WO2021079967A1 - Racket analysis system, racket analysis device, racket analysis program, and racket analysis method - Google Patents

Abstract

A racket has a striking portion and a grip. The striking portion has a striking surface. The grip extends parallel to the striking surface, outside the striking surface on a virtual line that passes through the center of the striking surface. This analysis system analyzes the movement of the racket, and includes: an angular velocity sensor fixed to the racket; and a processing portion which, on the basis of a detected value obtained from the angular velocity sensor when the racket has been hit by a ball, estimates the position at which the ball hit, within the striking surface.

Description

Racket analysis system, racket analysis device, racket analysis program and racket analysis method

This disclosure relates to a racket analysis system for analyzing the movement of a racket, a racket analysis device, a racket analysis program, and a racket analysis method.

A technique for improving a player's skill by attaching a sensor to a racket and analyzing the movement of the racket is known (for example, Patent Document 1 below). Patent Document 1 discloses a technique in which an acceleration sensor and a contact sensor are provided on a racket for table tennis. The contact sensor is composed of a surface contact position coordinate detection type sensor in a tablet or the like, and is provided over the entire surface of the striking surface (face). Then, the contact sensor contributes to estimating which position of the striking surface the ball hits.

Japanese Unexamined Patent Publication No. 2009-183455

The racket analysis system according to one aspect of the present disclosure analyzes the movement of the racket. The racket has a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the central portion of the striking surface. .. The racket analysis system estimates the position of the ball on the striking surface based on the angular velocity sensor fixed to the racket and the detection value of the angular velocity sensor when the ball hits the racket. It has a processing unit and a processing unit.

The racket analysis device according to one aspect of the present disclosure analyzes the movement of the racket. The racket has a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. The racket analysis device has a processing unit that estimates the position of the ball hitting the striking surface based on the angular velocity of the racket when the ball hits the racket.

The racket analysis program according to one aspect of the present disclosure is for analyzing the movement of the racket. The racket has a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. The racket analysis program causes a computer to perform an estimation step of estimating a position of the hitting surface where the ball hits, based on the angular velocity of the racket when the ball hits the racket.

The racket analysis method according to one aspect of the present disclosure analyzes the movement of the racket. The racket has a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. The racket analysis method includes an estimation step of estimating the position of the ball hitting the striking surface based on the angular velocity of the racket when the ball hits the racket.

It is a schematic perspective view which shows an example of the structure of the racket analysis system which concerns on embodiment. It is a block diagram which shows the structure of the signal processing system of the analysis system of FIG. 3 (a) and 3 (b) are diagrams for explaining an impact detection method. 4 (a) and 4 (b) are diagrams for explaining a method of estimating the striking surface hit by the ball. 5 (a), 5 (b) and 5 (c) are diagrams for explaining a method of estimating the position where the sphere hits in the left-right direction. 6 (a) and 6 (b) are diagrams for explaining a method of estimating the position where the sphere hits in the vertical direction. It is a flowchart which shows an example of the procedure of the sensing process executed by the sensor device of the analysis system of FIG. It is a flowchart which shows an example of the procedure of the analysis processing executed by the analysis apparatus of the analysis system of FIG. It is a figure for demonstrating the impact detection method.

FIG. 1 is a schematic perspective view showing an example of the configuration of the analysis system 1 according to the embodiment.

The analysis system 1 has, for example, a sensor device 5 fixed to the racket 3 and an analysis device 7 that analyzes the movement of the racket 3 detected by the sensor device 5. Specifically, the analysis device 7 estimates, for example, which position of the racket 3 the sphere hits based on the detected value of the sensor device 5. In the following explanation, the position where the ball hits may be referred to as the impact position.

The racket 3 may be considered not to be included in the analysis system 1 or may be considered to be included. In the description of this embodiment, the former is used for convenience. Further, the combination of the racket 3 and the sensor device 5 may be conceptualized as a racket with a sensor.

(racket)
The racket 3 may be used for various sports, and in the description of the present embodiment, a racket for table tennis is taken as an example. Sports other than table tennis include, for example, tennis (hard and soft), badminton, squash and racquetball. As can be understood from the fact that badminton is mentioned, in the present disclosure, the "sphere" may be a ball hit by a racket, and is not only a ball in a narrow sense but also a non-sphere such as a shuttle (feather ball). Also includes.

The racket 3 has, for example, a striking portion 9 (head) for striking a ball and a grip 11 for a player to grip.

The striking portion 9 has, for example, a flat striking surface 9a for striking a ball. The racket 3 may be a type having a pair of front and back striking surfaces 9a (rubber from another viewpoint) (usually a shake hand type), or a type having only one striking surface 9a. (Usually a pen holder type) may be used. In the description of this embodiment, the former will be taken as an example. As is clear from the fact that the striking surface 9a of the racket 3 for table tennis may have irregularities and that the striking surface of the racket for tennis is composed of guts, the flat surface referred to here is. It does not have to be a plane in the strict sense.

The grip 11 is a long member having a thickness that can be gripped by a player. The striking portion 9 and the grip 11 are fixed to each other by being integrally formed with each other. In rackets other than table tennis, the grip and the striking part may be fixed to each other via a throat.

Assume a virtual line CL that is parallel to the striking surface 9a and passes through the center of the striking surface 9a (that is, the striking surface 9a is roughly bisected). At this time, the grip 11 extends along the virtual line CL on the outside of the striking surface 9a. “Along the virtual line CL” here does not require that the center line (not shown) of the grip 11 and the virtual line CL are located on the same line or parallel to each other. Within the rules and / or common sense in each competition, the center line of the grip 11 and the virtual line CL may be inclined to each other or may be deviated from each other.

(Coordinate system)
In FIG. 1, a relative coordinate system xyz (sensor coordinate system) fixed to the racket 3 and an absolute coordinate system XYZ (spatial coordinate system) fixed to the space in which the racket 3 can move are shown. The orientation of each axis of these coordinate systems may be set as appropriate.

In the description of this embodiment, it is assumed that the z-axis is parallel to the virtual line CL (generally parallel to the grip 11). It is assumed that the x-axis is orthogonal to the striking surface 9a. The y-axis shall be orthogonal to the x-axis and z-axis. In other words, the y-axis is parallel to the striking surface 9a and orthogonal to the z-axis. In the following description, for convenience, the y-axis direction may be referred to as the left-right direction, and the z-axis direction may be referred to as the up-down direction. The direction of the arrow shown around each axis indicates the positive or negative of the angular velocity around each axis used in the following description.

(Divided area of striking surface)
In FIG. 1, a division region R (RC, RE, RNE, RN, RNW, RW, RSW, RS and RSE) in which the striking surface 9a is divided into a plurality of parts is defined. The estimation of the position (impact position) where the ball in the striking surface 9a hits may be, for example, an estimation of which of the divided regions R in which the striking surface 9a is divided into an appropriate number. However, the impact position may be estimated in a plurality of regions of the division region R. Of course, instead of classifying such an impact position into any of the divided regions R, the coordinates of the impact position or the distance from the reference position to the impact position may be estimated.

The number of division areas R (the number of divisions of the striking surface 9a), the position of each division area R, the area ratio of each division area to the striking surface 9a, and the like may be appropriately set. In the illustrated example, the number of divisions is 9. More specifically, the striking surface 9a is roughly divided into three equal parts in the y-axis direction with respect to the length of the striking surface 9a (for example, the maximum length), and the length of the striking surface 9a in the z-axis direction (for example, the maximum length). It is roughly divided into three equal parts based on the length). That is, the striking surface 9a is divided into 3 × 3. The central division area R is defined as the division area RC. It is assumed that the other division regions R are the division regions RE, RNE, RN, RNW, RW, RSW, RS, and RSE in order from the division region R on the −y side in a counterclockwise direction.

