velocity sensor capable of detecting the angular velocity in the triaxial direction, and the swing state determination unit detects a swing start point based on sensor data in the triaxial angular velocity sensor and the impact of the object, and a swing data analysis unit Is the sensor data detected before the detected swing start point. Configured for not using at least some swing analysis.

  Further, the swing state determination unit detects the negative peak value of the Z-axis sensor data of the angular velocity sensor and the positive peak value of the X-axis sensor data, and goes back from the time of impact to the start of measurement. The point at which one of the signs of the plus peak value is inverted may be determined as the swing start point. Further, the swing state determination unit goes back from the time of impact to the start of measurement, and selects the point closer to the start of measurement among the point where the sign of the minus peak value is inverted and the point where the sign of the plus peak value is inverted. You may determine with a swing start point.

  The motion sensor may further include an acceleration sensor capable of detecting acceleration in three axes, and the swing state determination unit may determine that the negative peak value of the Y-axis sensor data of the acceleration sensor is an impact time. Alternatively, the swing state determination unit may determine that there is an impact when there is a difference of a predetermined value or more between two sensor data values in the Y-axis sensor data of the acceleration sensor.

  According to the present invention, it is possible to provide a swing analysis system, a swing analysis method, and a swing analysis program capable of obtaining a more accurate analysis result in consideration of individual differences between people and sensors.

It is a block diagram which shows the structure of the swing analysis system of embodiment of this invention. It is an external view which shows attachment of the sensor in embodiment of this invention. It is a flowchart which shows the swing analysis process in embodiment of this invention. It is a flowchart which shows the swing analysis process in embodiment of this invention. It is a figure for demonstrating the positional relationship of a golf club, a sensor, and a target line. It is a figure for demonstrating the relationship between a swing track | orbit and a target line. It is a figure which shows the user coordinate system in embodiment of this invention. It is a figure which shows the waveform by the sensor data in embodiment of this invention.

  Hereinafter, a swing analysis system according to an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, a case where golf swing analysis is performed will be described as an example.

  FIG. 1 is a diagram showing a schematic configuration of a swing analysis system 1 in the embodiment of the present invention. As shown in FIG. 1, the swing analysis system 1 of the present embodiment includes a sensor 10 and a terminal device 20 that can wirelessly communicate with each other. The sensor 10 includes a motion sensor 100, a data processing unit 110, a communication unit 120, an operation button 130, and an LED 140. The motion sensor 100 includes an acceleration sensor that detects triaxial acceleration and an angular velocity sensor that detects triaxial angular velocity. The motion sensor 100 may further include a geomagnetic sensor. The data processing unit 110 is a processing unit that synchronizes each data detected by the motion sensor 100, performs processing such as bias correction and temperature correction, and outputs the processed data to the communication unit 120. The communication unit 120 is a communication unit for performing narrow-range wireless communication with the terminal device 20, and in this embodiment, data is transmitted / received to / from the terminal device 20 by Bluetooth (registered trademark). Note that the communication unit 120 may perform wireless communication using WiFi (wireless LAN) in addition to Bluetooth (registered trademark).

FIG. 2 is an external perspective view showing the sensor 10 attached to the golf club 6. As shown in FIG. 2, the sensor 10 is a small unit formed so as to have a width substantially the same as the diameter of the shaft of the golf club 6, and is detachable near the boundary between the grip of the golf club 6 and the shaft. Attached to. Further, the sensor 10 is fixed to the golf club 6 by a holder 15 such as a rubber band so that the sensor 10 does not move due to the swing speed or impact when hit.
By adjusting the size of the holder 15 and the tightening of the rubber, it can be attached to golf clubs of all thicknesses, baseball bats and tennis rackets other than golf clubs. Further, since the sensor 10 is reduced in size and weight, the balance of the center of gravity of the golf club 6 after the sensor 10 is attached is not lost, so that swing measurement can be performed while swinging as usual.

