JPWO2016185894A1 - Golf equipment fitting system, golf equipment fitting method, golf equipment fitting program, golf swing classification method, golf shaft fitting system, golf shaft fitting method, and golf shaft fitting program - Google Patents

Abstract

The golf equipment fitting system (2) includes a swing data acquisition unit (21) for acquiring swing data from a sensor (11) attached to the golf club (1), and swing data acquired by the swing data acquisition unit (21). The swing data storage unit (22) for storing the swing data, the data classification unit (23) for classifying the swing data stored in the swing data storage unit (22), and the swing data classified by the data classification unit (23) And a data predicting unit (24) that predicts swing data of a spec that is not measured by the sensor (11).

Description

The present invention relates to a golf equipment fitting system, a golf equipment fitting method, a golf equipment fitting program, a golf swing classification method, a golf shaft fitting system, a golf shaft fitting method, and a golf shaft fitting program.
This application claims priority on May 20, 2015 based on Japanese Patent Application No. 2015-102548 for which it applied to Japan, and uses the content here.

  In 2008, regulations on the rebound of the head were enforced by the golf rules, and it became difficult to pursue the flight distance with the head alone. In response to this, each manufacturer focused on the shaft and tried to increase the flight distance by utilizing the bending of the shaft. For this reason, shaft variations increased as of 2012, and club variations further increased by combination with the head. As an adverse effect, when a player purchases a club, it is difficult to select the best tool for the player, especially the best shaft.

  A solution to this problem is the fitting technique. As a fitting technique focusing on the entire golf club, a technique described in Patent Document 1 can be given. In this technique, the head behavior at the moment of impact is photographed with a high-speed camera, and the head posture is quantified by replacing it with three-dimensional coordinates by a DLT (Direct Linear Transformation) method. Thereby, the head posture at the time of impact can be specified for each club, and an appropriate club can be selected. However, in this method, it is necessary for the player to actually hit the golf ball with many clubs, and it is difficult to make the best selection even though it can be better.

  As a technique that can solve this problem, there is a technique described in Patent Document 2. In this technique, first, a swing is measured, and a head behavior is simulated based on the swing data. Using this technique, a player can obtain results equivalent to hundreds of test hits by performing only one swing measurement. However, in practice, when a club having a different hardness or weight is used, the player hits the ball by changing the swing itself. For this reason, there is a problem that even if a club or shaft with extremely different performance is simulated for one swing data, the simulation result does not match the actual solution.

  As a technique that can solve this problem, a technique described in Patent Document 3 can be cited. This technique is a technique for calculating a swing that changes for each shaft performance (flex, tone, and torque) using a response surface method. By simulating based on the swing after the change, it is possible to simulate more realistic phenomena.

  However, the technique described in Patent Document 3 has a problem that it is difficult to simulate by changing the “weight” and “length” of the members constituting the golf club, for example. The reason for this is that when the weight and length are changed, even the same player, from the address (the golf club is attached to the ground) to the top (the golf club is raised), the impact from the top (the head is the ball) This is because a swing time such as the time of hitting is greatly different and sufficient calculation accuracy cannot be obtained. Therefore, in the technique described in Patent Document 3, the specs of the golf club that can be changed are limited to specs excluding those that greatly affect the swing time such as weight and length.

  As a technique for solving this, there is a technique described in Patent Document 4. This technique is capable of accurately simulating specifications that greatly affect the swing time by using a swing response curved surface and a time response curved surface.

JP 2005-31734 A Japanese Patent No. 4871218 JP 2011-425 A International Publication No. 2014/132858

  However, in the technique described in Patent Document 4, the player needs to swing a plurality of golf clubs. For this reason, when the player who makes a trial is a beginner whose swing is not stable, there is a possibility that a desired result cannot be obtained because the variation between the swings becomes large.

  The present invention has been made in view of such circumstances, and a golf equipment fitting system, a golf equipment fitting method, a golf equipment fitting program, a golf shaft fitting system, and a golf equipment fitting system capable of selecting an appropriate golf equipment with a small number of trial hits. A golf shaft fitting method and a golf shaft fitting program are provided.

The present invention has the following aspects.
[1] A swing data acquisition unit that acquires swing data from sensors attached to a plurality of golf clubs having different specifications, a swing data storage unit that stores swing data acquired by the swing data acquisition unit, and the swing data A data classifying unit for classifying swing data stored in the storage unit, and a data predicting unit for predicting swing data of a specification not measured by the sensor by referring to the swing data classified by the data classifying unit; , Having a golf equipment fitting system.

[2] A swing data acquisition procedure for acquiring swing data from sensors attached to a plurality of golf clubs having different specifications, and a swing data storage procedure for storing the swing data acquired in the swing data acquisition procedure in a swing data storage unit The swing data stored in the swing data storage unit is classified, and the swing data classified by the data classification procedure is referred to, so that swing data of a spec not measured by the sensor is predicted. And a golf equipment fitting method comprising: a data prediction procedure.

[3] A swing data acquisition procedure for acquiring swing data from sensors attached to a plurality of golf clubs having different specifications, and a swing data storage procedure for storing the swing data acquired in the swing data acquisition procedure in a swing data storage unit The swing data stored in the swing data storage unit is classified, and the swing data classified by the data classification procedure is referred to, so that swing data of a spec not measured by the sensor is predicted. A golf equipment fitting program for causing a computer to execute a data prediction procedure.

  According to the present invention, an appropriate golf equipment can be selected with a small number of trial hits. Therefore, it can be suitably used even for beginners who do not have a stable swing. In addition, since the measurement time is shortened, it is possible to spend time for hearing to golfers and trial hitting of equipment desired for purchase, which is necessary for actual use.

It is a block diagram which shows the golf equipment fitting system of 1st Embodiment. It is a figure which shows an example of the bending rigidity (flex) of a shaft. It is a figure which shows an example of the tone (bending rigidity distribution L (x)) of a shaft. It is a figure which shows an example of the torsional rigidity (torque) of a shaft, and an example of the specification of a head. It is a figure which shows an example of head centroid depth. It is a figure which shows an example of head center-of-gravity height and head center-of-gravity distance. It is a figure which shows the specification of nine golf clubs used for the trial hit for obtaining swing data. It is a figure which shows an example of swing data (acceleration). It is a figure which shows an example of swing data (angular velocity). It is a figure which shows the classification result of swing data. It is a flowchart which shows operation | movement of a golf equipment fitting system. It is a figure for demonstrating the prediction method of swing data. It is a figure which shows the swing data in case the shaft weight is the same. It is a figure which shows the swing data in case a shaft weight differs. It is a figure which shows the swing data calculated only using the swing response curved surface. It is a figure which shows the swing data calculated using the swing response curved surface and the time response curved surface. It is a figure which shows a simulation result. It is a figure which shows an example of an influence degree. It is a figure which shows the table used by the data classification part in 2nd Embodiment. It is a figure which shows the table used by the data estimation part in 2nd Embodiment.

