KR20240128407A - A method for golf lesson and replay of golf round using wearable sensors - Google Patents

Abstract

The present invention relates to a golf lesson method using a wearable sensor, a golf round reconstruction method, a golf lesson device, and a golf round reconstruction device. The golf lesson method of the present invention may include a first step of obtaining, from wearable sensors attached to a plurality of points on a body of a golfer, first swing data including at least one of spatial coordinates of the points and changes (x, y, z, t) of the spatial coordinates on a time axis, and second swing data including changes (p, t) of pressure on a time axis of the points; a second step of generating swing information including at least one of the first swing data and the second swing data; a third step of comparing the swing information with reference swing information defined in advance in response to a skeletal model; and a fourth step of transmitting a signal to the wearable sensor so as to generate a notification at a corresponding point when the swing information deviates from the reference swing information by a predetermined range or more.

Description

{A method for golf lesson and replay of golf round using wearable sensors}

The present invention relates to a golf lesson method using a wearable sensor, a golf round review method, a golf lesson device, and a golf round review device.

In order to play golf well, it is important to have the correct swing form. For this, golfers receive lessons to learn the correct swing form when they first learn golf or after learning golf.

Golf lessons can be offline lessons or online lessons. In offline lessons, you receive instruction from a lesson pro offline. Online lessons can be obtained through apps that provide golf lessons available on the Internet.

Conventional offline and online lessons have the following problems:

Offline lessons have the disadvantage of being expensive and time-consuming. Also, since the lessons depend on the knowledge and memory of the person giving the lesson, the effectiveness is not constant and the teaching effect can vary depending on the individual differences of the person giving the lesson.

Most conventional online lessons require the use of cameras to capture swing videos from the front or side of the golfer. This is not easy in the typical narrow practice range environment. In particular, online lessons are nearly impossible in actual golf round situations on the golf course.

Also, when golfers encounter various situations that occur on actual golf courses, they forget what they learned and swing and respond based on instinct. Therefore, it is difficult to improve skills during actual golf rounds. You can play on the golf course with a lesson pro and receive coaching, but it is expensive and limited.

Therefore, there is a need to develop a golf lesson method that can consistently coach one's swing on the practice range and golf course.

Additionally, there is a need to develop golf lesson methods that can deal with various situations that occur during an actual golf round.

In addition, it is generally difficult to systematically analyze the content of a golf round actually performed, as it relies on the memory of the player or his/her companion. Therefore, a method is needed to systematically analyze the content of the golf round after the golf round to improve golf.

Additionally, there is a need for a way for golfers to continually improve their swing to suit their physical condition and changes in it.

The present invention is intended to solve the above-described problems.

An object of an embodiment of the present invention is to provide a golf lesson method and a golf lesson device using a wearable sensor that can save cost and time.

Another object of an embodiment of the present invention is to provide a golf lesson method and a golf lesson device using a wearable sensor that can consistently coach one's swing at a practice range and a golf course.

Another object of an embodiment of the present invention is to provide a golf lesson method and a golf lesson device that can provide lessons suited to various situations occurring in an actual golf round.

Another object of an embodiment of the present invention is to provide a golf round review method and a golf round review device using a wearable sensor that can systematically analyze and utilize the results after a golf round has ended.

The tasks of the present invention are not limited to the tasks described above, and other tasks not mentioned can be understood by those skilled in the art from the description below.

A golf lesson method of an embodiment of the present invention may include a first step of obtaining, from wearable sensors attached to a plurality of points on a body of a golfer, first swing data including at least one of spatial coordinates of the points and changes (x, y, z, t) of the spatial coordinates on a time axis, and second swing data including changes (p, t) of pressure on a time axis of the points; a second step of generating swing information including at least one of the first swing data and the second swing data; a third step of comparing the swing information with reference swing information defined in advance in response to a skeletal model; and a fourth step of transmitting a signal to the wearable sensor to generate a notification at a corresponding point when the swing information deviates from the reference swing information by a predetermined range or more.

In an exemplary embodiment, the wearable sensor includes an inertial sensor and a pressure sensor, the first swing data and the second swing data are obtained from the inertial sensor and the pressure sensor, and the skeletal model can be defined using a node that is a location to which the inertial sensor is attached.

According to an exemplary embodiment, the skeletal model includes at least one of a static skeletal model and a dynamic skeletal model, and at least one of the upright posture, the setup posture, and the backswing top posture of the golfer can be used as an origin coordinate to determine a change in the coordinates of a node of the skeletal model.

According to an exemplary embodiment, the method may include a fifth step in which lie information of a point where the golfer is located is obtained, and the reference swing information is defined to further reflect the lie information.

According to an exemplary embodiment, the lie information can be determined using map data of a golf course on which the golfer is playing and location information of the golfer.

In an exemplary embodiment, the lie information can be determined using the first swing data and the second swing data.

In an exemplary embodiment, the target point information intended by the golfer is obtained, and the reference swing information can be defined to further reflect the target point information.

According to an exemplary embodiment, the target point can be determined using map data of a golf course on which the golfer is playing and location information of the golfer.

According to an exemplary embodiment, the method may include a sixth step of obtaining motion information on various motions of the golfer by using the position information of the golfer, the first swing data, and the second swing data.

In an exemplary embodiment, a practice swing motion and a shot swing motion of the golfer can be determined from the motion information, and a golf lesson can be selectively provided to the golfer.

In an exemplary embodiment, using the motion information, if it is determined that the golfer has not moved for a certain period of time, at least one of transmitting a signal to the wearable sensor to generate a notification and transmitting a message to a preset contact can be performed.

According to an exemplary embodiment, the method may include a seventh step in which the reference swing information is changed by the golfer to generate individual reference swing information.

In an exemplary embodiment, the individual reference swing information may be generated by changing the first swing data and the second swing data in the skeletal model.