  As will be described later, the user can start transmission / reception of data between the sensor 10 and the terminal device 20 and start swing measurement by pressing the operation button 130 of the sensor 10 before the swing operation. Further, the LED 140 is lit according to the situation such as whether or not the terminal device 20 side is ready for swing analysis or whether a communication error has occurred between the terminal device 20 and the sensor 10. Thereby, the user should continue the swing measurement before starting the swing operation by checking the lighting state of the LED 140 without looking at the error screen or the operation instruction displayed on the display unit 220 of the terminal device 20. , Can decide whether to interrupt.

  Returning to FIG. 1, the terminal device 20 includes a CPU 200, a communication unit 210, a display unit 220, an operation unit 230, a ROM 240, a RAM 250, and a nonvolatile memory 260. In addition, although the smart phone is used as the terminal device 20 in this embodiment, a mobile phone terminal, PDA (Personal Digital Assistants), PHS (Personal Handy phone System), a portable game machine, a digital home appliance, a car navigation system, Other types of terminals such as desktop PCs and laptop PCs can be used. The terminal device 20 is placed within a range where wireless communication with the sensor 10 is possible (for example, a practice field shelf) while the swing is measured.

  The CPU 200 communicates with each element included in the terminal device 20 and performs overall control of the device. The ROM 240 and the nonvolatile memory 260 are memories for storing various programs and data executed by the terminal device 20. The RAM 250 is a volatile memory that temporarily stores various data and serves as a load destination of various programs stored in the ROM 240 or the nonvolatile memory 260. Various programs stored in the ROM 240 or the nonvolatile memory 260 are executed by the CPU 200.

  Further, the CPU 200 includes an operation determination unit 201, a stationary state detection unit 202, a posture determination unit 203, a target line setting unit 204, a swing state determination unit 205, and a swing data analysis unit 206. These are functional units implemented by the CPU 200 executing an application program stored in the nonvolatile memory 260. The function of each part will be described in detail later. Note that all or part of each functional unit may be mounted on the terminal device 20 as an ASIC (Application Specific Integrated Circuit), and the hardware may be realized by the ASIC.

  The communication unit 210 is a communication unit for performing wireless communication with the sensor 10 and Bluetooth (registered trademark). The communication unit 210 further establishes a connection with an external network in order to transmit / receive data to / from an external network (such as the Internet) (call incoming / outgoing, transmission / reception of e-mail, acquisition of Web content, etc.) A network interface may be included.

  The operation unit 230 is configured by a touch panel, and input by pen touch or finger touch for content displayed on the display unit 220 configured by LCD (Liquid Crystal Display), organic EL (Electro Luminescence), or the like, Various operations such as screen scrolling by a flick operation (operation with a finger on the screen) and zoom-in / zoom-out by a pinch operation (operation to expand or contract between two fingers on the screen) can be performed. The operation unit 230 may be configured with various keys provided separately from the display unit 220.

  Next, the flow of the swing analysis process in this embodiment will be described with reference to the flowcharts of FIGS. 3A and 3B. The swing analysis process according to the present embodiment is started when the user operates and activates the swing analysis application program downloaded in advance to the nonvolatile memory 260 by operating the operation unit 230 of the terminal device 20. When the swing analysis application program is activated, a screen for inputting information on the golf club to be used is displayed on the display unit 220. And the information of the golf club used by a user is input, the information input in the operation determination part 201 is acquired, and it memorize | stores in the non-volatile memory 260 (S1).

  Subsequently, the sensor 10 used for the swing analysis process is set (S2). Here, in the swing analysis system 1, there is a possibility that a plurality of sensors 10 exist within the wireless communication area of the terminal device 20 (for example, when one is mounted on each of a plurality of golf clubs). Therefore, in the terminal device 20, it is necessary to specify the sensor used in the current swing analysis process. Therefore, when the terminal device 20 and the sensor 10 are paired for the first time, the sensor 10 transmits the serial number at the time of manufacture to the terminal device 20 so that the terminal device 20 can identify the sensor 10. Yes. In S2, a screen for selecting the sensor 10 to be used is displayed on the display unit 220, and the operation determination unit 201 identifies the sensor selected by the user as the sensor 10 to be used this time, and the nonvolatile memory 260. To remember.