  Hereinafter, a golf equipment fitting system, a golf equipment fitting method, a golf equipment fitting program, a golf swing classification method, a golf shaft fitting system, a golf shaft fitting method, and a golf shaft fitting program according to an embodiment of the present invention will be described with reference to the drawings. I will explain. Here, although a golf shaft is illustrated and explained as a specification of a golf equipment, it is not limited to this.

(First embodiment)
FIG. 1 is a block diagram illustrating a golf equipment fitting system according to a first embodiment. In this figure, the golf club 1 is mounted with a golf shaft whose specifications are known in advance. The golf club 1 includes a sensor 11 and a transmission unit 12 inside the shaft of the grip portion. The transmission unit 12 transmits the output of the sensor 11 to the outside by wireless communication.

  The sensor 11 may be attached to the outside of the shaft. For example, the sensor 11 may be attached to the outside of the shaft at the lower part of the grip. Thus, the sensor 11 can be attached to the golf club 1 owned by the player. The sensor 11 may be attached to the inside of the shaft at the grip end of the golf club 1. Thereby, more accurate swing data can be measured. The sensor 11 and the transmission unit 12 are preferably configured integrally. In the case where the sensor 11 and the transmission unit 12 are configured in a non-integral manner, it is preferable that the transmission unit 12 and the battery incorporated in the transmission unit are attached to a golfer's arm or the like. With this configuration, the weight of the sensor main body attached to the club can be reduced, and the change in club specifications due to the attachment of the sensor 11 can be prevented.

  The sensor 11 is a 6-axis sensor that detects and outputs 3-axis acceleration and 3-axis angular velocity, but is not limited thereto. For example, the sensor 11 may be a nine-axis sensor that measures the three-axis direction using geomagnetism in addition to the three-axis acceleration and the three-axis angular velocity. The sensor 11 may be composed of two sensors, a 6-axis sensor and a 3-axis geomagnetic sensor. Further, two acceleration sensors having different measurement ranges and one angular velocity sensor may be used. In this case, a slow operation such as a backswing uses a sensor with a narrow measurement range, and a quick operation such as a downswing uses a sensor with a wide measurement range. Thereby, the measurement accuracy can be improved. The sensor frequency used here is 200 Hz, but is preferably 500 Hz or more, more preferably 1000 Hz or more. The higher the frequency, the more detailed changes in operation can be captured.

  The golf equipment fitting system 2 is a computer device that includes a processor such as a CPU (Central Processing Unit) and a memory that stores a program executed by the processor. The golf equipment fitting system 2 includes a reception unit 20, a swing data acquisition unit 21, a swing data storage unit 22, a data classification unit 23, a data prediction unit 24, a swing response curved surface calculation unit 25, a time response curved surface calculation unit 26, and a simulation execution unit. 27 and a result output unit 28. These functions are realized by the CPU executing a program stored in the memory.

  The receiving unit 20 receives sensor output data (swing data) transmitted by the transmitting unit 12. The swing data acquisition unit 21 acquires swing data via the reception unit 20. The swing data storage unit 22 stores in advance a plurality of swing data obtained by many players swinging a plurality of golf clubs. The data stored in the swing data storage unit 22 is not limited to swing data. For example, head behavior data calculated based on swing data may be stored in the swing data storage unit 22.

  The data classification unit 23 classifies the swing data stored in the swing data storage unit 22. Although details will be described later, the data classification unit 23 classifies the swing data using the self-organizing map. The classification method is not limited to the self-organizing map. For example, the data classification unit 23 uses various clustering methods such as neural network, support vector machine, Bayesian network, hidden Markov model, k-means method, cluster classification, principal component analysis, machine learning, and multivariate analysis. Also good. Golf experts may also classify based on their experience. Further, the data classifying unit 23 may classify the head behavior calculated based on the swing data without being limited to the swing data.

  The data prediction unit 24 refers to the swing data storage unit 22 and extracts swing data similar to the swing data acquired by the swing data acquisition unit 21. Then, the data predicting unit 24 predicts the swing data of the golf club that has not been swung based on the player data corresponding to the extracted swing data. The data prediction unit 24 may predict the head behavior calculated based on the swing data, not limited to the swing data.

  The swing response surface calculation unit 25 calculates a swing response surface when the golf club 1 is swung based on the swing data acquired by the swing data acquisition unit 21. The time response curved surface calculation unit 26 calculates a response surface of the swing time when the golf club 1 is swung based on the swing data acquired by the swing data acquisition unit 21. The simulation execution unit 27 performs a simulation by FEM (Finite Element Method) using the swing response curved surface and the time response curved surface. The result output unit 28 outputs the result of the simulation executed by the simulation execution unit 27.

  Next, the golf club 1 used for obtaining swing data will be described. The swing data obtained by swinging the golf club 1 is used to construct the swing data storage unit 22. The player swings in accordance with the specifications of the golf club 1 (shaft weight, bending rigidity, torsional rigidity, etc.). If the specifications of the golf club 1 are different, the player's swing is different. For this reason, even if a simulation is performed based on swing data obtained using one golf club, it is difficult to obtain a reasonable simulation result.

  Therefore, an example of a method for selecting the golf club 1 used for the simulation will be described. First, based on the experimental design method, three different specifications are selected from the specifications of many golf clubs 1. A total of 27 clubs with 3 levels for each spec are prepared. Among them, nine golf clubs 1 are selected based on the L9 orthogonal table. If three specifications and three levels of data are to be obtained without using the experimental design method, it is necessary to perform simulation for 33 = 27 golf clubs 1. Therefore, the load concerning swing data acquisition can be reduced by using the experimental design method.

  Here, as an example of the three specifications of the golf club 1, the weight of the shaft, the bending rigidity (hereinafter referred to as flex) of the shaft, and the bending rigidity distribution (hereinafter referred to as tone) of the shaft are used. The specs of the golf club 1 used exemplify those in which 3 specs are different at 3 levels, but are not limited thereto.

  FIG. 2A is a diagram illustrating an example of the bending rigidity (flex) of the shaft. In FIG. 2A, a position of 920 mm from the end on the small diameter side of the shaft is supported from the lower side, and a position in the direction of 150 mm on the large diameter side (1070 mm from the end on the small diameter side) is further supported from above. A load of 3.0 kgf is applied at a position 10 mm to 10 mm. The amount of displacement of the narrow-diameter end at this time is the flex (bending rigidity) of the shaft.