The golf round reconstruction method of the embodiment of the present invention may include a first step in which information obtained from the golf lesson method described above is stored as golf round information; a second step in which a golf round desired to be reconstructed is selected; a third step in which golf course graphic data and golfer graphic data are generated using golf round information corresponding to the selected golf round; and a fourth step in which the golf course graphic data and the golfer graphic data are output.

In an exemplary embodiment, the golf round information may include golfer location information, first swing data, second swing data, and golf course information.

In an exemplary embodiment, in the third step, the golfer graphic data may be generated as one of a golfer's skeletal model and a golfer's avatar.

According to an exemplary embodiment, the method includes a fourth step in which the first swing data and the second swing data are changed by the golfer to generate changed first swing data and changed second swing data, and the third step and the fourth step can be performed using the changed first swing data and the changed second swing data.

According to an exemplary embodiment, in the fourth step, the modified first swing data and the modified second swing data can be generated using the skeleton model.

According to an exemplary embodiment, in the fourth step, the first swing data and the second swing data obtained as a result of the golfer swinging while wearing the wearable device may be used to generate the modified first swing data and the modified second swing data.

The features of each of the above-described embodiments may be implemented in combination in other embodiments as long as they are not contradictory or exclusive to other embodiments.

The present invention described above has the following effects.

According to the golf lesson method and golf lesson device of the embodiment of the present invention, there is an advantage in that the cost and time of golf lessons can be reduced.

According to the golf lesson method and golf lesson device of the embodiment of the present invention, there is an advantage in that one can receive consistent coaching of one's swing at a practice range and a golf course.

According to the golf lesson method and golf lesson device of the embodiment of the present invention, there is an advantage in that lessons can be received that are suitable for various situations occurring during an actual golf round.

According to the golf round reconstruction method and golf round reconstruction device of the embodiment of the present invention, there is an advantage in that the golf round can be systematically analyzed and utilized after it is finished.

According to the golf lesson method and golf lesson device of the embodiment of the present invention, there is an advantage in that swing information can be continuously collected at a practice range, golf course, etc., and recorded and managed as a personal history for life.

The effects of the present invention are not limited to the effects described above, and other effects not mentioned can be understood by those skilled in the art from the description below.

Figure 1 is a configuration diagram of an embodiment of a golf lesson device and a golf round review device according to the present invention.
Figures 2 and 3 are schematic diagrams of the wearable sensor of Figure 1.
Figure 4 is a flowchart showing the overall flow of an embodiment of a golf lesson method according to the present invention.
Figure 5 is a conceptual diagram of an embodiment of a skeletal model according to the present invention.
Figure 6 is a flow chart of an embodiment of a general golf lesson method according to the present invention.
Figure 7 is a flow chart of an embodiment of a golf lesson method on a golf course according to the present invention.
Figure 8 is a flow chart of an embodiment of a method for distinguishing between a practice swing and a shot swing according to the present invention.
Figure 9 is a flow chart of an embodiment of a golf round restoration method according to the present invention.

Hereinafter, an embodiment of the present invention will be described with reference to the attached drawings.

The embodiments described below are intended to aid understanding of the present invention, and therefore the present invention is not limited to the embodiments described below. In addition, in the attached drawings, certain components may be exaggerated or reduced in order to aid understanding of the invention, and the present invention is not limited to the forms drawn in the drawings.

Referring to FIG. 1, the overall configuration of an embodiment of a golf lesson device and a golf round review device according to the present invention will be described.

The golf lesson and golf round review system (1) according to the present embodiment may include a server (50), a smart device (30), and a wearable sensor (10, 20).

The server (50) and the smart device (30) can be connected through a communication network (40). The wearable sensor (10, 20) and the smart device (30) can be connected through short-range communication.

Wearable sensors (10, 20) can be attached to a golfer's body to sense the golfer's swing and collect swing data. (Details of the swing data will be described later.) Wearable sensors (10, 20) can include a body sensor (10) and a shoe sensor (20).

The smart device (30) can create swing information using swing data received from the wearable sensor (10, 20). In addition, the smart device (30) can compare the golfer's swing information with the reference swing information to create corrected swing information and output it to the golfer. (Details of the swing information, reference swing information, and corrected swing information will be described later.)

The server (50) can collect, analyze, process, and store data acquired or generated from wearable sensors (10, 20) and smart devices (30).

Describe each component.

First, the wearable sensor (10, 20) is described.

As described above, the wearable sensor (10, 20) may include a body sensor (10) and a shoe sensor (20).

The body sensor (10) is attached to major parts of the golfer's body and can acquire various data on the golfer's swing (hereinafter, for convenience, 'swing data').

The body sensor (10) can obtain x, y, and z coordinates for a point in space that changes with respect to the time axis while the body part on which the body sensor (10) is installed is swinging. That is, data that can be obtained from the body sensor (10) can be expressed in the form of (x, y, z, t) (hereinafter, for convenience, referred to as 'first swing data'). Here, the x, y, and z coordinates require a reference point, and this reference point can be arbitrarily determined. (This will be described later.)

The main parts of the body to which the body sensor (10) is attached can be selected as the head, shoulders, pelvis, and hands. However, it is not limited to this, and other parts such as the knees and elbows may be included depending on the situation, and it may be worn only on some of the above parts.

In situations where you want to focus on correcting your golf swing in a limited space, such as a practice range, you can correct your swing more precisely by attaching additional sensors to the area you want to correct, depending on the golfer's level or the problems they have.

If you want to record only the swing data that is absolutely necessary for review after a golf round and not for the purpose of swing correction on an actual golf course, you can minimize the sensor attachment area and play more conveniently.

Below, for convenience of explanation, an example of a body sensor (10) being attached to the head, shoulder, pelvis, and hand is described.

Golfers usually wear a hat on their heads, a belt on their hips, and gloves on their hands. Therefore, instead of directly attaching it to their bodies, the body sensor (10) can be installed on a hat, belt, or gloves. In addition, women sometimes do not wear belts, and in such cases, it can be attached to clothing in the hip area instead of a belt. In addition, it can be easily attached to clothing using an attachment means such as a pin on the shoulder.