  Subsequently, the sensor 10 is attached to the golf club 6 as shown in FIG. 2, and the power is turned on (S3). Thereby, wireless communication is established between the terminal device 20 and the sensor 10, and data can be transmitted and received (S4). Then, the user prepares to swing by holding the golf club 6, and presses the operation button 130 of the sensor 10 when the preparation is completed (S5). At this time, if it is confirmed by the lighting of the LED 140 that the terminal device 20 is not ready for swing measurement or that a communication error has occurred, either wait for a while without pressing the operation button 130, or You can start over from processing.

  The operation determination unit 201 of the terminal device 20 detects that the operation button 130 of the sensor 10 has been pressed, and instructs the sensor 10 to start transmission of sensor data output from the motion sensor 100 (S6). Thereby, in the sensor 10, each sensor data detected by the triaxial acceleration sensor and the triaxial angular velocity sensor of the motion sensor 100 is transmitted to the terminal device 20. Specifically, in the present embodiment, a predetermined number of data is detected per second in each sensor and is transmitted to the terminal device 20 as needed via the data processing unit 110 and the communication unit 120.

  Subsequently, the terminal device 20 starts receiving the sensor data transmitted from the sensor 10, and stores the received sensor data in the RAM 250 as needed (S7). Then, the user is instructed to take an address posture via the display unit 220 or a speaker (not shown) (S8). Thereafter, the stationary state detection unit 202 detects the stationary state based on the received sensor data (S9). The stationary state is a state where the sensor 10 is stationary, that is, an address state before the start of the swing. In S9, the average value and standard deviation value of several (for example, 350 consecutive) sensor data from the start of reception of sensor data in S7 to the latest are calculated. Then, when the obtained standard deviation value becomes lower than a predetermined threshold value, it is determined that the stationary state is detected. Here, by adjusting the sensitivity of the motion sensor 100, the age of the user, and the golf club to be used, the number of data for calculating the average value and the threshold value to be compared with the standard deviation value, the type of golf club and the individual difference of the user Regardless of whether or not, the stationary state can be appropriately determined. If the stationary state is detected (S10: Yes), the process proceeds to S11. If the stationary state is not detected (S10: No), the process returns to S9 and the stationary state is detected again.

  Subsequently, the posture determination of the sensor 10 is performed using the sensor data when the stationary state is determined by the posture determination unit 203 (S11). Here, the ROM 240 stores in advance the sign and average value of the sensor data of each axis in the acceleration sensor of the motion sensor 100 when the mounting direction of the sensor 10 is correct. The posture determination unit 203 reads each value stored in the ROM 240 and compares it with sensor data when a stationary state is detected. If the sensor data at the time when the stationary state is detected does not match the sign of the stored sensor data, and the average value is not within a certain range with respect to the average value stored in the ROM 240, the sensor 10 It is determined that the mounting position is not correct or the address posture is completely different from the normal golf address posture. In addition, the setting of the allowable range in the posture determination of the sensor 10 is changed according to the sensitivity of the motion sensor 100, the age of the user, and the golf club to be used, similarly to the detection of the stationary state, so that the type of golf club and the user Regardless of individual differences, it is possible to determine the posture appropriately. As a result of the posture determination, if it is determined that the mounting direction of the sensor 10 is correct (S12: Yes), the process proceeds to S13. On the other hand, when it is determined that the mounting direction of the sensor 10 is not correct (S12: No), the process returns to S3, the sensor 10 is mounted again, and the subsequent processing is repeated. By performing the posture determination in this way, it is possible to prevent the sensor from being swung in a state where it is attached in error.