FIG. 2B is a diagram illustrating an example of shaft tone (bending stiffness distribution L (x)). The tone of the shaft is defined by the coefficient C of the function P indicating the point at which the bending of the shaft becomes the largest. In FIG. 2B, EI (i) indicates bending rigidity, and x indicates a position on the shaft with respect to the tip of the shaft. The condition of the shaft is expressed by the following formula.
L (x) = C0 * P0 (x) + C1 * P1 (x) + C2 * P2 (x) + C3 * P3 (x)

Also, the tone may be defined as follows. In a state where the shaft is curved, a point that protrudes most in the circumferential direction of the shaft due to the curve is set as a vertex, and the vertex, the tip end portion, and the distance Lk are measured. The ratio of the distance Lk to the shaft length Lb (the distance obtained by connecting both ends of the shaft during bending with a straight line) when the above-described bending is performed is defined as a kick point value. That is, the kick point is obtained using the following equation.
Kick point (%) = (Lk / Lb) × 100

  In this specification, a value measured using a shaft kick point gauge “FG-105RM” manufactured by Fourteen is used. For example, a shaft having a kick point of less than 44.0% is a low kick point shaft (first tone), a shaft having a kick point of 44.0% or more and less than 45.0% is a middle kick point shaft (medium tone), and a kick point is 45.0% or more. Can be classified as a high kick point shaft (original tone).

Lk and Lb are strictly defined as follows.
Lk: A straight line connecting both ends of the shaft when the ends of the shaft are curved by applying a compressive load so that the linear distance between the ends of the shaft is 98.5 to 99.5% of the shaft length. The distance between the intersection when the perpendicular is drawn from the apex of the curve and the tip end of the shaft.
Lb: between the ends of the shaft when the ends of the shaft are curved by applying a compressive load so that the linear distance between the ends of the shaft is 98.5 to 99.5% of the shaft length. direct distance.

  The shaft tone is classified into an original tone closer to the grip, a tip tone closer to the head, and a middle tone in the middle where the shaft has the largest bending. The tone is assigned to one of the original tone, the first tone, and the middle tone depending on the value defined by the coefficient C of the function P. Swing data is acquired using the nine golf clubs 1 described above.

  Other specifications include shaft torsional rigidity (hereinafter referred to as torque), shaft torsional rigidity distribution, shaft weight distribution, golf club length, head weight, club balance, head center of gravity depth, head center of gravity height, head center of gravity. It is possible to apply distance, grip weight, loft angle, lie angle and face angle.

  FIG. 3A is a diagram illustrating an example of torsional rigidity (torque) of the shaft. As shown in FIG. 3A, the torque of the shaft is defined by the twist angle of the shaft. In FIG. 3A, a position of 1035 mm is fixed from the end portion on the small diameter side of the shaft, and a torsional load is applied to a position of 45 mm from the tip of the shaft. Further, a torsional load of 1.152 kgf is applied at a position 120 mm away from the shaft axis. This is synonymous with a torsional load of 1 ft · lb being applied to the shaft tip. At this time, the twist angle of the end portion on the small diameter side of the shaft is defined as the torque of the shaft.

  FIG. 3B is a diagram illustrating an example of the head center-of-gravity depth. As shown in FIG. 3B, the head center-of-gravity depth is a depth (distance) from the face surface to the center of gravity of the head. FIG. 3C is a diagram illustrating an example of the head center-of-gravity height and the head center-of-gravity distance. The head center-of-gravity height is the length from the leading edge to the center of gravity on the face surface. The head center-of-gravity distance is the length of a perpendicular extending from the shaft axis toward the center of gravity on the face surface. Also, the torsional rigidity distribution of the shaft and the weight distribution of the shaft can be expressed in the same manner as the relationship between the bending rigidity distribution and the tone.

  The club balance (also referred to as swing weight) is how the head of the golf club works. The goodness of the head means the weight of the head during a swing or waggle. The club balance is measured using a club balance meter “Golf Scale Scale” manufactured by Kenneth Smith.

  The club hardness is a frequency of the golf club 1 in which a grip and a head are attached to a shaft. This frequency is measured using “Golf Club Timing Harmonizer” manufactured by Fujikura Rubber Industries. For example, the club head is vibrated in a state where 180 mm is fixed from the grip end. At this time, the vibration frequency per minute at a position of 760 mm from the grip end is defined as the club hardness.

  FIG. 4 is a diagram showing specifications of nine golf clubs used for trial hits for obtaining swing data. The specifications of the nine golf clubs are determined based on the experimental design L9 type orthogonal table. The shaft weight indicates a normalized value, and a smaller value means that the weight is heavier. In this embodiment, the shaft weight is 80 g when “0”, the shaft weight is 70 g when “0.5”, and the shaft weight is 60 g when “1”.

  Flex indicates a normalized value, and a smaller value means higher rigidity. In this embodiment, the flex is 130 mm for “0”, the flex is 180 mm for “0.5”, and the flex is 220 mm for “1”.

  Moreover, the normalized value is also shown regarding the tone. “0” is the first tone, “0.5” is the middle tone, and “1” is the original tone. The number of clubs used is not limited to nine as long as it is a quantity that can be practically tested.

  The same head is attached to the nine golf clubs 1 shown in FIG. The nine golf clubs 1 have the same shaft length, and the nine golf clubs 1 have the same weight. Thus, by using the golf club 1 having the same specifications other than the specifications to be fitted, changes due to other factors can be eliminated.

  Further, the sensor 11, the transmission unit 12, and devices necessary for operating these are inserted into the shafts of the nine golf clubs 1. The weight of the sensor 11, the transmission part 12, and the whole apparatus required in order to operate these is restrained to 30g. Thereby, the influence on the swing by the weight increase of the golf club 1 is suppressed. More preferably, it is 20 g or less. If it is 30 g or less, even if the device is attached to the club, a general lightweight grip can be used without changing the total weight or balance of the club.

  Next, a method for classifying swing data will be described. In the present embodiment, 126 players swing 9 golf clubs 1 each. As a result, a lot of swing data can be obtained. Specifically, since a 6-axis sensor is used as the sensor 11, 126 people × 9 × 6 axes = 6804 data is stored in the swing data storage unit 22.

  FIG. 5A is a diagram illustrating an example of swing data (acceleration). In FIG. 5A, swing data (acceleration) of five players is shown. The solid line indicates acceleration in the x-axis direction, the broken line indicates acceleration in the y-axis direction, and the alternate long and short dash line indicates acceleration in the z-axis direction. The vertical axis is the acceleration value normalized so that the absolute value is 1. The horizontal axis represents the time from the top (in a state where the golf club is swung up) to the impact (in the state where the head hits the ball). The time from top to impact is shown divided into 40.