Meanwhile, since the swing basically rotates along the center line connecting the head and the spine, the body sensor (10) must be attached far from this rotation axis to obtain swing data that changes reliably according to the swing.

To this end, it is desirable to attach a body sensor (10) to the brim or side of a hat on the head, place the shoulder on the left side for right-handed people (on the right side for left-handed people), and place the pelvis on the left side of the belt (for right-handed people) at a point farthest from the center line of the spine among the belt positions, where the swing is not obstructed. It is desirable to attach the hand to the back of the hand of the glove so that it is not obstructed by the grip. It is also possible to wear it on the wrist in the form of a bracelet.

Meanwhile, the shoe sensor (20) is a sensor that measures the movement of the golfer's feet.

For example, the shoe sensor (20) can be installed in the insole of a golfer's shoe.

Through the shoe sensor (20), the x, y, z coordinates (x, y, z, t) of the moving feet and the change in pressure (p, t) applied to the feet can be obtained while the golfer is swinging. That is, the shoe sensor (20) can obtain (x, y, z, t) data (first swing data) and (p, t) data (hereinafter, for convenience, 'second swing data'). The first swing data and the second swing data of the shoe sensor (20) can be included in the golfer's swing data.

Next, the smart device (30) is described.

The smart device (30) may be a regular smartphone, but is not limited thereto.

For example, a smart device (30) may be a smart device in the form of a watch or a laser rangefinder suitable for golfers to wear, if it has computing power and communication capabilities to perform the necessary role between the sensor (10, 20) and the server (50) with a certain level of intelligence.

When a smart device (30) other than a smartphone, such as a watch, is used, it is possible to use a separate smartphone to secure a display (screen) necessary for reviewing the golf round content described later.

The smart device (30) can receive, from the body sensor (10), the spatial coordinates of the moving body part to which the sensor is attached and the change in the time axis of these coordinates (hereinafter, for convenience, '(x, y, z, t)'), i.e., the first swing data. In addition, the smart device (30) can receive, from the shoe sensor (20), the change in pressure applied to both feet (hereinafter, for convenience, '(p, t)'), i.e., the second swing data.

In addition, the smart device (30) can obtain location data (hereinafter, for convenience, '(X,Y)') about where the golfer is located based on the GPS information it possesses.

The software built into the smart device (30) can infer various movements of the golfer using the first swing data and the second swing data.

For example, the smart device (30) can use the first swing data and the second swing data to infer all information about a light swing (so-called wagging) such as a golfer's body warm-up, a setup (so-called address) posture, a practice swing, a swing actually hitting a ball (hereinafter, for convenience, 'shot swing'), and its trajectory and swing speed.

Because the first swing data and the second swing data of each of the above movements are different. Therefore, by using the difference in these data, the movement that the golfer is currently taking can be inferred. (The detailed method of inferring will be described later.)

For example, the swing speed of the warm-up (waggling) is slower than the swing speed of the shot swing, and the swing size of the warm-up (waggling) is smaller than the swing size of the shot swing. Also, the setup posture is a state where the entire body is still for a considerable period of time. In this way, since each movement of the golfer is very different from the first swing data and the second swing data, the movements can be distinguished using this.

In addition, the smart device (30) can also infer abnormal postures, such as when a golfer falls or gets injured on a golf course or does not move for a long time. That is, the first swing data and the second swing data of the abnormal posture are very different from the first swing data and the second swing data of the posture when walking or exercising normally, so the golfer's abnormal posture can be inferred through this.

Meanwhile, the smart device (30) can provide golf lessons to golfers.

Golf lessons are intended to find problems in a golfer's swing and correct them. This can be done by analyzing the golfer's first swing data and second swing data and making corrections.

That is, the smart device (30) improves the golfer's swing by suggesting desirable first swing data and second swing data for the first swing data and second swing data performed by the golfer.

In the process of correcting a golfer's swing, the first swing data is used to correct the golfer's swing trajectory and tempo. Afterwards, the swing can be additionally corrected using the second swing data. The second swing data corresponds to the movement of the center of gravity (so-called weight shift) during the golfer's movement, and is important information for determining whether a normal swing has occurred.

In order to correct a golfer's swing in this way, the smart device (30) may have data on a desirable swing (hereinafter, for convenience, referred to as a 'reference swing'). The data on the reference swing may be built into the smart device (30) or received from a server (50).

The smart device (30) can compare the golfer's first and second swing data with the reference swing data. If there is a difference as a result of the comparison, feedback can be provided to the golfer by sending an alarm to the corresponding part of the body at each stage using a method such as vibration.

For example, if the head moves significantly during the swing, the body sensor (10) attached to the head can sound an alarm, for example, a haptic vibration. Also, if the hand moves significantly outside the target line during the backswing, the body sensor (10) attached to the hand or wrist can sound a vibration. (Details will be described later.)

Referring to FIGS. 2 and 3, embodiments of a body sensor (10) and a shoe sensor (20) are described.

Describes the body sensor (10).

An inertial sensor (11), an alarm unit (15), and a communication unit (14) may be connected to the central processing unit (12). In addition, a memory (13) may be connected to the central processing unit (12). In addition, the body sensor (10) may include a power supply unit (not shown) such as a battery and a charging device.

Describes the shoe sensor (20).

An inertial sensor (21), an alarm unit (25), and a communication unit (24) may be connected to the central processing unit (22). In addition, a memory (23) may be connected to the central processing unit (22). The shoe sensor (20) may include a pressure sensor (26). In addition, the shoe sensor (20) may include a power supply unit (not shown) such as a battery and a charging device.

Describe each component.