  Depending on the location where the sensor 10 is attached, the thickness, material and shape of the object to be attached, and the material and shape of the holder 15, the sensor 10 may not be attached at the position shown in FIGS. In such a case, the mounting position of the sensor 10 can be changed from the position shown in FIGS. 4 and 6 by changing the sign and average value of the sensor data when the mounting position stored in the ROM 240 is correct. is there. When the sensor position shown in FIG. 6 can be rotated 90 ° clockwise and attached, the sensor data from each axis of the sensor 10 is replaced with the position shown in FIG. Even if the position is not set to the front as shown in FIG. 6, the posture can be determined in the same manner. Further, the operation unit 230 may allow the user to select an attachment position and save it in the RAM 250, and based on the attachment position saved in the RAM 250, the sign and average value of sensor data for determining the posture as normal may be changed.

  In subsequent S13, the target line setting unit 204 sets the target line using the sensor data when the stationary state is detected. The target line is used as a reference for classifying the measured swing trajectory into inside-in, inside-out, or outside-in (swing plane) when performing a swing analysis. The target line setting process in this embodiment will be described in detail later.

  Subsequently, an instruction is given to the user to start a swing from the display unit 220 or the speaker (not shown) of the terminal device 20 (S14). Then, the swing motion by the user is measured by the sensor 10, and each sensor data is received by the terminal device 20. Subsequently, the swing state determination unit 205 performs a swing state determination process based on the received sensor data (S15). In this processing, each state of the swing is sequentially determined and stored in the RAM 250. Each state of the swing is an address state (before swing), swing start (start of takeback), takeback swing, top of swing, downswing, impact, and follow-through. In the swing state determination process of the present embodiment, a process for correcting a sensor error with respect to received sensor data and a waggle removal process for detecting a swing start point are performed. These processes will be described in detail later.

  Next, the swing state determination unit 205 determines whether or not an impact has been detected (S16). When the impact is detected (S16: Yes), the terminal device 20 instructs the sensor 10 to stop the measurement, and the sensor 10 stops the measurement and the data transmission to the terminal device 20 (S17). ).

  Thereafter, the swing data analysis unit 206 performs swing data analysis based on the data stored in the RAM 250 (S18). Specifically, the swing data analysis unit 206 analyzes various data necessary for swing measurement from the sensor data. Various data required for swing measurement includes head speed, face angle, real loft angle, lie angle, left and right approach angle, up and down approach angle, wrist speed, wrist rotation speed, swing plane, wrist turn coefficient, tempo and swing trajectory. There are coordinates. After the analysis is performed by the swing data analysis unit 206, the analysis result is stored in the nonvolatile memory 260 and displayed on the display unit 220 of the terminal device 20 (S19). The analysis result by the swing data analysis unit 206 may be transmitted to a server that can communicate with the terminal device 20 via the Internet or the like, and may be stored and managed by the server.

  The display unit 220 displays the coordinate information of the swing trajectory in 3D animation, the swing trajectory that can rotate 360 degrees, and the measurement results at each measurement point from the swing start to the swing end (head speed, face angle, real loft angle, etc.) ) Is displayed. In this case, various displays can be performed so that the user can easily grasp the measurement result. Specifically, it is conceivable to color the swing trajectory according to the change in the state of the swing, the color of the seek bar, the color of the shaft display, and the switching of the display method of the shaft and swing arc.

  In S19, it is desirable not only to display and record each measurement result and swing trajectory, but also present the swing improvement points so that the user can easily understand. Therefore, in the present embodiment, it is possible to display advice on the display unit 220 of the terminal device 20 based on the result of the swing analysis. Specifically, the user first operates the operation unit 230 of the terminal device 20 and inputs the direction in which the ball actually flew after the swing motion. The swing data analysis unit 206 gives appropriate advice from pre-stored advice based on the input ball direction, the head speed obtained by analyzing the sensor data, the wrist speed ratio, and the type of the swing plane. Select and display on the display unit 220. Thereby, the user can easily understand the improvement point.