Examples of division modes other than 9 divisions are given below. For example, the striking surface 9a may be divided only in the y-axis direction, and the number of divisions may be 2, 3, or 4 or more. Further, for example, the striking surface 9a may be divided only in the z-axis direction, and the number of divisions may be 2, 3, or 4 or more. Further, the striking surface 9a is divided in each of the y-axis direction and the z-axis direction, and is divided into 2 × 2, 2 × 3, 3 × 2, 2 × 4, 4 × 2, 3 × 4, 4 × 3 or 4 ×. It may be divided into four. From another viewpoint, when n and m are integers of 2 or more, the striking surface 9a may be divided into 1 × m, n × 1 or n × m. Further, the striking surface 9a may be divided by a concept different from the concept of vertically and horizontally divided, such as being divided into one circular region on the center side and a plurality of fan-shaped regions arranged around the circular region. Good.

As understood from the above, the number of divisions in the y-axis direction and the number of divisions in the z-axis direction may be the same or different. Further, a region embodying the center of the striking surface 9a in the y-axis direction or the z-axis direction (for example, the divided region RC in the illustrated example) may or may not exist.

(Overview of sensor device)
The sensor device 5 is fixed to the racket 3 and functions as an inertial sensor that detects at least an angular velocity. The mounting position, shape, size, and the like of the sensor device 5 with respect to the racket 3 may be appropriately set. In the illustrated example, the sensor device 5 is located at the end of the grip 11 opposite to the striking portion 9. Further, the sensor device 5 has substantially the same size as the end face of the grip 11 when viewed in the z-axis direction. The sensor device 5 may be detachable from one type of racket 3 or various rackets 3, or may be non-detachably fixed to the racket 3.

(Overview of analyzer)
The analysis device 7 includes a computer. In the illustrated example, the analyzer 7 is configured by a smart device. The smart device is, for example, a smartphone (illustrated example), a tablet, and a notebook computer, but is not limited to the above as long as it can estimate the impact position. The hardware and basic software of the computer (eg, OS (Operating System)) may be similar to various known ones (common in another respect). The analyzer 7 can be obtained by installing a predetermined application on a general computer.

(Structure of signal processing system)
FIG. 2 is a block diagram showing a configuration of a signal processing system of the analysis system 1. In this figure, the direction indicated by the arrow connecting the blocks indicates the direction in which the main signal is transmitted, but in reality, there may be a signal transmitted in the direction opposite to the arrow.

The sensor device 5 includes, for example, an angular velocity sensor 13 (gyro sensor) that detects the angular velocity of the racket 3, an acceleration sensor 15 that detects the acceleration of the racket 3, and a communication unit 17 that communicates with the analysis device 7. It has a detection processing unit 19 that processes a signal to be input or output.

The analysis device 7 processes, for example, an input unit 21 that accepts user operations, a display 23 that displays an image, a communication unit 25 that communicates with the sensor device 5, and signals that are input or output to these. It has an analysis processing unit 27 that performs the above and the like. The analysis processing unit 27 functions as a processing unit that estimates the impact position.

(Structure of each part of the sensor device)
The angular velocity sensor 13 is composed of, for example, a three-axis angular velocity sensor capable of detecting the angular velocities of each of the three axes of the x-axis, the y-axis, and the z-axis. As such an angular velocity sensor, for example, although not particularly shown, a combination of an angular velocity sensor that detects an angular velocity around the x-axis, an angular velocity sensor that detects an angular velocity around the y-axis, and an angular velocity sensor that detects an angular velocity around the z-axis is used. Can be mentioned. However, the angular velocity sensor 13 may be a combination of three sensors that detect the angular velocity around each axis of the coordinate system inclined to the relative coordinate system xyz. Even in this case, the angular velocities of the three axes of the x-axis, the y-axis, and the z-axis can be specified by the coordinate transformation by the angular velocity sensor 13 or an external device thereof. Therefore, it can be said that the angular velocity sensor 13 is a sensor capable of detecting the angular velocity of the x-axis, the y-axis, and / or the z-axis.

The configuration of the sensor that detects the angular velocity around each of the x-axis, y-axis, and z-axis may be, for example, various known configurations. For example, the angular velocity sensor for detecting the angular velocity around the z-axis in US Patent Application Publication No. 2019/265033 (hereinafter referred to as Prior Document 1; corresponding to International Publication No. 2018/021166) is shown in FIG. It may be used as an angular velocity sensor that detects the angular velocity around the z-axis defined in. The angular velocity sensor that detects the angular velocity around the y-axis in the prior document 1 may be used as the angular velocity sensor that detects the angular velocity around the y-axis defined in FIG. The angular velocity sensor for detecting the angular velocity around the y-axis in the prior document 1 is arranged so that the y-axis of the prior document 1 coincides with the x-axis of FIG. 1 and the x-axis of the prior document 1 coincides with the y-axis of FIG. , May be used as an angular velocity sensor that detects the angular velocity around the x-axis defined in FIG. The content of Prior Document 1 may be cited by reference in the present disclosure.

The acceleration sensor 15 is composed of, for example, a three-axis acceleration sensor capable of detecting the acceleration of each of the three axes of the x-axis, y-axis, and z-axis. The configuration of the 3-axis accelerometer may be of various known configurations. For example, a piezoresistive 3-axis accelerometer as disclosed in WO 2009/11840 may be used. The content of this document may be cited by reference in this disclosure. Further, for example, a capacitance type 3-axis acceleration sensor or a heat detection type 3-axis acceleration sensor may be used. The acceleration sensor 15 may be formed by combining two or more acceleration sensors that detect acceleration of one or two axes of an appropriate type. Similar to the angular velocity sensor 13, the direction in which the acceleration sensor 15 directly detects acceleration may be inclined with respect to the x-axis, y-axis, and z-axis.

The communication unit 17 has, for example, a configuration capable of communicating with at least the analysis device 7. The communication may only be able to transmit to the analysis device 7, or may be capable of transmitting and receiving to and from the analysis device 7. The communication may be wireless communication (illustrated example), wired communication, or switchable between wireless communication and wired communication. Examples of wireless communication include those using radio waves and those using infrared rays. In addition, examples of those using radio waves include those for short distances such as Bluetooth (registered trademark) and Wi-Fi (registered trademark).

The detection processing unit 19 has, for example, a CPU 19a (Central Processing Unit, processor) and a memory 19b. The memory 19b includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), an external storage device, and the like. The detection processing unit 19 is constructed by the CPU 19a executing a predetermined program stored in the memory 19b (ROM and external storage device). For example, the detection processing unit 19 acquires the detected angular velocity information from the angular velocity sensor 13, acquires the detected acceleration information from the acceleration sensor 15, and transmits the acquired information from the communication unit 17.

(Structure of each part of the analyzer)
The input unit 21 may have various known configurations. In the illustrated example, a plate-shaped pointing device constituting the touch panel 29 (see FIG. 1) is included. In addition, examples of the input unit 21 include those including a keyboard, a push button switch, and the like.

The display 23 can display an arbitrary image, for example, and is composed of a liquid crystal display or an organic EL (Electro-Luminescence) display. In the illustrated example, the touch panel 29 is configured by being combined with at least a part of the input unit 21.

Since the communication unit 25 communicates with the communication unit 17 of the sensor device 5, the above-mentioned explanation about the communication unit 17 may be appropriately referred to the communication unit 25. The communication unit 25 may only be able to receive from the communication unit 17, or may be capable of transmitting and receiving to and from the communication unit 17.

The analysis processing unit 27 has, for example, a CPU 27a (processor) and a memory 27b. The memory 27b includes, for example, a ROM, a RAM, an external storage device, and the like. The analysis processing unit 27 is constructed by the CPU 27a executing the program 31 stored in the memory 27b (ROM and external storage device). For example, the analysis processing unit 27 acquires information on the angular velocity and acceleration detected by the sensor device 5 via the communication unit 25, performs analysis based on the acquired information, and obtains the analysis result (for example, impact position). It is displayed on the display 23.