  FIG. 5B is a diagram illustrating an example of swing data (angular velocity). In FIG. 5B, swing data (angular velocity) of five players is shown. The solid line shows the angular velocity around the x axis, the broken line shows the angular velocity around the y axis, and the alternate long and short dash line shows the angular velocity around the z axis. The vertical axis represents the angular velocity value normalized so that the absolute value is 1. The horizontal axis is the time from top to impact. The time from top to impact is shown divided into 40.

  In the present embodiment, swing data of 126 players is acquired as described above. If such complex swing data exists for 126 people, it is difficult to classify these swing data with human eyes. Therefore, in this embodiment, the data classification unit 23 classifies these swing data using a self-organizing map.

  A self-organizing map (SOM) is an algorithm for performing clustering without a teacher, and is a data analysis method for nonlinearly mapping high-dimensional data onto a two-dimensional plane. By mapping high-dimensional and complex data onto a two-dimensional plane using a self-organizing map, clustering of swing data can be easily performed.

  The self-organizing map has an input layer and an output layer. The input layer in the present embodiment is swing data. In the present embodiment, the data classification unit 23 uses the swing data of the golf club 1 with the club number 5 shown in FIG. 4 among the swing data of the player. Each swing data is data obtained by dividing the time from the top to the impact by 40 as described above. Each swing data is data that is standardized so that the absolute value is 1.

  An input vector s in the following equation (1) represents swing data for one person. Since the time from the top to the impact is divided into 40, D = 40.

  The swing data S of 126 players can be expressed by the following formula (2). Note that N = 126.

  Next, the output layer will be described. The reference vector m is defined by the following equation (3) using the neuron u. Note that i is a positive integer.

  The matrix M can be expressed by the following equation (4). L is the number of nodes of the matrix M.

  The data classification unit 23 assigns a random value to each neuron u. Thereafter, the data classification unit 23 gives the input vector s to the input layer. The data classification unit 23 determines an output layer node having a reference vector m closest to the input vector s given to the input layer. The node determined here is the winner node c. The data classification unit 23 calculates the winner node c based on the Euclidean distance between the input vector s and the reference vector m, as shown in the following equation (5).

  When the winner node c is determined, the data classification unit 23 learns the winner node c and nodes around the winner node c. Specifically, the data classifying unit 23 brings the winner node c and the surrounding nodes in the reference vector m closer to the input vector s according to the following equation (6).

  Here, the data classification unit 23 calculates a function hci that determines the magnitude of learning according to the following equation (7). The coefficient α is a positive constant less than 1 that represents the strength of learning. In this embodiment, α = 0.5. ri is a position on the output layer. rc is the coordinate of the winner node c. σ is a positive constant and corresponds to the standard deviation of a normal distribution.

  The data classification unit 23 classified the swing data of 126 players into six accelerations x, y, z and angular velocities x, y, z separately by repeating the above learning a predetermined number of times. In this embodiment, the data classification unit 23 has repeated learning 10,000 times. Six data may be connected in series to form one data.

  FIG. 6 is a diagram showing a classification result of swing data of one axis (acceleration x-axis) as an example. As shown in FIG. 6, the data classification unit 23 mathematically classifies the swing data in the swing data storage unit 22 using the self-organizing map. In FIG. 6, the swing data is classified as one of hexagonal cells, but is not limited to this. For example, a regular polygon (for example, square) square may be used. Moreover, although online SOM was used here, batch SOM may be used. Actually, since six data of acceleration x, y, z and angular velocity x, y, z are classified separately, six maps are generated.

  FIG. 7 is a flowchart showing the operation of the golf equipment fitting system. First, the fitting target player swings the golf club 1 having the club number “5” out of the nine golf clubs 1. The sensor 11 outputs a detection result (swing data) during the swing operation to the transmission unit 12. When the swing data is output from the sensor 11, the transmission unit 12 wirelessly transmits the swing data to the reception unit 20. The receiving unit 20 outputs the received swing data to the swing data acquiring unit 21. The swing data acquisition unit 21 acquires the swing data output from the reception unit 20 (step S1).

  Next, the data prediction unit 24 refers to the swing data storage unit 22 classified in advance by the data classification unit 23, and the other eight golf clubs 1 (club numbers “1” to “4” and “6”). ”To“ 9 ”) is predicted (step S2).

  The swing data prediction method executed in step S2 will be described below. FIG. 8 is a diagram for explaining a swing data prediction method. The swing data SD1 to SD9 are swing data of club numbers “1” to “9”, respectively. The difference data DD1 to DD4 and DD6 to DD9 are data indicating differences between the swing data SD1 to SD4 and SD6 to SD9, respectively, and the swing data SD5. The difference data DD1 to DD4 and DD6 to DD9 are stored in the swing data storage unit 22 for each of a plurality of players.

  The data prediction unit 24 predicts swing data of the other eight golf clubs 1 only from the classification of the golf club 1 with the club number “5”. This is based on the hypothesis that players with similar swings have similar adjustments when swinging on another shaft. Therefore, as shown in FIG. 8, the data prediction unit 24 subtracts the data of the club number “5” from the eight data of the club numbers “1 to 4, 6 to 9” of each cluster. Data DD1 to DD4 and DD6 to DD9 are calculated.

  When the new data ND of the golf club 1 with the club number “5” is acquired by the swing data acquisition unit 21, the data prediction unit 24 determines where the acquired new data ND is classified in the self-organizing map. Determine. Specifically, the data prediction unit 24 searches all the swing data in the swing data storage unit 22 and extracts the swing data classification closest to the new data ND. As a result, the swing data of a player whose swing is similar to the player who made the trial can be specified. Note that the data classification unit 23 may perform classification again by entering new data ND.

  For example, when predicting the swing data whose club number is “1”, the data prediction unit 24 calculates the average value of all the difference data DD1 in the extracted classification. Then, the data prediction unit 24 calculates the swing data with the club number “1” by adding the calculated average value of the difference data DD1 to the new data ND. The data prediction unit 24 similarly calculates swing data for swing data of other golf clubs 1 (club numbers “2” to “4” and “6” to “9”). Here, the average value of DD1 of the classified group is added to ND, but DD1 of the player with the closest swing data may be added to ND. In addition, although six data of the six-axis sensor are handled separately here, the six data may be connected in series and handled as one data.

  With the above processing, the data prediction unit 24 holds time-series swing data that is considered to have been swung once by each of the nine golf clubs 1. The data prediction unit 24 obtains one piece of swing data (club number “5”) and eight pieces of predicted swing data (club numbers “1” to “4” and “6” to “9”). Are converted into movement speed data of the grip portion of the golf club 1 and shaft rotation data of the shaft. This conversion is performed by geometric conversion based on the mounting position of the sensor 11 inserted into the golf club 1 and the predetermined two-point positional relationship of the grip portion of the golf club 1. .