  Furthermore, it is generally considered that the user is most interested in the behavior immediately before the head of the golf club 6 hits the ball during the swing. Therefore, in the present embodiment, after detecting the impact, dozens of sensor data before and after the impact are extracted, and the user is able to correct the behavior of the face surface (face angle displacement and trajectory information) just before the head hits the ball. A close-up as a trajectory seen from above is displayed on the display unit 220. Thereby, the user can know the behavior before and after the head of the golf club 6 hits the ball in more detail.

  Subsequently, it is determined whether or not the operation button 130 of the sensor 10 has been pressed (S20). When the operation button 130 is pressed (S20: Yes), the process returns to S6, and measurement by the sensor 10 and transmission of sensor data to the terminal device 20 are performed again. The subsequent processing is repeated for the new swing. On the other hand, when the operation button 130 is not pressed (S20: No), the communication between the terminal device 20 and the sensor 10 is disconnected (S21), and this process is terminated.

  Next, the target line setting process in the present embodiment will be described with reference to FIGS. As described above, in the prior art, since a swing machine is used, the attachment position of the sensor 10 to the golf club 6 and the attitude of the sensor 10 are always constant. Specifically, in the prior art, as shown in FIG. 4, it is possible to statically set the target line direction as a direction orthogonal to the face surface of the golf club 6. However, when actually analyzing a swing by a person, the loft angle and the lie angle of the golf club 6 cannot always be constant at the time of addressing, and the direction perpendicular to the face surface is changed from the target line by the change of these angles. There was a problem that it was not parallel.

  If the target line serving as a reference for measuring the swing trajectory deviates from the original direction, the measurement result becomes correct. FIGS. 5A and 5B are diagrams showing the relationship between the swing trajectory and the target line. The swing trajectory refers to a trajectory when the trajectory drawn by the head of the golf club 6 is viewed from directly above as shown in FIGS. 5 (a) and 5 (b). When performing a swing analysis, the swing trajectory is classified as one of inside-in, inside-out, and outside-in with the target line as a reference line. In general, it is said that an ideal swing trajectory is an inside-in trajectory. Here, even if the swing trajectory should be determined inside-in as shown in FIG. 5A, the target line direction is different from the original direction as shown in FIG. 5B. If it is set to, the swing trajectory is erroneously determined as inside-out. Therefore, in the target line setting process of the present embodiment, the target line is dynamically set according to the direction of the sensor 10 at the time of addressing (that is, the direction of the shaft of the golf club 6).

  The flow of the target line setting process will be described in detail below. As a premise of this processing, as shown in FIG. 4, it is assumed that the sensor 10 is mounted on the golf club 6 so that the surface on which the operation button 130 of the sensor 10 is provided is parallel to the target line. More specifically, the sensor 10 is attached to the golf club 6 so that the X axis in the coordinate system shown in FIG. 4 (hereinafter referred to as “sensor coordinate system”) is parallel to the target line and the Z axis is orthogonal to the target line. It shall be installed. Further, at the time of addressing, the user takes a stance substantially parallel to the target line as shown in FIG. 6, that is, substantially parallel to the Y axis of the coordinate system shown in FIG. It is assumed that the golf club 6 is held.

  In the target line setting process, first, from the sensor data of the motion sensor 100 when the stationary state is detected, the direction of the sensor 10 at the time of addressing (that is, the posture information of the golf club 6) is obtained, and the target line in the sensor coordinate system is obtained. Is calculated. Then, the calculated target line of the sensor coordinate system is converted into the user coordinate system, and compared with the state shown in FIG. 6, how much the orientation of the sensor 10 is rotated is determined, and the original target line is set. .