(Analysis items and analysis methods)
Hereinafter, the analysis items and the method thereof performed by the analysis device 7 (more specifically, the analysis processing unit 27) based on the detection result of the sensor device 5 will be described. In the following, basically, a case where the ball hits the striking surface 9a on the −x side will be described as an example. Similarly, the case where the ball hits the striking surface 9a on the + x side may be estimated.

(Impact estimation)
The analysis device 7 may estimate the presence or absence of an impact (the ball is hit by the hitting surface 9a. From another viewpoint, the ball hits the hitting surface 9a) and the time of impact. The estimation method is as follows, for example.

FIG. 3A is a diagram showing an example of a time-dependent change in acceleration in the x-axis direction detected by the acceleration sensor 15. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis Ax (G) indicates acceleration in the x-axis direction. The symbol G is a unit for expressing how many times the standard gravity is. This figure is obtained by an experiment in which a player swings a racket 3 and hits a ball.

In this figure, the change with time when the ball is struck by the striking surface 9a on the −x side of FIG. 1 is shown. As indicated by the extremum P1, when the ball hits the striking surface 9a, the acceleration to the side opposite to the side facing the striking surface 9a suddenly rises. From another point of view, various parameters (for example, absolute value, rate of change and / or frequency) are different between the acceleration at the time of impact and the acceleration at the time of shaking the racket 3. Therefore, by detecting the rising (or falling) of the acceleration in the x-axis direction, it is possible to estimate the presence or absence of the impact and the time point at that time.

FIG. 3B is a diagram showing an example of a time-dependent change in the angular velocity around the z-axis detected by the angular velocity sensor 13. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis Gz (dps: degree per second) indicates the angular velocity around the z axis. This figure is obtained from the same experiment as in FIG. 3 (a).

In this figure, the change with time when the ball is struck by the region on the + y side (divided region RW) of the striking surface 9a on the −x side in FIG. 1 is shown. As indicated by the extremum P2, when the ball hits a position of the striking surface 9a away from the virtual line CL, the striking surface 9a rotates around the virtual line CL, and the angular velocity suddenly drops (or rises). ). From another point of view, various parameters (for example, absolute value, rate of change and / or frequency) are different between the angular velocity at the time of impact and the angular velocity at the time of swinging the racket 3. Therefore, by detecting such rising and falling angular velocities around the z-axis, the presence or absence of impact and the time point can be estimated.

FIG. 9 is a diagram showing an example of a time-dependent change in acceleration in the z-axis direction detected by the acceleration sensor 15. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis Az (G) indicates acceleration in the z-axis direction. This figure is obtained from the same experiment as in FIGS. 3 (a) and 3 (b).

As shown in this figure, when a swing is made to hit a ball, the acceleration to the + z side temporarily increases due to centrifugal force (t is in the range of approximately 2.7 to 3.2). This acceleration drops momentarily, as indicated by the extremum P3. It is considered that this is because the centrifugal force also changed due to a slight change (decrease) in the swing speed at the moment when the ball hit. Therefore, by detecting such a momentary change in acceleration in the z-axis direction, the presence or absence of impact and the time point can be estimated.

Here, three types are exemplified as physical quantities (acceleration and / or angular velocity) for estimating impact, but it is also possible to estimate impact based on physical quantities related to other axes. Specifically, an angular velocity around the x-axis, an angular velocity around the y-axis, or an acceleration in the y-axis direction may be used. In estimating the impact, only one of a plurality of physical quantities (for example, acceleration in the x-axis direction, angular velocity around the z-axis, and acceleration in the z-axis direction) may be used, or two or more are combined. It may be used. An example of the latter is a method of estimating an impact when it can be estimated that there was an impact based on at least one of two or more physical quantities, and conversely, an impact is based on all of two or more physical quantities. There is a method of estimating that there was an impact only when it could be estimated that there was.

In estimating the presence or absence of impact, various parameters related to acceleration (or angular velocity; the same shall apply hereinafter in this paragraph) may be used. For example, focusing only on the absolute value of acceleration, it may be estimated that there was an impact when the acceleration exceeded a predetermined threshold value. Further, for example, it may be estimated that the impact occurs when the absolute value of the rate of change of acceleration (jerk) exceeds a predetermined threshold value in place of the absolute value of acceleration or in addition to the absolute value of acceleration. Further, for example, instead of or in addition to the above, it may be estimated that the impact occurred when the frequency of the acceleration vibration was higher than the predetermined value.

In estimating the time of impact, any time of change in the acceleration (or angular velocity; the same shall apply hereinafter in this paragraph) that was the basis for determining that there was an impact may be used as the time of impact. It may be appropriately set within a range that is practically acceptable in the analysis. For example, the time point at which the absolute value of acceleration becomes the largest may be set as the time point of impact, such as the time point of the extreme value P1 in FIG. Further, for example, the time point at which the sudden increase in the absolute value of acceleration starts may be set as the time point of impact. Further, for example, the time point of impact may be treated as having a certain time length within a practically permissible range in the analysis, instead of one point of the literal time. For example, the range from the start of the rapid increase in the absolute value of acceleration to the maximum absolute value of acceleration may be treated as the time of impact.

The estimation may be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI (Artificial Intelligence) technology. In the former case, the determination procedure may be set based on the above-mentioned estimation method. In the latter case, for example, using teacher data in which at least one of the above-mentioned various parameters of acceleration in the x-axis direction (and / or angular velocity around the z-axis) is input and the presence / absence of impact and the time point are output. It may be estimated by generating a learning model and incorporating this learning model into the program 31.

The presence / absence and time point of the impact included in the teacher data can be specified by, for example, shooting the racket 3 with a video camera, or teacher data creation provided with an appropriate sensor (for example, a plate-shaped pointing device extending over the striking surface 9a). It may be specified by using a racket for. The AI technology will also be touched upon for front and back estimation, left-right estimation, central estimation, vertical estimation, etc., which will be described later. The information output from the teacher data at this time may also be specified by the photographing and the sensor as described above. Depending on the information, it may be visually identified.

(Estimation of front and back)
In the present embodiment, the racket 3 has a pair of front and back striking surfaces 9a. In such a case, the analysis device 7 may estimate which of the pair of striking surfaces 9a the ball hits. The estimation method is as follows, for example.

FIG. 4A is a diagram showing an example of a change over time in the angular velocity detected by the angular velocity sensor 13. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis ω (dps) indicates angular velocity. ωx, ωy and ωz indicate the angular velocities around the x-axis, y-axis and z-axis, respectively. This figure is obtained by an experiment in which a player swings a racket 3 and hits a ball.

When the angular velocities of the three axes are obtained from the time-series data in this way, the time-series data indicating the posture of the racket 3 can be obtained based on the time-series data. Postures may be represented, for example, by Euler angles. Euler angles use, for example, roll angles, pitch angles and yaw angles. In this Euler angle, for example, when the absolute coordinate system XYZ is rotated in order around the Z axis, the Y axis, and the X axis, the angle around the Z axis is the yaw angle, and the angle around the Y axis is the pitch angle and X. The angle around the axis may be the roll angle.

Specifically, for example, the angular velocity around each axis of the relative coordinate system xyz (sensor coordinate system) is converted into the angular velocity around each axis of the absolute coordinate system XYZ (spatial coordinate system). The angular velocity in this absolute coordinate system XYZ is integrated to obtain the roll angle, pitch angle, and yaw angle. Such an operation is repeated along the time axis. Calculation formulas (determinants) for obtaining such Euler angles are known in various fields such as robots and drones, and these formulas may be used. The initial value of Euler angles when calculating along the time axis may be an arbitrary value by, for example, specifying the posture of the racket 3 at the start of measurement in advance. Alternatively, the initial value may be specified by combining the angular velocity sensor with another sensor. For example, the initial values of the roll angle and the pitch angle may be specified based on the acceleration detected by the acceleration sensor 15 (in another viewpoint, the gravitational acceleration). The initial value of the yaw angle may be specified based on the detection value of the geomagnetic sensor, for example, by providing the sensor device 5 with a geomagnetic sensor. Such sensor fusions are also known and may be utilized.