  The data prediction unit 24 converts only the swing data from the top (the state where the golf club is raised) to the impact (the state where the head hits the ball) into the movement speed data and the shaft rotation data. Not limited to. For example, the data prediction unit 24 may convert swing data from an address (a state where the golf club is attached to the ground) to an impact (a state where the head hits the ball) into movement speed data and shaft rotation data. Thus, by performing data prediction using data obtained by swinging a plurality of golf clubs having different specifications as a database, it is possible to clarify the swing change amount for each specification. Moreover, the predicted value regarding two or more different specification data (for example, hardness, tone, weight, etc.) can be obtained at a time. Therefore, fitting can be performed more simply, and the interaction between specifications is also taken into consideration.

  Next, the swing response surface calculation unit 25 reads the swing data of the nine clubs held in the data prediction unit 24, and calculates a swing response surface obtained by converting the skill and skill of the test batter into a linear function ( Step S3). The swing response surface is the relationship between the moving speed data and the shaft rotation data obtained when the golf club 1 is tested with nine types of golf clubs shown in FIG. 4 and the three specifications of the golf club 1 (shaft weight, flex, and tone). It is a formula. Although a linear function is used here, a quadratic function, a high-order function, and various functions can be used.

  Hereinafter, the swing response curved surface calculation process executed in step S3 will be described. Swing data measured by trial hitting a plurality of golf clubs 1 is expressed by f1 to f9 as shown in Expression (8). Time t is discretized into t = {t1,..., Tn}. In the present embodiment, since nine golf clubs 1 have been trial hits, the swing data is from f1 to f9, but the value varies depending on the number of trial hits of the golf club 1. Further, fj (ti) is a measurement amount by the j-th golf club 1. Specifically, fj (ti) indicates the respective amounts of the three-direction acceleration {ax, ay, az} and the three-direction angular velocity {ωx, ωy, ωz}.

  When the three specifications (design variables) of the nine golf clubs are {xj, yj, zj} (j = 1... 9), the relationship of the formula (8) is obtained. The swing response curved surface calculation unit 25 solves Equation (8) for each ti.

  Here, x, y, and z are design variables, respectively. x is the first spec (shaft weight), y is the second spec (flex), and z is the third spec (tone). The numbers 1 to n in x1 to xn, y1 to yn, and z1 to nz correspond to the numbers of the clubs. The x bar, y bar, and z bar are given arbitrary values convenient for each analysis. For example, intermediate values of design variables may be set for x bar, y bar, and z bar, respectively.

  The swing response curved surface calculation unit 25 calculates coefficients a1 to a4 of the response curved surface as shown in Equation (9) by solving Equation (8). When n = 5 or more, since the equation (8) is in an overcondition, there is no exact solution. Therefore, the swing response surface calculation unit 25 uses a generalized inverse matrix A + (also referred to as Moore-Penrose inverse matrix or pseudo inverse matrix). This method is a method for calculating an approximate solution when there is no exact solution. That is, this method is a method for calculating a solution that minimizes the error | Ax−b | 2. Since this method is a general mathematical method, detailed description is omitted. In order to solve Equation (8), numerical calculation software “MATLAB” manufactured by MathWorks was used.

  Coefficients a1 to a4 obtained by the equation (9) are values corresponding to the skill of the test hitter and the heel of the swing. The swing response curved surface calculation unit 25 calculates swing data f of specifications that are not actually measured according to the equation (10). In other words, the swing response curved surface calculation unit 25 calculates an approximate value of the three-way acceleration and the three-way angular velocity for an arbitrary specification {x, y, z} according to the equation (10).

  The swing response curved surface calculation unit 25 calculates swing data of the m-th golf club 1 that is not measured according to the equation (11). Fm (t) shown in Expression (11) is a swing response curved surface.

  Next, the time response curved surface calculation unit 26 reads the swing data of the nine clubs held in the data prediction unit 24. And the time response curved surface calculation part 26 calculates the time response curved surface for prescribing | regulating a player's swing time (step S4).

  The time response curved surface calculation process executed in step S4 will be described below. When the swing time changes according to the golf club 1, if the club number of the golf club 1 is k (k = 1,..., 9), the sampling time is normalized as shown in Expression (12). Note that tk is the swing time of each golf club 1.

  If there is no significant difference in swing time, the time response curved surface calculation unit 26 may stop the calculation process at tmin with the shortest swing time among the nine clubs in order to save the time and effort of normalization.

  The time response curved surface is an expression showing the relationship between the respective swing times of the nine types of golf clubs 1 shown in FIG. 4 and the three specifications (shaft weight, flex, and tone) of the golf club. The time response curved surface calculation unit 26 can calculate a time response curved surface according to the equations (13) to (15).

  In formula (13), g1 to g9 are swing times of nine golf clubs 1. The time response curved surface calculation unit 26 calculates the coefficients b1 to b4 according to the equation (14). The coefficients b1 to b4 are values corresponding to the player's swing time. By this processing, the time response curved surface calculation unit 26 can calculate the difference in swing time due to the difference in weight of the golf club 1.

  Next, the time response curved surface calculation unit 26 calculates gm according to the equation (15). Gm shown in Expression (15) is a time response curved surface.

  Further, as shown in Expression (16), swing data fm ′ (t) is calculated. The swing data fm '(t) is swing data newly calculated based on the swing response curved surface and the time response curved surface.

  The simulation execution unit 27 uses the swing response curved surface fm calculated by the swing response curved surface calculation unit 25 and the time response curved surface gm calculated by the time response curved surface calculation unit 26 to swing the golf club 1 that has not been measured. Data fm ′ (t) is calculated. Then, the simulation execution unit 27 simulates the movement of the head of the golf club 1 by the dynamic finite element method based on the calculated swing data fm ′ (t) (step S5).

  The simulation results showing the motion of the golf club 1 obtained by this analysis are the club speed at the time of impact, the face angle, and the impact loft (the loft angle at the time of impact). Club speed affects the flight distance of the ball. The face angle affects the flying ball direction. Impact loft affects ballistic height. The simulation result calculated by the simulation execution unit 27 is not limited to the club speed, the face angle, and the impact loft. For example, as a simulation result, a club pass, an attack angle, a ball speed, a flight distance, and the like may be calculated.

  Since a known method (for example, commercially available finite element method software) is used as an analysis method by the dynamic finite element method, a detailed description of the processing operation is omitted. The simulation execution unit 27 analyzes the motion of the golf club head using a swing response surface obtained by linearizing the player's skill and wrinkle and a time response surface formulated by the player's swing time. Therefore, the simulation execution unit 27 can simulate the motion of the golf club 1 in consideration of the player's skill, habit, and swing time (change in swing time according to the specifications of the golf club 1).