A more specific flow in this processing will be described below.
1. First, when a stationary state is detected, a target line vector target ′ in the sensor coordinate system is obtained from sensor data output from the acceleration sensor at that time. The target line vector target ′ is a direction orthogonal to the gravitational acceleration direction and the Z-axis direction in the sensor coordinate system. Therefore, as shown in the following formula (1), the target line vector target ′ can be obtained from the outer product of the gravity acceleration direction vector representing the posture information of the golf club 6 and the Z-axis direction vector.
target '= Z × g (1)

尚、ｅ１−ｅ４の初期条件（ｔ＝０）は、以下のとおりである。

2. Subsequently, in order to obtain a target vector of the target line in the user coordinate system, conversion from the sensor coordinate system to the user coordinate system is performed. Specifically, The target line vector target ′ in the sensor coordinate system calculated in step (1) is normalized (ie, unit vectorized), and the rotation line (2) and (3) below is used to calculate the target line in the user coordinate system. Find the vector target. The target line vector target obtained here is set as the original target line and stored in the nonvolatile memory 260.

The initial conditions (t = 0) of e1-e4 are as follows.

  3. In addition, 2. The target vector of the target line obtained in step 1 is not parallel to the Y axis in the user coordinate system. Therefore, in order to perform the subsequent swing analysis, the product of the known rotation matrix P that rotates the vector target of the target line with respect to the Y axis and the above-described expression (2) is the rotation matrix of expression (2). And correct to be parallel to the Y axis.

  Thus, in the present embodiment, the target line can be dynamically set according to the direction of the sensor 10 at the time of addressing. Therefore, even when the golf club 6 is held at various angles by a person, it is possible to obtain an accurate swing trajectory measurement result reflecting these individual differences. Moreover, appropriate advice display can be performed by setting the original target line in this way.

  Next, a process for correcting an output error of the sensor, which is performed in the swing state determination process of the present embodiment, will be described. General acceleration sensors are digital and analog methods, and there are detection methods such as a capacitance detection method, a piezoresistive method, and a heat detection method. However, since any method is a machine, there is a physical bias. These biases are calibrated and corrected when the sensor is manufactured. However, depending on the environment at the time of calibration, the accuracy of calibration may be low and errors may not be corrected. Further, even if calibration is performed at the time of manufacture, calibration cannot be performed for less than 1 LSB (Least Significant Bit). However, when a man swings using a driver, the swing speed is generally 40 m / s or more, and it is necessary to set the sensitivity settings of the acceleration sensor and the angular velocity sensor to low sensitivity and increase the measurement range. For this reason, the measurement accuracy per 1 LSB is lowered, and the error factor is increased. As a result, if there is an error below 1 LSB, the calculation accuracy will be lowered although the sensor output cannot be corrected. Further, there is a problem that the sensor itself generates heat in proportion to the sensor usage time, causing distortion of the substrate in micro units due to heat generation, and further increasing the error in sensor output. As described above, if the accuracy of the sensor output is lowered due to an error due to the passage of time or an individual difference between the sensors, the measurement result becomes unstable and accurate swing analysis cannot be performed. Therefore, in this embodiment, an error of the motion sensor 100 (more specifically, an error of the acceleration sensor) is obtained, and the sensor data is corrected based on the error, thereby stabilizing the measurement result.

  Generally, the user starts the swing after placing the head of the golf club 6 immediately behind the ball at the address posture of the golf swing. Therefore, it can be assumed that the coordinate point at the start of the swing matches the coordinate point of the impact. On the other hand, if the coordinate point at the start of the swing and the coordinate point at the time of impact are significantly separated, it is considered that the position has shifted due to the influence of some error. Therefore, in this embodiment, the difference between the coordinate point at the start of the swing and the coordinate point at the time of impact is obtained as an excess / shortage movement distance, and the acceleration that moves the distance is added or subtracted and recalculated. By doing so, an error of less than 1 LSB of the sensor 10 and an error due to the passage of time are corrected.