An inertial sensor (IMU, Inertial Measurement Unit) (11,21) measures gravitational acceleration when the sensor is stationary, and measures movement acceleration (hereinafter referred to as “acceleration”) and angular velocity within space of a specific part to which the sensor is attached when moving. Based on this, through calculations by a central processing unit (12,22), the spatial coordinates to which the part to which the sensor (10,20) is attached has moved and the changes (x, y, z, t) of these coordinates on the time axis, i.e., the first swing data, can be obtained.

These coordinates are determined based on a reference point. That is, the location corresponding to the origin, which is indicated by (0,0,0) in the coordinates. This can be determined arbitrarily. For example, in the case of a sensor attached to a hand, the location of the hand that is being set up before a swing can be set as the origin. That is, the hand is still for a long time during the set-up, and this location can be set as the coordinate (0,0,0), and the path (x,y,z,t) moved from it to perform the swing can be calculated using the acceleration and angular velocity. Other body parts, such as the head or shoulders, can be set as a reference point in the same way, and the path moved from it can be set as (x,y,z,t).

In addition, the first swing data (x, y, z, t) may include, in addition to the coordinates of the moved path, information about the acceleration and angular velocity of the corresponding point, which are its original information, and may also include extracted information such as speed, movement path, distance, angle, and inclination that appear during its calculation. It may be understood that using some of the above extracted information is also using the first swing data.

The first swing data can be stored in the memory (13, 23) within the body sensor (10) and the shoe sensor (20).

By command of the central processing unit (12, 22), the stored data can be transmitted to a smart device (30) through the communication unit (14, 24). The communication unit (14, 24) can be a low-power communication device, such as Zigbee or low-power Bluetooth.

Typically, a golfer carries a smart device (30) or places it around the golfer. That is, the smart device (30) is located at a distance where it can communicate with the low-power communication device of the wearable sensor (10, 20), so the smart device (30) and the wearable sensor (10, 20) can communicate using the low-power communication device.

If the smart device (30) is not near the golfer, data can be stored in memory (13, 23) and data transmission can be initiated when the smart device (30) is detected and communication is determined to be possible.

Golfers who find it inconvenient to carry or keep a smart device (30) on their body while playing can add a GPS (not shown) to either the shoe sensor (20) or the body sensor (10). In this case, the location of the golfer is recorded at set intervals through the sensor (10, 20) and stored in the memory (13, 23) within the sensor (10, 20). When the sensors (10, 20) encounter the smart device (30), the first swing data, the second swing data, and the GPS location information (X, Y) can be transmitted to the smart device (30).

Meanwhile, the alarm unit (15, 25) has a function to notify when the golfer's swing is different from the standard swing. For example, the alarm unit (15, 25) can be used to notify the golfer of vibration, electrical stimulation, buzzer, voice, etc. If the golfer is wearing earphones (not shown), voice feedback can be provided for the relevant area through a smart device (30).

As described above, the shoe sensor (20) may include a pressure sensor (26). The pressure sensor (26) may measure changes in pressure (p, t) applied to both feet to obtain second swing data.

Pressure sensors (26) can be simply installed one on each foot, two on each foot. Alternatively, a total of four pressure sensors (26) or more can be installed to obtain sufficient data on the movement of the body's center of gravity by distinguishing between the front and back of each foot, or distributed sensors that can measure the distribution of pressure over the entire foot can be installed.

In addition, the shoe sensor (20) may further include an inertial sensor (21). The inertial sensor (21) may include two or more in total, one for each foot, to measure the difference in inclination and position between the two feet and to distinguish the difference in inclination and position between the front and the back of each foot.

This allows the golfer to determine the lie at the point where he is addressing the ball. This feature can be especially useful when visiting a golf course and actually playing.

For example, there are four types of slopes in the address point lie. That is, when the ball is above the feet, when the ball is below the feet, when the left foot is high, and when the right foot is high. When a golfer is positioned on these slopes, the (x, y, z, t) data and (p, t) data transmitted from the inertial sensor (11, 21) and the pressure sensor (26) detected by the golfer's body sensor (10) and shoe sensor (20) differ in each case. Through this, the smart device (30) can infer slope information where the golfer is located. (Details will be described later.)

Referring to FIG. 4, a golf lesson method according to the present invention is described.

First, the initial setup process of the golf lesson method according to the present invention is described.

Golfers can enter basic information (s1) through a smart device (30).

Basic information can include basic information about the body, such as gender, age, height, weight, right/left handedness, and arm/leg length. In addition to numbers or letters, basic information can include photos or videos showing the golfer's physical condition. This makes it easy to understand the overall structure of the body, such as the golfer's height and arm/leg length.

The golfer first wears a sensor (10, 20) (s2), enters golf club information (s3), and assumes a setup (aka address) posture for each club (s5).

The golf club information input step (s3) and the setup posture step (s5) can be performed sequentially for all clubs, from the driver to the shortest wedge.

Through this, you can obtain a skeleton model of the entire body from head to feet for each club. (Detailed skeleton models will be described later.)

The setup posture of s5 is a posture in a stationary state. As described above, the sensor (10, 20) measures acceleration and angular velocity in response to movement, so that the reference point can be set before reaching the setup posture and the path moved from there can be extracted to determine the coordinates of the final setup posture.

That is, the sensors attached to each part of the body move from the reference origin and set the final (after time t) stationary position (x, y, z) as (x, y, z, t), and these data are set as a stationary skeletal model at the time of setup (hereinafter, the stationary skeletal model at the time of setup is referred to as the 'stationary skeletal model', and the 'skeletal model' referred to throughout this specification means a dynamic skeletal model obtained when the sensor (10, 20) is in a moving state, unless there is a special technique). A more detailed method of obtaining the stationary skeletal model is described later in Fig. 5.

The above process is repeated for each club to obtain a stationary skeleton model, and then a swing is performed to obtain a normal skeleton model, that is, a skeleton model in a moving state. As described above, the skeleton model is summarized as time-dependent path data (x, y, z, t) created when the part to which the sensor is attached moves. In other words, the s7 process obtains both the setup process and the skeleton model during the swing process.