  As described above, the weight and the club length greatly affect the swing time. FIG. 9 is a diagram illustrating swing data when the shaft weight is the same. FIG. 10 is a diagram showing swing data when shaft weights are different. 9 and 10, an angular velocity ωx around the x-axis is shown as an example of the swing data. When the shaft weight is the same, the variation in swing time is small as shown in FIG. On the other hand, when the shaft weight is different from 40 g, 60 g, and 80 g, respectively, the swing time varies greatly as shown in FIG.

  In this case, in order to solve Equation (9) using the generalized inverse matrix, it is necessary to keep t constant. If the swing time does not change significantly as shown in FIG. 9, no major error will be produced no matter how the swing time is made constant. However, when the swing times are greatly different as shown in FIG. 10, a large error is generated if t is forced to be constant. Therefore, the time response surface calculation unit 26 calculates a time response surface after each t is standardized. The simulation executing unit 27 can calculate swing data reflecting an appropriate swing time by converting t again according to the calculated time response curved surface.

  Thus, even when the specifications of the golf club 1 are changed, swing data reflecting an appropriate swing time can be created. Therefore, an accurate simulation result can be obtained even when specifications (weight and length) that greatly affect the swing time are changed. In addition, strictly speaking, the swing time fluctuates even when a specification that has little influence on the swing time is changed. The simulation execution unit 27 calculates swing data using the swing response curved surface and the time response curved surface, so that the calculation accuracy can be improved even when a specification that has little influence on the swing time is changed.

  FIG. 11 is a diagram showing swing data calculated using only the swing response curved surface. FIG. 12 is a diagram showing swing data calculated using a swing response curved surface and a time response curved surface. As is clear from these figures, the simulation execution unit 27 can perform a simulation using the swing response surface and the time response surface, thereby obtaining a more accurate simulation result.

  Next, a result data generation processing operation performed by the result output unit 28 will be described. In the present embodiment, the shaft weight level (first specification division level number i) is 3, the flex level (second specification division level number j) is 5, and the tone level (third specification division level). The number of levels k) was 5. The number of division levels is set in order to divide the 9 specs actually tested into an arbitrary number.

  Specifically, in this embodiment, the shaft weight has three levels (60 g, 70 g, and 80 g), and the flex has five levels (X, S, R, A, and L in order from the hardest). The tone has five levels (0 (first tone), 0.25 (first tone), 0.5 (medium tone), 0.75 (middle tone), and 1 (original tone)).

  In this case, the simulation execution unit 27 can obtain a simulation result of the motion of the golf club 1 for each of 3 × 5 × 5 = 75 types of shafts. The result output unit 28 displays the simulation result, for example, with the club speed as the vertical axis and the result number as the horizontal axis.

  FIG. 13 is a diagram illustrating a simulation result. FIG. 13 shows that the greater the club speed value, the faster the speed (the flying distance increases), and the smaller the club speed value, the slower the speed (the flying distance does not appear). As shown in FIG. 13, in order to maximize the club speed, the golf club 1 having a shaft weight of “60 g”, a flex of “X”, and a tone of “medium tone” is selected. The selection of the golf club 1 is performed by the simulation execution unit 27. Note that the user may check the simulation result and select the golf club 1 having the optimum specifications. In addition to the club speed, the face angle, impact loft, and the like can be expressed by the same method.

  Further, the result output unit 28 may display the trend of specifications in a natural language. That is, the result output unit 28 converts the simulation result by the simulation execution unit 27 into a natural language and displays it according to the relationship between each absolute value and inclination. For example, if the purpose of the fitting is to increase the club speed, the result output unit 28 indicates that “the harder the flex, the higher the club speed”, “the lighter the shaft, the higher the club speed”, “when the tone is the original tone "The club speed increases the most" and "the influence of the shaft weight is the largest" is displayed. If the purpose of the fitting is to increase the impact loft (raise the ball), the result output unit 28 indicates that “the harder the flex is, the higher the ball is”, “the lighter the shaft is, the higher the ball is”, In the case of, the ball goes up the most, “the influence of the flex is the greatest”, etc. are displayed. Thereby, since the user of the golf equipment fitting system 2 can grasp the tendency of specifications in a natural language, the golf equipment can be easily selected.

  In this way, the result output unit 28 outputs the simulation result in association with the purpose of fitting and the specification. Therefore, the user can understand the association with the spec at a glance. In addition, the result output unit 28 outputs all possible specs instead of outputting only one point of the specifications of the golf shaft when the club speed is maximum. Thereby, the user can grasp the specs that can achieve an appropriate flight distance, an appropriate trajectory height, and an appropriate trajectory bend. Therefore, the specifications of the golf club 1 can be selected according to the player's preference. This is particularly effective when the face angle is calculated. If the maximum bending shaft is selected because it is desired to bend the trajectory, the trajectory will bend too much. For this reason, it is preferable for the user that an appropriate trajectory curve can be selected.

  Although the example in which the result output unit 28 displays the character information indicating the specification is shown in FIG. 13, it may be displayed by color-coding for each specification. In order to make it easy to see, the result output unit 28 may devise a plotting method, display a bar graph, or display a three-dimensional graph. In order to make it easier to see, the result output unit 28 may change and display the line color density so that the one containing the best value is conspicuous.

  In the above description, the first spec is described as the shaft weight (3 levels), the second spec as flex (5 levels), and the third spec as the tone (5 levels), but other specifications are used. May be. For example, the first specification may be torque (5 levels), the second specification may be flex (5 levels), and the third specification may be tone (5 levels). Alternatively, the first specification may be flex (5 levels), the second specification may be torque (5 levels), and the third specification may be weight (3 levels).

  In the present embodiment, the shaft specifications are described as an example. However, the simulation execution unit 27 may perform the simulation of the present embodiment by using the club specifications, the head specifications, the grip specifications, and the like.

  Next, an operation for outputting a complex condition such as club speed and ballistic height, club speed and ballistic curve as a fitting result will be described. In general, the player does not seek only one performance for the golf club 1 but seeks complex performance. For example, maximizing club speed to prevent slicing and simultaneously increase trajectory. The ball trajectory is determined by the height of the trajectory and the direction of the trajectory. In the present embodiment, the ballistic height has three levels (High (high trajectory), Mid (medium trajectory), and Low (low trajectory)). The ballistic direction has three levels (Fade (Fade), Straight (Draw)), and Draw (Draw). In this case, based on the player's request, one of the nine trajectories in which the trajectory height and the trajectory direction are combined is selected.