FIG. 4B is a diagram showing an example of changes over time in Euler angles calculated as described above. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis indicates Euler angles. φ indicates the roll angle, θ indicates the pitch angle, and ψ indicates the yaw angle.

When the time series data of Euler angles are obtained in this way and the presence or absence of impact and the time point are specified as described above, the posture of the racket 3 at the time of impact can be specified. Based on this specified posture, it is possible to specify which of the pair of striking surfaces 9a the ball hits.

For example, at the time of impact, the hitting surface 9a on which the ball hits is usually facing the opponent's court. Therefore, for example, when considering the normal of one of the striking surfaces 9a extending to the side facing the one striking surface 9a, this normal has a component in the alignment direction from one's own court to the other's court. In this case, it can be estimated that the ball hits the one hitting surface 9a. In other words, when Euler angles are included in a predetermined first range, it can be estimated that the ball hits one of the striking surfaces 9a, and Euler angles are included in a second range different from the first range. When it is, it can be estimated that the ball hits the other striking surface 9a.

In the above, we focused only on Euler angles. However, in addition to Euler angles, other detected values may be considered. For example, even if the position of the racket 3 in the absolute coordinate system XYZ is detected based on the detected value of the acceleration sensor 15, the first range and the second range are changed according to the change in the positional relationship between the racket 3 and the mating court. Good.

The estimation may be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI technology. In the former case, the determination procedure may be set based on the above-mentioned estimation method. In the latter case, for example, using teacher data that inputs Euler angles at the time of impact (and the position of the racket 3 if necessary) and outputs information that identifies the striking surface 9a hit by the ball. It may be estimated by generating a learning model and incorporating this learning model into program 31. As already described, the information for identifying the hitting surface 9a hit by the ball, which is included in the teacher data, may be obtained by visual inspection, photographing, a sensor, or the like.

(Right and left estimation)
The analysis device 7 may estimate the position of the position where the ball hits the striking surface 9a (impact position) in the y-axis direction. For example, the analysis device 7 may estimate whether the impact position is on the positive side or the negative side in the y-axis direction with respect to the central side (for example, the virtual line CL) of the striking surface 9a. The estimation method is as follows, for example.

5 (a) to 5 (c) are diagrams showing an example of changes in the angular velocity around the z-axis detected by the angular velocity sensor 13 with time. In this figure, the horizontal axis t (msec) indicates time, and the vertical axis ωz (dps) indicates angular velocity. All of these figures show the change with time when the ball hits the striking surface 9a on the −x side. FIG. 5A shows the change with time when the sphere hits the divided region RE on the −y side. FIG. 5B shows the change with time when the sphere hits the divided region RC at the center in the y-axis direction. FIG. 5C shows the change with time when the sphere hits the divided region RW on the + y side. These figures are obtained by an experiment in which a ball is hit against a racket 3 held in a fixed position by a player.

As described in the explanation of impact estimation, when a ball hits a position of the striking surface 9a away from the virtual line CL, the striking surface 9a rotates around the virtual line CL, and the angular velocity suddenly drops or rises. The rotation direction at this time is when it hits the region on the + y side with respect to the virtual line CL (for example, the division region RW) and when it hits the region on the −y side with respect to the virtual line CL (for example, the division region RE). And the opposite. As a result, as shown in FIGS. 5 (a) and 5 (c), the angular velocity around the z-axis at the time of impact (for example, the maximum value in FIG. 5 (a) and the minimum value in FIG. 5 (b)) are set. The positive / negative of (see) is opposite when it hits the + y side region and when it hits the −y side region. Therefore, it can be estimated whether the impact position is on the −y side or the + y side with respect to the center side of the striking surface 9a based on the positive or negative of the angular velocity around the z-axis at the time of impact.

The estimation may be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31 based on the above estimation method. However, it may be realized by AI technology. For example, as input of teacher data, not only the positive / negative of the angular velocity at the time of impact (and information for identifying the striking surface 9a hit by the ball as necessary) but also other parameters of the angular velocity and acceleration can be used for manufacturing. It is expected that error corrections that were not noticed by the person will be made. As already described, the information included in the teacher data as an output, that is, the information on whether the impact position is on the −y side or the + y side may be obtained by visual inspection, photographing, a sensor, or the like.

(Estimation of distance from the center)
5 (b) corresponding to the case where the sphere hits the center in the y-axis direction, and FIGS. 5 (a) and 5 (c) corresponding to the case where the sphere hits a position away from the center in the y-axis direction. As can be understood from the comparison, the distance of the impact position from the virtual line CL and the absolute value of the angular velocity around the z-axis are correlated. Specifically, the longer the distance from the virtual line CL, the larger the absolute value of the angular velocity. Therefore, the position of the impact position on the y-axis may be estimated based on the absolute value of the angular velocity.

As described above, the estimation of the position of the impact position in the y-axis direction may be to estimate which of the divided regions R in which the striking surface 9a is divided in the y-axis direction the ball hits, or the impact. It may estimate the y-axis coordinates of the position.

When estimating which of the division regions R the sphere hits in the y-axis direction, for example, the number of divisions in the y-axis direction may be 3 or more. This is because when the number of divisions is two, it is sufficient to estimate the left and right based on the positive and negative of the angular velocity at the time of the above impact. The estimation of which of the divided regions R the sphere hits may be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI technology. Good.

When the estimation procedure is directly specified, for example, the larger the absolute value of the angular velocity, the more the procedure for estimating that the outer region is hit is specified. Specifically, for example, the manufacturer (which may be a user) stores the threshold value of the number corresponding to the number of divisions in the y-axis direction in the memory 27b. For example, if the number of divisions is 3 or 4, the threshold value is 1, and if the number of divisions is 5, the threshold value is 2. Then, the analysis processing unit 27 estimates which region is hit depending on whether or not the absolute value of the angular velocity around the z-axis at the time of impact exceeds the threshold value. In the example shown in FIG. 1, when the absolute value of the angular velocity is smaller than a predetermined threshold value, it is estimated that the sphere hits one of the division regions R (RC, RN and RS) on the central side, and in other cases, it is estimated that the sphere hits. , It may be presumed that the sphere hits any of the outer division regions R (RE, RNE, RNW, RW, RSW and RSE).

In the case of being realized by AI technology, for example, using teacher data in which the absolute value of the angular velocity around the z-axis at the time of impact is input and the information on which region divided in the y-axis direction is hit is output. A learning model may be generated and the learning model may be incorporated into the program 31. As already described, the information on which area is hit, which is included in the teacher data, may be obtained by visual inspection, photographing, a sensor, or the like.

The estimation of the y-axis coordinates of the impact position may also be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI technology.

In the former case, for example, the manufacturer indicates the correlation between the distance from the virtual line CL to the impact position (y-axis coordinates from another viewpoint) and the absolute value of the angular velocity (for example, an approximate expression or a map). Is stored in the memory 27b. Then, the analysis processing unit 27 (CPU27a) estimates the y-axis coordinates corresponding to the detected angular velocities with reference to the above-mentioned information indicating the correlation.

In the case of being realized by AI technology, for example, using teacher data in which the absolute value of the angular velocity around the z-axis is input and the distance of the impact position from the virtual line CL (y-axis coordinate in another viewpoint) is output. It may be estimated by generating a learning model and incorporating this learning model into the program 31. As already described, the distance of the impact position from the virtual line CL, which is included in the teacher data, may be obtained by photographing or a sensor or the like.

The distance of the impact position from the virtual line CL (y-axis coordinate from another viewpoint) is not only the absolute value of the angular velocity around the z-axis, but also, for example, the frequency of vibration of the angular velocity around the z-axis and the angular velocity around the z-axis. It also affects the time it takes to decay to a given size. Therefore, in any of the above cases, the values of these parameters may be used in addition to the absolute value of the angular velocity or in place of the absolute value of the angular velocity. For example, the values of these parameters may be used as input for teacher data.