そして、時刻 t における真の速度 v(t)は、下記の式（９）で表わされる。

また、アドレス時点でのクラブヘッド位置をr(0)とすると、時刻 t における真の位置 r(t)は、下記の式（１０）で表わされる。

さらに、インパクト時刻をＴとすると、真のインパクト位置r(T)は、下記の式（１１）で表わされる。

ここで、下記の式（１２）に示されるように、重力加速度ｇにセンサ１０による誤差ｇｅが加算されていたとすると、時刻 t における加速度 a(t)、速度v(t)、および位置 r(t)はそれぞれ下記の式（１３）−（１５）で求められる。

また、誤差 ｇｅは、下記の式（１６）および（１７）によって求められる。

A specific calculation method in this process will be described below. In the following description and mathematical formulas, the true value is represented by italic characters.
First, if the acceleration detected by the sensor 10 at time t is ai (t) and the true gravitational acceleration is g, the true acceleration a (t) at time t is expressed by the following equation (8).

The true speed v (t) at time t is expressed by the following equation (9).

If the club head position at the time of address is r (0), the true position r (t) at time t is expressed by the following equation (10).

Furthermore, if the impact time is T, the true impact position r (T) is expressed by the following equation (11).

Here, as shown in the following equation (12), if the error ge from the sensor 10 is added to the gravitational acceleration g, the acceleration a (t), the velocity v (t), and the position r ( t) is obtained by the following equations (13) to (15).

Further, the error ge is obtained by the following equations (16) and (17).

  As described above, in the present embodiment, the error ge is obtained based on the measurement result first obtained based on the sensor data, and the error ge is corrected again for the sensor data to obtain the measurement result. . By configuring in this way, more accurate measurement results can be obtained. Thereby, the behavior of the head of the golf club 6 before and after the impact can be accurately displayed.

  Next, a process for removing waggle performed in the swing state determination process of the present embodiment will be described. In general, in a user's swing motion, a preliminary motion (waggles) unrelated to the swing may be performed before the initial motion of the swing. Since this waggle movement is a trajectory-less movement, if the waggle movement is also calculated as a swing trajectory, it becomes a large error factor in subsequent swing measurement results. In addition, in order to detect a swing by a driver whose maximum speed is 40 m / s or more, the sensitivity setting of the acceleration sensor and the angular velocity sensor is set to the maximum speed that can be measured with a finite number of bits. The detectable unit is large. When the entire range width that can be measured is expanded as described above, there is a problem that it is impossible to accurately detect an operation at a very slow speed such as a waggle operation.

  Therefore, in this embodiment, the swing state determination unit 205 detects a swing start point obtained by removing the swing trajectory due to the waggle from the sensor data, and the sensor data based on the waggle operation is not used for the swing analysis.

  A specific flow in this processing will be described below. FIG. 7 shows the waveforms of the X-axis and Z-axis sensor data in the sensor coordinate system of the angular velocity sensor of the motion sensor 100 and the Y-axis sensor data of the acceleration sensor when the sensor 10 is mounted on the golf club 6 and swung. It is. Here, the swing state determination unit 205 detects the negative peak value Z1 of the Z-axis sensor data of the angular velocity sensor and the positive peak value X1 of the X-axis sensor data of the angular velocity sensor from the data measurement start point, and stores them in the RAM 250. To do. After detecting the impact point T1, both the X-axis sensor data and the Z-axis sensor data are traced back to the point where the sign is inverted with respect to the peak values Z1 and X1 recorded in the RAM 250. Specifically, as shown in FIG. 7, a point Z2 where the Z-axis sensor data turns from a minus peak value Z1 to a plus point and a point X2 where the X-axis sensor data turns from a plus peak value X1 to a minus point are detected. . Then, at the detected points Z2 and X2, the one having fewer measurement points with the data detection start point is determined as the swing start point S. In the swing data analysis process, the sensor data of the acceleration sensor before the determined swing start point is filtered as only gravitational acceleration and is not used for swing analysis, so that a more accurate analysis result can be obtained. .