Here, the fade is a trajectory that turns to the right in the case of a right strike. A draw is a trajectory that turns to the left when hitting right. For example, the result output unit 28 outputs a condition that satisfies the selected trajectory while maximizing the club speed. In order to output a complex condition, a specification that maximizes the objective function F is required. The objective function F is expressed by the following equation.
F = α × f1 + β × f2 + γ × f3

  In the above formula, f1 is first result data (for example, club speed), f2 is second result data (for example, face angle), and f3 is third result data (for example, impact loft). ), And α, β, and γ are weighting factors. α, β, and γ are appropriately set according to the player's request. In general, α is preferably 1 to 3 times the value of (β + γ). This is because club speed is most important for players.

  Note that the result output unit 28 may convert and display the specification indicating the maximum value of the simulation result (club speed, impact loft, face angle, etc.) into the product name of the golf equipment. Thereby, the golf equipment fitting system 2 can be practically used.

  In this way, the result output unit 28 can display whether or not it is suitable for realizing the trajectory selected based on the player's request. For this reason, the user of the golf equipment fitting system 2 can easily select a golf equipment that realizes a desired trajectory. In other words, the golf equipment fitting system 2 enables the fitting of the golf equipment. In addition, measurement, analysis, and result display can be performed in a short time, and the user can visually determine the specifications of the optimum golf equipment.

  In the above example, the shaft weight, flex, and tone are selected, but not limited thereto. Further, the number of specifications is not limited to three. In addition, the specification used for the simulation is preferably a specification in which the swing time is easily changed. Others include shaft bending rigidity, shaft torsional rigidity, shaft weight, shaft bending rigidity distribution, shaft torsional rigidity distribution, shaft weight distribution, golf club length, head weight, club balance, head center of gravity depth, head center of gravity. Any specs of height, head center-of-gravity distance, grip weight, loft angle, lie angle, and face angle may be used.

  Further, the time response curved surface calculation unit 26 may calculate the time response curved surface by extending each swing data so as to match the longest swing time so that all the swing times are equal. In this case, the calculation time can be shortened.

  FIG. 14 is a diagram illustrating an example of the degree of influence. The influence degree is a value indicating how much the head behavior is affected when the spec is changed. When a simulation is performed by selecting a plurality of different specifications, it is better to combine the ones having the closest influence as much as possible. The reason for this will be described below.

  For example, when the simulation execution unit 27 performs a simulation based on three specifications including a head weight (influence degree: 5), a shaft torsional rigidity distribution (influence degree: 1), and a head center of gravity height (influence degree: 1). The data is greatly influenced by the spec (head weight) having a large influence, and the result of the spec (shaft torsional rigidity distribution and head center of gravity height) having a small influence is almost unknown. For this reason, it is preferable to select a spec that has a difference in influence level within a predetermined range (in this embodiment, within 2). More preferably, the influence levels of the three different specifications are the same.

  When simulation is performed by selecting three specifications, the combination of shaft weight, flexural rigidity (flex), and flexural rigidity distribution (tone) is the best. By using this combination, optimum shaft fitting can be performed.

  Also, a combination of flexural rigidity (flex), torsional rigidity (torque), and shaft weight may be used. Also, a combination of bending stiffness (flex), torsional stiffness (torque), and bending stiffness distribution (tone) may be used. Further, a combination of flexural rigidity (flex), shaft weight, and shaft weight distribution may be used. By using these combinations, this embodiment can be applied to shaft design.

  Also, a combination of bending stiffness (flex), club length, and head weight may be used. In this case, the club can be suitably fitted.

  A combination of the center of gravity height, the center of gravity depth, and the center of gravity distance of the head may be used. In this case, the fitting of the head can be suitably performed. Further, a combination of the loft angle, lie angle, and face angle of the head may be used. Also in this case, the fitting of the head can be suitably performed.

  A combination of club hardness, club length, total club weight, and club balance may also be used. As mentioned above, although the suitable combination was illustrated, it is not limited to this.

(Second Embodiment)
Next, a second embodiment will be described with reference to FIG. In the first embodiment, the head behavior includes a swing data acquisition unit 21, a swing data storage unit 22, a data classification unit 23, a data prediction unit 24, a swing response surface calculation unit 25, a time response surface calculation unit 26, and a simulation execution. It was decided to be calculated by the unit 27. In addition, the simulation execution unit 27 simply calculates the head behavior including the position coordinates and the posture of the head with the shaft as a rigid body. In the first embodiment, the simulation execution unit 27 requires a high-performance computer. Therefore, it is desirable that the computer that executes the simulation execution unit 27 is installed in a server or the like and provided separately from the swing data acquisition unit 21.

  On the other hand, in the second embodiment, the head behavior is measured without using the simulation execution unit 27. Specifically, the swing data acquisition unit 21 directly measures the head behavior using a camera, a sound wave, or the like. The swing data storage unit 22 stores the head behavior measured by the swing data acquisition unit 21. In the second embodiment, a high-performance computer is not necessary and can be executed by, for example, a smartphone.

  The data classification unit 23 classifies the head behavior stored in the swing data storage unit 22 based on the table shown in FIG. The data classification unit 23 determines to which group the newly measured head behavior data belongs among the preset groups 1 to 45 shown in FIG. In this embodiment, the data classification unit 23 performs classification by paying attention to three items of club speed, club pass, and attack angle.

  For example, when the measured club speed is 33 [m / s], the club path is 1.2 [deg], and the attack angle is −2.0 [deg], the data classification unit 23 determines the measured head behavior. Is determined to belong to group 14.

  The data predicting unit 24 predicts the optimum shaft specifications based on the table shown in FIG. In the present embodiment, the weight and flex are predicted from the club speed, the flex level is predicted from the club pass, and the tone is predicted from the attack angle.

  For example, as described above, when the data classification unit 23 determines that the measured head behavior belongs to the group 14, the data prediction unit 24 determines that the optimum spec (weight is 40 to 50 [g], flex is “ A ”and the tone is“ medium tone ”).

  The result output unit 28 outputs the recommended golf club 1 based on the specifications of the shaft predicted by the data prediction unit 24. In the present embodiment, the swing response curved surface calculation unit 25, the time response curved surface calculation unit 26, and the simulation execution unit 27 can be omitted from the golf equipment fitting system 2.

  The data classification unit 23 is not limited to the table of FIG. 15, and can classify the head behavior stored in the swing data storage unit 22 based on various classification methods. Further, the data prediction unit 24 is not limited to the table of FIG. 16, and can predict the specifications of the golf club 1 based on various prediction methods.

  As described above, the golf equipment fitting system 2 of this embodiment includes the swing data acquisition unit 21 that acquires swing data from the sensor 11 attached to the golf club 1 and the swing data acquired by the swing data acquisition unit 21. Is measured by the sensor 11 by referring to the swing data storage unit 22 for storing the data, the data classification unit 23 for classifying the swing data stored in the swing data storage unit 22, and the swing data classified by the data classification unit 23. A data predicting unit 24 that predicts swing data of specifications that are not performed. Thereby, an appropriate golf equipment can be selected with a small number of trial hits.