(Estimation of top and bottom)
The analysis device 7 may estimate the position of the position where the ball hits the striking surface 9a (impact position) in the z-axis direction. The estimation method is as follows, for example.

6 (a) and 6 (b) are diagrams showing an example of changes in the angular velocity around the y-axis detected by the angular velocity sensor 13 with time. In this figure, the horizontal axis t (sec) indicates time, and the vertical axis ωy (dps) indicates angular velocity. All of these figures show the change with time when the ball hits the striking surface 9a on the −x side. FIG. 6A shows the change with time when the sphere hits the divided region RN on the + z side. FIG. 6B shows the change with time when the sphere hits the divided region RS on the −z side. These figures are obtained by an experiment in which a player swings a racket 3 and hits a ball.

When the ball hits the striking surface 9a, the striking portion 9 rotates parallel to the y-axis and around a rotation axis (not shown) located on the grip 11 side. As a result, the angular velocity around the y-axis suddenly drops or rises. As can be understood from the comparison between FIGS. 6 (a) and 6 (b), the distance of the impact position from the grip 11 (in other words, the z-axis coordinate of the impact position) and the amplitude of the angular velocity around the y-axis are It correlates with the time Ta until it decays to a predetermined magnitude. Specifically, the longer the distance from the grip 11, the longer the time Ta. Therefore, the position of the impact position on the z-axis may be estimated based on the time Ta.

The beginning of time Ta may be, for example, the time of impact determined by impact estimation. However, the start of time Ta may be set based on the angular velocity around the y-axis. The end of the time Ta is when the amplitude of the angular velocity around the y-axis is attenuated to a predetermined magnitude (threshold value) as described above, and this threshold value may be appropriately set based on an experiment or the like. For example, in the illustrated example, it is shown that the threshold value may be approximately 0.

As described above, the estimation of the position of the impact position in the z-axis direction may be to estimate which of the divided regions R in which the striking surface 9a is divided in the z-axis direction the sphere hits, or the impact. It may be used to estimate the z-axis coordinates of the position.

When estimating which of the division regions R the sphere hits in the z-axis direction, for example, the number of divisions in the z-axis direction may be 2 or more (3 in the illustrated example). The estimation of which of the divided regions R the sphere hits may be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI technology. Good.

When the estimation procedure is specified, for example, the longer the time Ta, the more the procedure for presuming that the area away from the grip 11 is hit is specified. Specifically, for example, the manufacturer (which may be a user) stores the threshold value in the memory 27b, which is one less than the number of divisions in the z-axis direction. For example, if the number of divisions is two, the threshold value is one, and if the number of divisions is three, the threshold value is two. Then, the analysis processing unit 27 estimates which region is hit depending on whether or not the time Ta until the angular velocity vibration generated at the time of impact is attenuated to a certain magnitude exceeds the threshold value. In the example shown in FIG. 1, when the absolute value of the angular velocity is smaller than the first threshold value, it is presumed that the ball hits any of the division regions R (RSW, RS and RSE) closest to the grip 11. On the other hand, when the absolute value of the angular velocity is equal to or higher than the first threshold value and smaller than the second threshold value (> first threshold value), it is presumed that the central divided region R (RE, RC and RW) is hit. When the absolute value of the angular velocity is other than the above, it may be estimated that the ball hits any of the divided regions R (RNE, RN and RNW) farthest from the grip 11.

In the case of being realized by AI technology, for example, a learning model is generated using teacher data in which time Ta is input and information on which region divided in the z-axis direction is hit is output, and this learning model is used. It may be incorporated into program 31. As already described, the information on which area is hit, which is included in the teacher data, may be obtained by visual inspection, photographing, a sensor, or the like.

The estimation of the z-axis coordinates of the impact position may also be realized, for example, by the manufacturer of the analysis system 1 directly defining the estimation procedure in the program 31, or by AI technology.

In the former case, for example, the manufacturer stores information (for example, an approximate expression or a map) indicating the correlation between the distance of the impact position from the grip 11 (z-axis coordinates from another viewpoint) and the time Ta in the memory 27b. To memorize. Then, the analysis processing unit 27 (CPU27a) estimates the z-axis coordinates corresponding to the detected angular velocity with reference to the above-mentioned information indicating the correlation.

In the case of being realized by AI technology, for example, a learning model is generated using teacher data in which time Ta is input and the distance from the grip 11 of the impact position (z-axis coordinate in another viewpoint) is output. This learning model may be estimated by incorporating it into program 31. As already described, the distance from the grip 11 included in the teacher data may be obtained by photographing or a sensor or the like.

The distance of the impact position from the grip 11 (z-axis coordinate from another viewpoint) affects not only the time Ta but also, for example, the absolute value of the angular velocity around the y-axis and the vibration frequency of the angular velocity around the y-axis. Therefore, in any of the cases described above, the values of these parameters may be used in addition to or in place of the time Ta. For example, the values of these parameters may be used as input for teacher data.

(Correction of angular velocity)
The speed of the racket 3 at the time of impact affects the magnitude of various angular velocities at the time of impact. As a result, the accuracy of estimating the impact position based on the angular velocity at the time of impact described above may decrease. For example, if the speed of the racket 3 at the time of impact is high, the absolute value of the angular velocity around the z-axis becomes large even though the distance from the virtual line CL is short, and it is erroneously determined that the distance from the virtual line CL is long. May be done. Therefore, the angular velocity at the time of impact may be corrected, and the various estimations described above may be performed using the corrected angular velocity.

Specifically, for example, the following corrections may be made. First, the speed of the racket 3 at the time of impact is specified based on the detected value of the acceleration sensor 15. Of this velocity, the component in the direction in which the striking surface 9a hit by the ball is facing is specified. Correction may be made so that the larger this component is, the smaller the absolute value of the angular velocity around the y-axis is, the smaller the absolute value of the angular velocity around the z-axis is, and / or the time Ta is shortened. The specific correction amount may be derived theoretically or may be obtained by an experiment or the like.

Also, the ball velocity at the time of impact (strictly, immediately before the impact) also affects the magnitude of various angular velocities at the time of impact, similar to the speed of the racket 3 described above. Therefore, the ball speed (vector quantity or absolute value) may be measured by a video camera and / or a speed measuring device, and correction may be made based on the ball speed. Even if such a correction is not made, for example, in the analysis of the serve and the analysis of the practice of hitting a ball at a constant velocity emitted from the machine, it is unlikely that the accuracy of estimating the impact position will decrease.

(Other estimation methods)
In the above description, a method of estimating the impact position by dividing it into a plurality of items, such as separating the estimation of the position in the y-axis direction and the estimation of the position in the z-axis direction, has been shown. However, two or more of the above estimation items may be estimated together. For example, when using AI technology, time-series data of angular velocity and acceleration for each of the three axes is input, and the time of impact and impact position (which of the nine divided regions R belongs to, or yz coordinates) are output. A learning model may be generated using the teacher data to be used. As shown above, there is a correlation between the angular velocity and acceleration and the time and position of impact, and it is clear that an appropriate learning model can be obtained.

(flowchart)
An example of the procedure executed by the detection processing unit 19 (CPU 19a) of the sensor device 5 and the analysis processing unit 27 (CPU 27a) of the analysis device 7 will be described with reference to the flowchart. The flowchart shown below is drawn to facilitate understanding of the concept of processing, and does not necessarily match the actual processing.

The analysis of the movement of the racket may be performed, for example, after the practice (or a match, etc., the same shall apply hereinafter), or may be performed (generally) in real time during the practice. In the former case, for example, the time series data of the detected value is accumulated in the sensor device 5 during the practice, and the time series data of the detected value is transmitted from the sensor device 5 to the analysis device 7 after the practice for analysis. It's okay. Further, for example, the detection value data may be sequentially transmitted from the sensor device 5 to the analysis device 7 during the practice, and the analysis may be performed after the practice. Further, in the case of performing in real time, for example, the detection value data may be sequentially transmitted from the sensor device 5 to the analysis device 7 during the practice, and the analysis device 7 may perform the sequential analysis.