Based on this, a golf lesson (s9) can be performed. (Details will be described later) In the golf lesson step (s9), the golfer's actual swing can be compared with a reference swing, and the golfer's swing can be corrected.

During the above process, acquisition of a club-specific skeleton model (s7) can be performed only for one or a few representative clubs, if necessary, and the setup process can be omitted and only the swing process can be performed.

This is because of the following two reasons. First, it is because it takes into account that the posture and swing form of a standard full swing shot taken by a golfer in a flat lie situation should not differ greatly depending on the club. In other words, since the length of the club varies depending on the club, the distance at which the club head and the ball fall from the body varies greatly, but the posture taken by the golfer with his or her body and the basic form of the swing do not differ greatly, so there is no problem in performing golf lessons (s9) even if a skeletal model is taken with some clubs. Second, as described above, since the sensor (10, 20) measures acceleration and angular velocity in response to movement, in order to obtain a stationary skeletal model during setup, the reference origin must be set before reaching the setup posture, and the path moved from there must be extracted to determine the coordinates of the final setup posture, so the process of obtaining such a skeletal model is somewhat complicated.

Accordingly, the static skeleton model for the setup posture for each club is limited to the initial setup stage or cases where the setup posture of the golfer is particularly suspicious and there is a strong need to correct it, and when correcting shots at most practice ranges or golf courses, the dynamic skeleton model, that is, the movement data (x, y, z, t) obtained by attaching a sensor and actually swinging can be used to correct the golfer's swing.

A predefined reference swing can be stored in the smart device (30). If the reference swing stored in the smart device (30) is not sufficient for teaching, the golfer's swing can be analyzed and the swing can be corrected based on data related to more reference swings stored in the server (50). The swing can be corrected by voice, display, or a combination thereof through an alarm attached to a sensor or through the smart device (30).

Referring to FIG. 5, a skeletal model of an embodiment according to the present invention is described.

Figure 5 is a drawing based on a right-handed person. Figure 5(A) is a front view of the golfer's setup posture, and Figure 5(B) is a side view.

The skeletal model refers to a model that views the inertial sensors (11, 21) among the wearable sensors (10, 20) as nodes and connects them to complete the posture and movement of the entire skeleton.

Nodes in the skeletal model may include a head node (61), a left shoulder node (62), a left pelvis node (64), a hand node (66), and two feet nodes (67, 68). In addition, nodes in the skeletal model may include a virtual shoulder node (63) and a virtual pelvis node (65), although sensors (10, 20) are not attached thereto. The distance between all nodes, including the virtual nodes, may be inferred through body structure such as basic body information entered in Fig. 4, such as height, weight, and arm/leg length, or may be entered as numbers to more accurately represent the distance between nodes.

The skeletal model can be broadly divided into a stationary skeletal model in which the golfer assumes a setup posture and a skeletal model in a moving situation such as a practice swing or shot swing.

As described above, the inertial sensor (11, 21) is a device that reports the acceleration and angular velocity of the point where the sensor is attached, and generates data when the sensor moves. Therefore, when a golfer takes a setup posture, the setup skeleton model is completed by looking at the data that appears in the moving section between the state before setup and the state where setup is completed and the situation where it remains stationary for a long time.

The starting position that a golfer takes before completing the setup position can be, for example, an upright position with both feet together and the hand holding the golf club facing downward. The upright position is the same position as when measuring height. As described above, the distance between sensors (nodes) determined using basic body information (height, weight, arm/leg length, etc.) is most consistent in the same position, so the upright position is desirable as a starting position before the setup.

When the setup posture is assumed from the upright posture, the inertial sensor (11, 21) can be used to determine how much each node has moved in space from the upright posture. That is, the position of each node in the upright posture is set as the origin of each node, and then the path that each node has moved while assuming the setup posture is determined to determine the final coordinates of each part. That is, the relative positions for postures such as the bending of the head, shoulders, and waist, the normal position of the hands, and the spreading of both feet starting from the upright posture and until the body bends and completes the setup posture can be used as a stationary skeleton model.

The stationary skeleton model must start moving from a reference posture (an upright posture holding a golf club in the above example) that the smart device (30) knows in advance before completing the setup posture. Through this, the smart device (30) sees how much each body part has moved from the reference point set in advance by the golfer for setup, and sets the coordinates and period of the final destination point as (x, y, z, t) to determine the stationary skeleton model.

The skeletal model in motion, or the general 'skeletal model', can be determined by the movement data (x, y, z, t) of each node in a dynamic situation where the body moves, such as a swing. That is, when a golfer makes a relatively normal swing, each node shows a different movement depending on its attached location. That is, the sensor located on the head shows some horizontal and vertical movement, and the sensor located on the shoulder generates data corresponding to the trajectory of the shoulder movement.

The sensors attached to the hands show the most active and large movements. The sensors attached to the pelvis and both feet also have different characteristics and produce swing data that matches the normal swing and the human body structure. The skeletal model uses the (x,y,z,t) data of these nodes. These (x,y,z,t) coordinates also need an origin.

It is desirable to set the origin of these coordinates as the position when the sensor is stationary. For example, since a golfer is stationary for a long time when taking a setup posture, the position where each node is stationary can be set as the origin (0,0,0) of each node, and the path moved from there can be set as (x,y,z,t). In other words, in the case of a sensor attached to a hand, the position of the hand when the golfer takes a setup posture and is stationary for a long time can be set as the origin of the coordinates, that is, (0,0,0), and then the trajectory moved while performing a backswing and forward swing can be calculated using the acceleration and angular velocity sent by the inertial sensor (11,21), thereby completing (x,y,z,t).