  It is to be noted that a program for realizing each function of the golf equipment fitting system 2 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system and executed, thereby executing the above-described units. You may perform the process of. Here, “loading and executing a program recorded on a recording medium into a computer system” includes installing the program in the computer system. The “computer system” here includes an OS and hardware such as peripheral devices. Further, the “computer system” may include a plurality of computer devices connected via a network including a communication line such as the Internet, WAN, LAN, and dedicated line. The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. As described above, the recording medium storing the program may be a non-transitory recording medium such as a CD-ROM. The recording medium also includes a recording medium provided inside or outside that is accessible from the distribution server in order to distribute the program. The code of the program stored in the recording medium of the distribution server may be different from the code of the program that can be executed by the terminal device. That is, the format stored in the distribution server is not limited as long as it can be downloaded from the distribution server and installed in a form that can be executed by the terminal device. Note that the program may be divided into a plurality of parts, downloaded at different timings, and combined in the terminal device, or the distribution server that distributes each of the divided programs may be different. Furthermore, the “computer-readable recording medium” holds a program for a certain period of time, such as a volatile memory (RAM) inside a computer system that becomes a server or a client when the program is transmitted via a network. Including things. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient.

  Moreover, you may implement | achieve part or all of the function mentioned above as integrated circuits, such as LSI (Large Scale Integration). Each function described above may be individually made into a processor, or a part or all of them may be integrated into a processor. Further, the method of circuit integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. Further, in the case where an integrated circuit technology that replaces LSI appears due to progress in semiconductor technology, an integrated circuit based on the technology may be used.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Golf club 2 Golf equipment fitting system 11 Sensor 20 Reception part 21 Swing data acquisition part 22 Swing data storage part 23 Data classification part 24 Data prediction part 25 Swing response curved surface calculation part 26 Time response curved surface calculation part 27 Simulation execution part 28 Result output Part

Claims (22)

A swing data acquisition unit for acquiring swing data from sensors attached to a plurality of golf clubs having different specifications;
A swing data storage unit for storing the swing data acquired by the swing data acquisition unit;
A data classification unit for classifying swing data stored in the swing data storage unit;
A data prediction unit that predicts swing data of a specification that is not measured by the sensor by referring to the swing data classified by the data classification unit;
A golf equipment fitting system comprising: Using the swing data predicted by the data prediction unit, a swing response surface calculation unit that calculates a swing response surface by a response surface method;
Using the swing data predicted by the data prediction unit, a time response surface calculation unit that calculates a time response surface by a response surface method,
A simulation execution unit that calculates swing data of a spec that is not measured by the sensor from the swing response curved surface and the time response curved surface, and executes a simulation of a golf club swing based on the calculated swing data;
The golf equipment fitting system according to claim 1, further comprising: The golf equipment fitting system according to claim 2, wherein the simulation execution unit calculates a head posture at the moment of impact using the swing data predicted by the data prediction unit and the finite element method. The golf equipment fitting system according to claim 2, further comprising a result output unit that outputs a simulation result by the simulation execution unit. The golf equipment fitting system according to claim 4, wherein the result output unit converts a simulation result by the simulation execution unit into a natural language and outputs the result. The golf equipment fitting system according to claim 2, wherein the specs of the plurality of golf clubs used in the simulation have a difference in influence between each other within a predetermined range. The golf equipment fitting system according to any one of claims 1 to 6, wherein the data classification unit classifies the swing data stored in the swing data storage unit using a classification method based on unsupervised learning. The golf equipment fitting system according to claim 7, wherein the data classification unit classifies the swing data stored in the swing data storage unit using a self-organizing map as a classification method by the unsupervised learning. The swing data storage unit stores difference data, which is a difference between swing data of a specific golf club and swing data of another golf club, for each of a plurality of players.
When the swing data acquisition unit newly acquires the swing data of the specific golf club, the data prediction unit uses the difference data of the swing data belonging to the same classification as the newly acquired swing data to the swing 9. The golf equipment fitting system according to claim 1, wherein swing data of the other golf club is predicted using the difference data read out from the data storage unit and read out. The swing data acquisition unit measures the head behavior of the golf club,
The data classification unit determines which group of a plurality of preset groups the head behavior measured by the swing data acquisition unit belongs,
The golf equipment fitting system according to claim 1, wherein the data prediction unit predicts an optimum specification of the golf club based on the group determined by the data classification unit. The golf equipment fitting system according to any one of claims 1 to 10, wherein the sensor is a six-axis sensor that detects a triaxial acceleration and a triaxial angular velocity. The golf equipment fitting system according to any one of claims 1 to 10, wherein the sensor is a nine-axis sensor that detects three-axis acceleration, three-axis angular velocity, and three-axis azimuth. 11. The golf equipment according to claim 1, wherein the sensor includes a six-axis sensor that detects a triaxial acceleration and a triaxial angular velocity, and a geomagnetic sensor that detects a triaxial direction. Fitting system. The golf equipment fitting system according to any one of claims 1 to 13, wherein the sensor is attached to a grip of the golf club. The golf equipment fitting system according to any one of claims 1 to 14, wherein the swing data storage unit is a database storing the swing data of a plurality of players. The golf equipment fitting system according to any one of claims 1 to 15, wherein the plurality of golf clubs having different specifications are selected based on an L4 type orthogonal table defined by an experimental design method. Swing data acquisition procedure for acquiring swing data from sensors attached to a plurality of golf clubs having different specifications;
A swing data storage procedure for storing the swing data acquired in the swing data acquisition procedure in a swing data storage unit;
A data classification procedure for classifying the swing data stored in the swing data storage unit;
A data prediction procedure for predicting swing data of a specification not measured by the sensor by referring to the swing data classified in the data classification procedure;
A golf equipment fitting method comprising: Swing data acquisition procedure for acquiring swing data from sensors attached to a plurality of golf clubs having different specifications;
A swing data storage procedure for storing the swing data acquired in the swing data acquisition procedure in a swing data storage unit;
A data classification procedure for classifying the swing data stored in the swing data storage unit;
A data prediction procedure for predicting swing data of a specification not measured by the sensor by referring to the swing data classified in the data classification procedure;
A golf equipment fitting program for causing a computer to execute.   A golf swing classification method for acquiring swing data from a sensor attached to a golf club and classifying the swing data by a self-organizing map.   A golf shaft fitting system using the golf equipment fitting system according to any one of claims 1 to 16.   A golf shaft fitting method using the golf equipment fitting method according to claim 17.   A golf shaft fitting program using the golf equipment fitting program according to claim 18.

Images