The following is an example of a flowchart in which the analysis is performed in real time during practice, which is the most complicated of the above aspects. The processing when the analysis is performed after the practice can be inferred from the processing when the analysis is performed in real time.

FIG. 7 is a flowchart showing an example of the procedure of the sensing process executed by the detection processing unit 19 of the sensor device 5. This process is started, for example, when an operation (for example, a switch) of the sensor device 5 (not shown) is performed.

In step ST1, the detection processing unit 19 determines whether or not the predetermined time T1 has elapsed. This time T1 is a sampling period for detecting the angular velocity and acceleration, as will be understood from the following description. The length of time T1 may be set by the manufacturer or user. The detection processing unit 19 waits for a negative determination (repeats step ST1), and proceeds to step ST2 for an affirmative determination. In step ST2, the detection processing unit 19 acquires information on the detected value of the angular velocity from the angular velocity sensor 13. Further, in step ST3, the detection processing unit 19 acquires information on the detected value of the acceleration from the acceleration sensor 15.

In step ST4, the detection processing unit 19 determines whether or not the predetermined time T2 has elapsed. This time T2 is a cycle for transmitting information on the detected values of the angular velocity and the acceleration, as will be understood from the following description. The time T2 is set to be twice or more the time T1. That is, in the process shown here, in order to reduce the burden of communication, the information of the detected values at a plurality of time points is collectively transmitted. However, transmission may be performed in the same cycle as the detection sampling cycle without providing step ST4. The length of time T2 may be set by the manufacturer or user. The detection processing unit 19 returns to step ST1 when a negative determination is made, and proceeds to step ST5 when an affirmative determination is made. In step ST5, the detection processing unit 19 transmits untransmitted information on the detected values of the angular velocity and the acceleration acquired so far.

In step ST6, the detection processing unit 19 determines whether or not the end condition for ending the sensing process is satisfied. The end condition is, for example, that an operation has been performed on an operation unit (not shown) of the sensor device 5. Then, the detection processing unit 19 returns to step ST1 to continue the process when the negative determination is made, and ends the process when the affirmative determination is made.

FIG. 8 is a flowchart showing an example of the procedure of the analysis process executed by the analysis processing unit 27 of the analysis device 7. This process is started, for example, when a predetermined operation is performed on the input unit 21 (not shown) of the analysis processing unit 27. Note that initialization processing such as setting the initial values of Euler angles is omitted here.

In step ST11, the analysis processing unit 27 determines whether or not data has been received from the sensor device 5. The analysis processing unit 27 waits for a negative determination (repeats step ST11), and proceeds to step ST12 for an affirmative determination. Here, the sensor device 5 actively transmits data and the analysis processing unit 27 passively receives the data, but the analysis processing unit 27 transmits data at a predetermined cycle (time T2). You may instruct.

As can be understood from the following explanation, here, the estimation regarding the impact is repeated in the cycle of receiving the data. However, the estimation may be repeated in a cycle longer than the cycle in which the data is received.

In step ST12, the analysis processing unit 27 estimates the posture of the racket 3 based on the information obtained in step ST11, as described with reference to FIGS. 4 (a) and 4 (b). In this estimation, for example, the posture at each time point (in time T1 increments) within the time T2 corresponding to the data received this time is estimated.

In step ST13, the analysis processing unit 27 specifies the presence / absence and the time point of impact as described with reference to FIGS. 3 (a), 3 (b) and 9. In this estimation, for example, the presence or absence of impact is estimated in the time series data of at least the time T2 minutes received this time. In addition, not only the data received this time but also the data received before that (data of the previous time T2, if necessary) so that the impact generated near the boundary between the current time T2 and the previous time T2 can be estimated. Further previous data) may be used.

As described above, when time-series data of the detected value is required when estimating the impact, the previous detected value may be used as appropriate. The same applies to the other estimates described below. This also applies when the time T2 is not set and the serial estimation is performed at the sampling cycle of the detected value at the time T1.

Further, in step ST13, the analysis processing unit 27 proceeds to step ST14 when it is estimated that there is an impact, and returns to step ST11 when it is estimated that there is no impact.

As can be understood from the following explanation, here, if it is estimated that there was an impact, various other estimations are sequentially performed. However, depending on the estimation item, not only the information at the time of impact but also the information before and after it may be required (for example, specifying the time Ta). Therefore, in the actual processing, various estimations may be performed in parallel with other estimation processing and the next impact estimation processing, for example, starting with the estimation that there was an impact as a trigger.

In step ST14, the analysis processing unit 27 estimates the estimation result at the time of impact in step ST13 and the posture of the racket 3 in step ST13, as described with reference to FIGS. 4 (a) and 4 (b). Based on the result, the posture of the racket 3 at the time of impact is estimated. As a result, the analysis processing unit 27 estimates which of the pair of front and back striking surfaces 9a the ball hits.

In step ST15, the analysis processing unit 27 calculates the speed of the racket 3 at the time of impact based on the estimation result at the time of impact in step ST13 and the detected value of the acceleration sensor 15. Next, the analysis processing unit 27 corrects the detected value of the angular velocity based on the calculated velocity and the estimation result of the posture at the time of impact. In various estimations described below, this corrected angular velocity may be used, but it may not be used.

In step ST16, the analysis processing unit 27 specifies the angular velocity around the z-axis at the time of impact based on the estimation result at the time of impact in step ST13. Then, the analysis processing unit 27 is based on the above-specified angular velocity and the estimation result of the striking surface hit by the ball in step ST14, as described with reference to FIGS. 5 (a) and 5 (c). Then, it is estimated whether the ball hits the + y side or the −y side.

In step ST17, the analysis processing unit 27 virtualizes the impact position based on the absolute value of the angular velocity around the z-axis at the time of impact, as described with reference to FIGS. 5 (a) to 5 (c). Estimate the distance from line CL.

In step ST18, the analysis processing unit 27 specifies the time Ta at which the amplitude of the angular velocity vibration around the y-axis generated at the time of impact is attenuated to a predetermined magnitude based on the estimation result at the time of impact in step ST13. .. Then, the analysis processing unit 27 estimates the distance of the impact position from the grip 11 based on the time Ta, as described with reference to FIGS. 6 (a) and 6 (b).

In step ST19, the analysis processing unit 27 has a plurality of impact positions (this implementation) based on the above front and back estimation (step ST14), left / right estimation (step ST16), central estimation (step ST17), and vertical estimation (step ST18). It is determined (classified) which of the nine division regions R the form belongs to.

In step ST20, the analysis processing unit 27 displays any of the estimation results so far on the display 23. The displayed contents may be appropriate. For example, in a figure schematically showing a plurality of division regions R, it may be indicated which division region R the latest impact position belongs to, or the frequency of impact for each division region R may be indicated.

In step ST21, the analysis processing unit 27 determines whether or not the end condition for ending the analysis process is satisfied. The end condition is, for example, that a predetermined operation has been performed on the input unit 21. Then, the analysis processing unit 27 returns to step ST11 to continue the process when the negative determination is made, and ends the process when the affirmative determination is made.

As described above, the analysis system 1 according to the present embodiment is a racket analysis system that analyzes the movement of the racket 3. The racket 3 has a striking portion 9 and a grip 11. The striking portion 9 has, for example, a flat striking surface 9a. The grip 11 extends outside the striking surface 9a along a virtual line CL that is parallel to the striking surface 9a and passes through the center of the striking surface 9a. The analysis system 1 estimates the position of the ball on the striking surface 9a based on the detection values of the angular velocity sensor 13 fixed to the racket 3 and the angular velocity sensor 13 when the ball hits the racket 3. It has a processor (CPU 27a).