Also, the origin can be redefined during the swing. For example, the momentary stopping point when the backswing stops and changes to the forward swing (aka downswing) can be set as another origin (0,0,0), and from that point on, (x,y,z,t) can be calculated to divide the trajectories of the backswing and forward swing into (x,y,z,t) data consisting of two origins. Since the inertial sensor obtains the trajectory from the acceleration and angular velocity by integration, errors can accumulate. To this end, errors can be reduced by resetting the coordinates in the middle.

In addition, when the smart device (30) creates an entire skeletal model using the data of each of these sensors, it additionally checks which sensor part of the body each data comes from. This is necessary to make it easy for people to see by graphically connecting nodes in the skeletal model with lines. To this end, a method of inferring the attachment location of each sensor by looking at the characteristics of the data can be used.

That is, as described above, the data generated by the sensors attached to each part of the body during operation have distinctly different patterns, so this is a method of identifying the attachment location of the sensor from which the data is generated by utilizing this. That is, when the sensors (10, 20) attached to each part of the body move and generate each data, the smart device (30) can see the pattern of the data and determine which part of the body the data comes from, and use it as a skeletal model of the entire golfer's body.

As an alternative to the above method, a method may be used in which the positions of each sensor (10, 20) are determined in advance and the golfer wears each sensor in the determined position. However, this may be inconvenient because the golfer must wear each sensor separately in the determined position.

The skeletal model can be useful for identifying a golfer's swing posture, ideal posture, slope, and reviewing a golf round.

The skeletal model can be provided in a graphic form by connecting nodes with lines so that a lesson pro (human) or a golfer can easily understand it with their eyes, but a smart device (30) can determine the situation by simply inferring the shape with only the node data. In other words, the smart device (30) can correct the swing, infer the abnormal posture, etc. with only the change in the node (x, y, z, t) over time without necessarily connecting the nodes with lines depending on the situation.

Referring to FIG. 6, an embodiment of a golf lesson method of the present invention is described.

The golf lesson method of this embodiment is not limited, but is a golf lesson method suitable for use in places where there is only one place to swing, such as a practice range, and where the lie is flat. (Hereinafter, for convenience, the golf lesson is in 'normal mode')

As described above, when a golfer wears a wearable sensor (10, 20) and swings, the wearable sensor can obtain the golfer's swing data, i.e., first swing data and second swing data (s91).

As described above, the first swing data may include spatial coordinates of a moving body part (node of the skeletal model) to which an inertial sensor is attached, and changes in these coordinates along a time axis (x, y, z, t). The second swing data may include changes in pressure (p, t) of a body part (node of the skeletal model) to which a pressure sensor is attached.

In a smart device (30), swing information can be created using the first swing data and the second swing data (s93).

Swing information can include swing path and swing tempo. Swing path is the path of movement in space that each body part moves during a swing. Swing tempo is the speed at which each body part moves.

Meanwhile, in order to correct a golfer's swing, swing information (hereinafter, for convenience, 'reference swing information') that is to be compared with the golfer's swing information can be defined in an appropriate manner. The reference swing information can be stored in a smart device or obtained in real time from a server, etc.

The baseline swing information is information about the posture and movements that the golfer should have throughout the entire process from setup to backswing and forward swing.

The reference swing information can be expressed as the position of each node in the skeletal model and its change over time (x, y, z, t).

Reference swing information for a golfer's posture (setup, etc.) can include appropriate positions of each body part in a stationary state for each skeletal model. As described above, reference swing information for a golfer's posture (setup, etc.) can be expressed as the position (x, y, z, t) of each node relatively distant from a reference point of coordinates determined in advance in a stationary skeletal model at the time of setup (t).

The reference swing information for a golfer's swing (shot) can be information about the swing trajectory and swing tempo suitable for each skeletal model. In other words, the golfer's swing motion can be divided into time-axis, and the body trajectory and the change speed of the body trajectory can be the timing. The reference swing information for a golfer's swing (shot) can be expressed as the position change (x, y, z, t) of each node in the skeletal model over time.

As described above, the golfer's swing information may be at least one set of the first swing data and/or the second swing data. In addition, the golfer's swing information may further include a swing trajectory, a swing tempo, etc., created by combining a plurality of the first swing data and/or the second swing data.

In substantially the same way, the reference swing information may also be a set of at least one first reference swing data and/or at least one second reference swing data. In addition, the reference swing information of the golfer may further include a reference swing trajectory, a reference swing tempo, etc., which are created by combining the first swing data and/or the second swing data. The reference swing information may be created for each skeletal model. For example, the skeletal model may be appropriately classified using factors such as height and body type, and the reference swing information may be standardized for each classified skeletal model.

In addition, the standard swing information can be continuously updated for each type of golfer when a lot of data on swings is accumulated in the server (50). For example, data on desirable swings for each skeletal model of the golfer can be collected and learned using artificial intelligence to create and continuously update standard swing information for each skeletal model.

In addition, the above-described standard swing information can be created using a skeleton model for each individual golfer by utilizing the knowledge of a lesson pro (person). In addition, a method can be used in which the golfer memorizes the most desirable swing among the swings he or she performed with the help of a lesson pro or on his or her own as standard swing information on a smart device (30).

As explained above, the reference swing information can be a set of reference swing data, i.e. (x,y,z,t) data and (p,t) data.

The smart device (30) compares the golfer's swing information with the reference swing information corresponding to the golfer's skeletal model (s95).

If the golfer's swing information deviates from the reference swing information, the golfer's swing can be determined to be undesirable. Therefore, in this case, the golfer can be induced to correct his/her swing by himself/herself by providing an alarm, such as vibration, through a sensor (10, 20) attached to the relevant part of the golfer's body (s97).

First, when a golfer sets up, the golfer's static skeletal model, that is, the static swing information (hereinafter, for convenience, 'static swing information') can be identified. The smart device (30) selects the setup information (hereinafter, for convenience, 'static reference swing information') from the reference swing information corresponding to the identified golfer's skeletal model. The smart device (30) compares the golfer's static swing information with the static reference swing information, and if there is a large difference, it can sound an alarm, such as a vibration, for the wearable sensor (10, 20) attached to the relevant part.