  Further, the detection of the impact point T1 is determined by detecting the rapid deceleration of the Y-axis sensor data of the acceleration sensor or the negative peak in the waveform of FIG. Further, it is generally considered that, when an impact occurs, the movement stops momentarily by hitting a ball, and a large difference occurs in the sensor data. Therefore, in addition to the above, it may be determined that the impact T1 is detected when there is a predetermined difference between the values of the sensor data before and after the Y-axis sensor data waveform of the acceleration sensor shown in FIG. Furthermore, by changing the difference constant before and after determining the impact according to the type of golf club or ball, it is possible to determine whether or not the ball has been hit according to the degree of impact when hit, Impact can be detected accurately.

  Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments and can be modified in various ranges. For example, the present embodiment relates to a golf swing analysis system, but the present invention is not limited to this, and can also be applied to a system for analyzing various swing motions such as tennis and baseball.

DESCRIPTION OF SYMBOLS 1 Swing analysis system 10 Sensor 20 Terminal device 100 Motion sensor 110 Data processing part 120 Communication part 130 Operation button 140 LED
200 CPU
210 Communication unit 220 Display unit 230 Operation unit 240 ROM
250 RAM
260 Nonvolatile memory

Claims (7)

A swing analysis system comprising a motion sensor attached to an object of swing analysis and a terminal device capable of wireless communication with the motion sensor,
The terminal device
Based on sensor data output from the motion sensor, a swing data analysis unit that performs a swing analysis,
An error correction unit for correcting an output error of the motion sensor;
With
The error correction unit estimates an output error of the motion sensor by obtaining a difference between a coordinate point at the time of addressing the object and a coordinate point at the time of impact based on sensor data output from the motion sensor. A swing analysis system that recalculates sensor data output from the motion sensor using the estimated error. The motion sensor includes an acceleration sensor,
The swing analysis system according to claim 1, wherein the error correction unit corrects an output error in the acceleration sensor. The direction parallel to the target line is the X axis, the direction parallel to the face surface of the object and orthogonal to the shaft of the object is the Z axis, and the direction orthogonal to both the X axis and the Z axis is the Y axis. As
The terminal device
A swing state determination unit that determines a swing state from sensor data output from the motion sensor;
The motion sensor includes an angular velocity sensor capable of detecting an angular velocity in three axial directions,
The swing state determination unit detects a swing start point based on sensor data in the triaxial angular velocity sensor and the impact of the object,
The swing analysis system according to claim 1 or 2, wherein the swing data analysis unit is configured not to use at least a part of sensor data detected before the detected swing start point for swing analysis.   The swing state determination unit detects a negative peak value of the Z-axis sensor data of the angular velocity sensor and a positive peak value of the X-axis sensor data, and goes back to the start of measurement from the impact time to the negative peak value. The swing analysis system according to claim 3, wherein a point at which one of a peak value and a sign of the plus peak value is inverted is determined as a swing start point.   The swing state determination unit goes back from the impact time to the start of measurement, and the point closer to the start of measurement among the point where the sign of the minus peak value is inverted and the point where the sign of the plus peak value is inverted The swing analysis system according to claim 4, wherein the swing start point is determined as a swing start point. The motion sensor further includes an acceleration sensor capable of detecting acceleration in three axial directions,
The swing analysis system according to any one of claims 3 to 5, wherein the swing state determination unit determines that the negative peak value of the Y-axis sensor data of the acceleration sensor is at the time of the impact. The motion sensor further includes an acceleration sensor capable of detecting acceleration in three axial directions,
6. The swing state determination unit according to claim 3, wherein, in the sensor data of the acceleration sensor, the Y-axis sensor data determines that the impact occurs when there is a difference between two sensor data values equal to or larger than a predetermined value. The swing analysis system according to one item.

Images