From another point of view, the analysis device 7 according to the present embodiment is a racket analysis device that analyzes the movement of the racket 3. The analysis device 7 has a processor (CPU 27a) that estimates the position of the hitting surface 9a where the ball hits, based on the angular velocity of the racket 3 when the ball hits the racket 3.

From yet another point of view, the program 31 according to the present embodiment is a racket analysis program for analyzing the movement of the racket 3. The program 31 causes the computer (CPU 27a and memory 27b) to execute an estimation step (for example, at least one of steps ST16 to ST18). The estimation step is a process of estimating the position where the ball hits the striking surface 9a based on the angular velocity of the racket when the ball hits the racket 3.

From yet another viewpoint, the racket analysis method according to the present embodiment is an estimation step of estimating the position where the ball hits the striking surface 9a based on the angular velocity of the racket 3 when the ball hits the racket 3. have.

Therefore, for example, it is not necessary to provide a plate-shaped pointing device on the racket 3. As a result, for example, the probability that the weight of the racket with a sensor deviates from the weight of the racket 3 is reduced. Further, for example, unlike the case where the impact position is specified by a video camera, the impact position can be specified even if the ball is hidden by the striking portion 9.

Further, in the present embodiment, when the axis parallel to the virtual line CL is the z-axis and the axis parallel to the striking surface 9a and orthogonal to the z-axis is the y-axis, the angular velocity sensor 13 determines the angular velocity around the z-axis. It is detectable. Based on the positive and negative of the angular velocity around the z-axis when the ball hits the racket 3, the CPU 27a has the position where the ball hits the positive side in the y-axis direction with respect to the center side (for example, the virtual line CL) of the striking surface 9a. And estimate which is the negative side.

In this case, for example, it is convenient to estimate whether the impact position is on the positive side or the negative side in the y-axis direction. As a result, for example, the burden for estimation can be reduced. As a result, for example, it becomes easy to perform analysis in real time by a general smart device.

Further, in the present embodiment, the angular velocity sensor 13 can detect the angular velocity around the z-axis. The CPU 27a estimates that the smaller the angular velocity around the z-axis when the ball hits the racket 3, the more the position where the ball hits is on the center side of the striking surface 9a in the y-axis direction. Specifically, for example, the CPU 27a estimates that the sphere hits any of the divided regions RC, RN, and RS when the angular velocity around the z-axis is smaller than a predetermined threshold value.

Here, in the competition using the racket, for example, the rotation around the z-axis of the racket is actively used as compared with the competition using the bat. Therefore, in order to detect this rotation, it is expected that an angular velocity sensor 13 capable of detecting the angular velocity around the z-axis with high accuracy will be provided. As a result, for example, when the position of the impact position in the y-axis direction is estimated based on the angular velocity around the z-axis as described above, it is expected that the accuracy of the impact position estimation will be improved as the angular velocity sensor 13 becomes more accurate. To.

The CPU 27a is a divided region in which the position where the ball hits divides the striking surface 9a into three or more in the y-axis direction based on the positive and negative values and the absolute value of the angular velocity around the z-axis when the ball hits the racket 3. You may estimate where it is located.

In this case, for example, when the impact position is repeatedly estimated and statistical processing is performed for a large number of estimated impact positions, statistical processing is performed for the divided region to which the impact position belongs. As a result, the burden of calculation can be reduced as compared with a mode in which statistical processing is performed on the coordinates of the impact position (such a mode is also included in the technique according to the present disclosure). Further, for example, when the accuracy of estimating the coordinates of the impact position is low and the previous impact position and the current impact position are close to each other, the relative relationship between the two impact positions is opposite to the actual relative relationship. Relative relationships may be estimated. Then, when such a relative relationship opposite to the actual one is presented to the user, there is a possibility that the improvement of the user's skill is rather hindered. However, when the divided region to which the impact position belongs is estimated and presented, since the minute error of the coordinates is not presented to the user, the possibility that the above-mentioned inconvenience will occur is reduced.

Further, in the present embodiment, the angular velocity sensor can detect the angular velocity around the y-axis. In the CPU 27a, the longer the time Ta at which the amplitude of the angular velocity vibration around the y-axis generated when the ball hits the racket 3 is attenuated to a predetermined magnitude, the more the position where the ball hits is separated from the grip 11 in the z-axis direction. Presumed to be. Specifically, for example, the CPU 27a estimates that the longer the time Ta, the more the ball hits the divided region R, which has a relatively longer distance from the grip 11.

Here, the maximum value or the minimum value of the angular velocity generated at the time of impact reflects the inertial force momentarily received by the racket 3. On the other hand, one of the factors causing the difference in the length of time Ta is that the speed at which the vibration energy is dissipated into the grip 11 and the player's hand differs depending on the distance from the grip 11. That is, time Ta is not a parameter governed only by the momentary inertial force. From this, for example, if the impact position is estimated based on the time Ta as described above, the influence of the ball speed on the estimation can be reduced.

Based on the time Ta, the CPU 27a may estimate which of the divided regions where the hitting surface 9a is divided into two or more in the z-axis direction.

In this case, for example, the same effect as that of estimating which of the divided regions in which the striking surface 9a is divided in the y-axis direction belongs to the impact position is produced. For example, the burden of calculation is reduced. Further, for example, with respect to two or more impact positions, the possibility of presenting the user with a relative relationship opposite to the actual one is reduced.

Further, in the present embodiment, the angular velocity sensor 13 can detect the angular velocity around the x-axis, the angular velocity around the y-axis, and the angular velocity around the z-axis. The CPU 27a estimates the posture of the striking portion 9 in the absolute coordinate system based on the angular velocity around the x-axis, the angular velocity around the y-axis, and the angular velocity around the z-axis. Then, the CPU 27a presumes that the ball hits the striking surface 9a when the posture of the racket 3 when the ball hits the racket 3 is included in the predetermined first range, and when the ball hits the racket 3. When the posture of the racket 3 is included in the second range different from the first range, it is presumed that the ball hits the back surface of the striking surface 9a (another striking surface 9a).

In this case, for example, unlike the case where a plate-shaped pointing device is provided, the number of angular velocity sensors 13 does not increase even if the number of striking surfaces 9a to be analyzed increases. Therefore, the probability that the weight of the racket with the sensor deviates from the weight of the racket 3 is reduced. In addition, posture estimation technology that has been put into practical use in other technical fields such as robots and drones can be applied.

Further, in the present embodiment, the angular velocity sensor 13 can detect the angular velocity around the z-axis. The CPU 27a estimates the time when the ball hits the racket 3 based on the angular velocity around the z-axis.

In this case, the configuration of the analysis system 1 is simplified as compared with a mode in which the presence / absence and a time point of impact are estimated by, for example, shooting with a video camera (the mode is also included in the present disclosure). Further, for example, since the data that is the basis of the estimation at the time of impact and the data that is the basis of the estimation of the impact position can be made the same, there is an inconvenience that the detection results are inconsistent or deviated between different sensors. The probability of occurrence is reduced.

Further, in the present embodiment, the analysis system 1 further includes an acceleration sensor 15 capable of detecting acceleration in the x-axis direction orthogonal to the striking surface 9a. The CPU 27a estimates the time when the ball hits based on the detected value of the acceleration in the direction orthogonal to the striking surface 9a.

In this case, for example, the influence of the impact position is reduced as compared with the method of estimating the time when the ball hits the racket 3 based on the above-mentioned angular velocity around the z-axis. That is, the estimation accuracy at the time of impact can be improved.

Further, in the present embodiment, the analysis system 1 further includes an acceleration sensor 15 capable of detecting acceleration in the z-axis direction parallel to the virtual line CL. The CPU 27a estimates the time when the sphere hits based on the detected value of the acceleration in the z-axis direction.