Next, when a golfer swings, the swing information of the golfer's dynamic state (hereinafter, for convenience, 'dynamic swing information') is compared with the reference swing information (hereinafter, for convenience, 'dynamic reference swing information') regarding the swing trajectory and/or swing tempo corresponding to the golfer's skeletal model, and if the difference is large, an alarm such as vibration can be sounded for the wearable sensor (10, 20) attached to the relevant part.

However, as described above, in order to teach about the setup posture, the golfer must take a reference posture and extract the path from there to the setup in order to obtain a static skeleton model, which is inconvenient to do for every shot on the golf course. Therefore, lessons using the setup and the resulting static swing information can be conducted at the initial setup stage or when the golfer's setup posture is particularly suspicious and there is a strong need to correct it.

If the swing trajectory of the golfer's swing information deviates from the swing trajectory of the reference swing information by a large amount, an alarm such as vibration may sound for the wearable sensor (10, 20) attached to the relevant area. That is, the smart device (30) may transmit a control signal to the wearable sensor (10, 20) attached to the relevant area to sound an alarm.

For example, if the hands are too close to or too far from the body during the backswing, this is considered a deviation from the trajectory suggested by the reference swing information, and vibration is generated in the body sensor (10) attached to the hand. As another example, if the head is lifted too quickly before the end of the swing, it is considered a head-up, and vibration can be generated in the body sensor (10) attached to the head.

When correcting a swing, tempo can also be considered important. In other words, if the speed at which each part of the body moves is too slow or fast, or if it is not constant and changes according to the orbit, you can find it and create vibration in that part.

For example, when the speed slows down during the impact section while coming down from the backswing to the downswing. Golf is a game where flow is important, so it is desirable for the sequential speed change to be constant for each stage of the swing. In other words, the club is gently sent back during the backswing, and the club is accelerated during the forward swing and is maximized at the impact. In this way, when the golfer's swing is different from the standard swing in tempo and timing, a method of applying vibration to the relevant part can be used.

Meanwhile, the present embodiment may additionally include a swing re-correction step (s99).

This is because, if a golfer determines that the standard swing suggested by the smart device (30) is not completely suitable for him/her due to his/her physical condition or other circumstances, it is desirable for the golfer to practice by modifying part of the swing.

An example of swing re-correction is as follows:

The golfer's own swing and the reference swing provided by the smart device (30), that is, the swing that serves as the reference for correction, can be expressed as a skeletal model (Fig. 5).

A smart device (30) can provide a display and an editing means. In the case of a smartphone, a touch sensor or, if a PC is used separately, a mouse or other input device can be the editing means.

A smart device (30) can use a skeletal model to temporally segment a golfer's swing into three-dimensional information and view it from various angles.

In other words, golfers can observe their swings and corrected swings by changing the direction, such as front, side, and top. In addition, golfers can use the editing tool to correct the swing suggested by the smart device to suit themselves, and through this, they can find and practice the swing path and tempo that suits them more effectively.

It is difficult to uniformly standardize and enforce a human swing, and there are individual differences in particular, so the concept of allowing correction of swings to suit each individual's physical condition is very useful in actual use.

As described above, a golf swing modified by the golfer himself can be stored as separate reference swing information (hereinafter, for convenience, 'personal reference swing information') in a smart device, etc. In this way, during a subsequent golf lesson, the golfer can select one of the reference swing information provided by the smart device (30) or the personal reference swing information, and correct his/her swing by comparing the selected swing information with his/her actual swing.

Referring to FIG. 7, the use of the golf lesson method of the present invention on a golf course is described.

The swing on the golf course and the swing on the practice range may be different. Therefore, in this embodiment, a golf lesson method that is more effective for use on the golf course is proposed. (Hereinafter, for convenience, 'field mode or golf course mode')

After arriving at the golf course, the golfer wears a sensor (10, 20) and carries a smart device (30) before arriving at the first hole.

The smart device (30) obtains golf course information through GPS information (s110). The golf course information may include the golf course where the golfer is located, map data of the golf course, and the hole currently located.

Golfers who do not like to always carry a smart device (30) can obtain location information through a shoe sensor (20) or body sensor (10) equipped with GPS. In this case, they can wear only the sensors (10, 20) on their body and play, and then transfer the information recorded in these sensors (10, 20) to a smart device (30) when the situation allows, to receive coaching or use it for future replay.

Below, an example is given of a situation where a smart device (30) plays together with a golfer in close proximity.

The smart device (30) can obtain the golfer's location information and the lie information of the golf ball (s120, s130). In addition, the smart device (30) can additionally obtain the target point information of the shot (s140).

In Fig. 7, for convenience, these steps are illustrated as being performed in advance, but the step of obtaining the golfer's position information (s120), the step of obtaining the golf ball's lie information (s130), and the step of obtaining the target point information (s140) can be performed at any point where the golfer intends to swing.

Although this information can be used from the tee box, it is often used after the second shot, so how to use this information is explained in detail in the second shot.

Meanwhile, the smart device (30) can track all movements made by the golfer (s150).

Actions that a golfer can take on a golf course include walking, riding a cart, warming up, stopping, preparatory actions using a golf club (so-called wagging), setup, practice swing, shot swing for hitting a shot, actions after a shot, or in rare cases, accidents such as falling.

When these movements are analyzed using the skeletal model described above, there is a clear difference between the first swing information and the second swing information of each movement, and among these, information necessary for the golfer's lessons can be extracted.

For example, all swings, including practice swings, have a certain swing range through hand movement and the feet are still in a certain position spread apart at the time, which makes them very different from other movements that are not directly related to the shot.

Through this, the smart device (30) can ignore simple preparatory movements and find meaningful movements among the movements that the golfer takes on the golf course. For example, the most meaningful movements can be a practice swing or a swing that actually hits the ball (hereinafter, for convenience, a 'shot swing').