In this case, for example, as compared with the method of estimating the time when the ball hits the racket 3 based on the acceleration in the x-axis direction described above, the normal line (x) of the hitting surface 9a in the direction in which the ball hits the hitting surface 9a. The effect of the tilt angle on the axis) is reduced.

Further, in the present embodiment, the acceleration sensor capable of detecting the acceleration in the x-axis direction orthogonal to the striking surface 9a is further provided. The CPU 27a corrects the detection value of the angular velocity so that the larger the detection value of the acceleration in the direction in which the hit surface 9a faces when the ball hits the hitting surface 9a, the smaller the absolute value of the angular velocity, and after the correction. The position where the ball hits the striking surface 9a is estimated using the angular velocity of.

In this case, for example, the influence of the speed of the racket 3 on the angular velocity at the time of impact can be reduced. As a result, the accuracy of estimating the impact position based on the angular velocity can be improved.

The angular velocity sensor 13 (sensor device 5) may be located at the end of the grip 11 opposite to the striking portion 9.

In this case, for example, as compared with the embodiment in which the angular velocity sensor 13 is located at the striking portion 9 (the embodiment is also included in the technique according to the present disclosure), the angular velocity sensor 13 is used as another member (for example, in the swing). , Ball, table or floor) is reduced.

The technique according to the present disclosure is not limited to the above embodiments, and may be implemented in various embodiments.

Depending on the performance required for the racket analysis system, the various elements or functions mentioned in the embodiment may be omitted. For example, as can be understood from the description of the embodiment, it is possible to specify the impact position based only on the angular velocity sensor, and if high accuracy is not required for the impact position, the acceleration sensor may be omitted. Good. Further, for example, when the posture of the racket is not estimated, it is not necessary to detect the angular velocity around the x-axis. That is, the angular velocity sensor may be a biaxial angular velocity sensor. Further, if the impact position is specified only in either the y-axis direction or the z-axis direction, the angular velocity sensor may only detect the angular velocity around one axis.

In the embodiment, it has been described that the analysis device 7 analyzes the impact position and the like. However, the sensor device 5 may perform analysis such as estimation of the impact position. That is, the step described as being performed by the CPU 27a of the analysis device 7 may be performed by the CPU 19a of the sensor device 5. Further, the analysis process may be shared between the CPU 27a and the CPU 19a. In other words, in the racket analysis system, the processor that performs analysis such as estimation of the impact position may be CPU 19a instead of CPU 27a, or may be a combination of both.

The technique according to the present disclosure may be used not only for analysis of competition practice, etc., but also for determining the quality of play in an arcade game machine that imitates competition.

1 ... analysis system (racket analysis system), 3 ... racket, 9 ... striking part, 9a ... striking surface, 11 ... grip, 13 ... angular velocity sensor, 27a ... CPU (processor), CL ... virtual line.

Claims (15)

1. The movement of a racket having a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the central portion of the striking surface. It is an analysis system for rackets that analyzes
   The angular velocity sensor fixed to the racket and
   A processing unit that estimates the position of the ball hitting the striking surface based on the detection value of the angular velocity sensor when the ball hits the racket.
   An analysis system for rackets that has.
2. When the z-axis parallel to the virtual line and the y-axis parallel to the striking surface and orthogonal to the z-axis are defined,
   The angular velocity sensor can detect the angular velocity around the z-axis and can detect the angular velocity.
   In the processing unit, the position where the ball hits is the positive side in the y-axis direction with respect to the center side of the striking surface, based on the positive and negative of the angular velocity around the z-axis when the ball hits the racket. The racket analysis system according to claim 1, wherein it estimates which of the two is on the negative side.
3. When the z-axis parallel to the virtual line and the y-axis parallel to the striking surface and orthogonal to the z-axis are defined,
   The angular velocity sensor can detect the angular velocity around the z-axis and can detect the angular velocity.
   The processing unit estimates that the smaller the angular velocity around the z-axis when the ball hits the racket, the more the position where the ball hits is on the center side of the striking surface in the y-axis direction. The racket analysis system according to 1 or 2.
4. Based on the positive and negative values and absolute values of the angular velocity around the z-axis when the ball hits the racket, the processing unit has three positions where the ball hits the striking surface in the y-axis direction. The racket analysis system according to claim 2 or 3, which estimates which of the divided regions is located.
5. When the z-axis parallel to the virtual line and the y-axis parallel to the striking surface and orthogonal to the z-axis are defined,
   The angular velocity sensor can detect the angular velocity around the y-axis and can detect the angular velocity.
   In the processing unit, the longer the amplitude of the angular velocity vibration around the y-axis generated when the sphere hits the racket decays to a predetermined magnitude, the longer the position where the sphere hits is in the z-axis direction. The racket analysis system according to any one of claims 1 to 4, which is presumed to be away from the grip.
6. The fifth aspect of the present invention, wherein the processing unit estimates, based on the time, which of the divided regions in which the hitting surface is hit is located in two or more divided regions in the z-axis direction. Analysis system for rackets.
7. When the z-axis parallel to the virtual line, the x-axis orthogonal to the striking surface, and the x-axis and the y-axis orthogonal to the z-axis are defined,
   The angular velocity sensor can detect the angular velocity around the x-axis, the angular velocity around the y-axis, and the angular velocity around the z-axis.
   The processing unit estimates the posture of the striking portion in the absolute coordinate system based on the angular velocity around the x-axis, the angular velocity around the y-axis, and the angular velocity around the z-axis, and the ball hits the racket. When the posture of the racket is included in the predetermined first range, it is estimated that the ball hits the striking surface, and the posture of the racket when the ball hits the racket is in the first range. The racket analysis system according to any one of claims 1 to 6, wherein it is estimated that the ball hits the back surface of the striking surface when it is included in the second range different from the above.
8. When the z-axis parallel to the virtual line is defined,
   The angular velocity sensor can detect the angular velocity around the z-axis and can detect the angular velocity.
   The racket analysis system according to any one of claims 1 to 7, wherein the processing unit estimates the time when the ball hits the racket based on the angular velocity around the z-axis.
9. It further has an acceleration sensor capable of detecting acceleration in the z-axis direction parallel to the virtual line.
   The racket analysis system according to any one of claims 1 to 8, wherein the processing unit estimates the time when the sphere hits based on the detected value of the acceleration in the z-axis direction.
10. It further has an acceleration sensor capable of detecting acceleration in the x-axis direction orthogonal to the striking surface.
    The racket analysis system according to any one of claims 1 to 9, wherein the processing unit estimates the time when the sphere hits based on the detected value of the acceleration in the x-axis direction.
11. It further has an acceleration sensor capable of detecting acceleration in the x-axis direction orthogonal to the striking surface.
    The processing unit determines that the larger the detected value of the acceleration in the x-axis direction and the side facing the striking surface when the ball hits the striking surface, the smaller the absolute value of the angular velocity. The racket analysis system according to any one of claims 1 to 10, wherein the detected value is corrected and the position where the ball hits the striking surface is estimated using the corrected angular velocity.
12. The racket analysis system according to any one of claims 1 to 11, wherein the angular velocity sensor is located at an end portion of the grip opposite to the striking portion.
13. The movement of a racket having a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. It is an analysis device for rackets to analyze,
    A racket analysis device having a processing unit that estimates the position of the ball hitting the striking surface based on the angular velocity of the racket when the ball hits the racket.
14. The movement of a racket having a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. A racket analysis program for analysis, which can be applied to a computer.
    An analysis program for a racket that executes an estimation step of estimating a position where the ball hits the striking surface based on the angular velocity of the racket when the ball hits the racket.
15. The movement of a racket having a striking portion having a striking surface and a grip extending outside the striking surface along a virtual line parallel to the striking surface and passing through the center of the striking surface. It is an analysis method for rackets to be analyzed.
    An analysis method for a racket, which comprises an estimation step of estimating a position of the hitting surface where the ball hits, based on the angular velocity of the racket when the ball hits the racket.

Images