The smart device (30) can provide feedback to the golfer during a practice swing to correct the swing. During an actual shot swing, the feedback can be stopped to provide psychological stability, but all aspects of coaching within the golf course can be selected by the golfer.

If a golfer does not want coaching within the golf course, as explained above, he or she can wear only the sensors (10, 20) on his or her body and play, and then transfer the data recorded by these sensors (10, 20) to a smart device (30) when the situation allows, and use it for coaching or replay. Next, we will explain a situation where a smart device actively intervenes in the golfer's play.

When the golfer steps up to the first hole tee box, the smart device intervenes in the practice swing and shot swing to provide a golf lesson (s160, s170, s180).

Generally, the tee box is a flat lie similar to a practice range. Therefore, golf lessons on the tee box can use the golf lesson method of the general mode (Fig. 6). That is, the golfer's swing information can be compared with the reference swing information to give an alarm to the golfer.

Once the golfer completes his swing on the tee box, he moves to position for his second shot.

Even at the second shot position, the smart device can intervene in the practice swing and shot swing to provide golf lessons (s160, s170, s180).

However, unlike the tee box, it is better to conduct a golf lesson in field mode rather than a golf lesson in general mode at the location of the second shot. This is because, unlike the practice range, the lie of the golf ball at the point where the golfer is trying to swing is different on the golf course at locations other than the tee box. It is better to conduct a golf lesson that reflects this.

To this end, when the golfer reaches a position where he or she must make a second shot, it is desirable to perform at least one of the following steps: obtaining golfer position information (s120), obtaining golf ball lie information (s130), and obtaining shot target point information (s130). This information can be used for golf lessons in field mode.

The golfer's location information is the current location information (X, Y) of the golfer, and can be obtained using the GPS information of the smart device. The GPS (X, Y) data can also be used as information for positioning the golfer's avatar within the virtual golf course during the process of reviewing the content of a golf round after play.

The lie information of a golf ball is information about the terrain on which the golf ball is placed.

The lie information of a golf ball can be extracted using the map data of the golf course and the location information of the golfer. The map data of the golf course can be downloaded and stored in advance in a smart device (30).

Therefore, by linking with the map data of the golf course, general situational information such as whether the golfer's current location is a fairway, rough, hazard, or bunker can be extracted.

The general situational information here is because the error of the location data (X, Y) provided by GPS technology is on the order of several meters, so it may be difficult to obtain information about the exact terrain, i.e. slope, of the golfer's location.

Therefore, the information obtained by the shoe sensor (20) can be additionally used to supplement the lie information of the golf ball.

The method for determining the slope of the golfer's location using a shoe sensor (20) is as follows.

The types of slopes and the postures that a golfer can take at each time and the signals from the sensors (10, 20) according to each are described. The description below is based on right-handedness.

First, the first swing data (x, y, z, t) for the left foot uphill situation can be obtained by the inertial sensor (21) located in the shoe sensor (20) for the posture in which the left foot is high and the right foot is low. In other words, the situation in which the left foot is high and the right foot is low can be extracted through the slope information transmitted by the inertial sensor (21). In addition, it can be known as supplementary data through the pressure sensor (26) located in the shoe sensor (20) that greater pressure is applied to the right foot than to the left foot.

However, the signal transmitted to the pressure sensor (26) may vary depending on whether the golfer is positioned parallel to the slope or positioned against the slope to match the gravity of the Earth. Therefore, it is desirable to determine the slope condition of the ground by synthesizing the signals (x, y, z, t) sent by the inertial sensor (21). In other words, it is desirable to determine the slope by synthesizing the signals of the pressure sensor (26) and the inertial sensor (21).

Next, in the case where the left foot is low and the right foot is high, the shoe sensor (20) sends a signal, which is the opposite of the above example, and the situation can be identified through this.

Also, when the toes are uphill, the front of both feet is higher than the back, and the slope is determined accordingly. Also, a lot of pressure is usually applied to the heels. Therefore, using the (x, y, z, t) and (p, t) data between them, it is possible to infer that the golfer is positioned on an uphill toes. When the toes are downhill, the interpretation is the opposite of the above case.

In the case of a compound slope, for example, when the toes are uphill and the left foot is higher than the right foot at the same time, the situation described in the above example is applied simultaneously, and the shoe sensor (20) extracts a signal from the inertial sensor (21) and the pressure sensor (26) indicating that the left foot is higher than the right foot and the front of both feet is higher than the back at the same time. Using this data, the smart device can identify the situation.

There is also a method of utilizing the slope information sent by the inertial sensor (11) attached to the body sensor (10) to determine the slope. That is, when a golfer takes a setup posture on a slope, the slope of the sensor (10) attached to the head or pelvis, etc., can react to point in a specific direction depending on the posture, and this can also be utilized.

After the information about the location of the second shot and its terrain has been determined, when the golfer takes a practice swing, the smart device (30) can determine whether the golfer is making a swing that is suitable for the terrain conditions of the current second shot location based on the pressure data transmitted from the shoe sensors (20) installed on both feet and the swing data transmitted from the body sensor (10).

To this end, the smart device (30) can use a swing that matches the terrain conditions as reference swing information. A swing that matches a specific terrain condition is different from a swing that matches a flat terrain such as a practice range. Therefore, it is desirable for the smart device (30) to have reference swing information (hereinafter, for convenience, 'field reference swing information') that matches each terrain condition. The field reference swing information can be defined using golf course information, golfer location information, golf ball lie information, shot target point information, etc.

That is, in a golf lesson in field mode, the golfer's swing information and the field-based swing information can be compared (s170) and an alarm for swing correction can be output to the golfer (s180).

For example, as a general rule for shots on slopes, it is advisable not to shift the golfer's weight too much, especially on steep slopes.

Therefore, the field-based swing information on such slopes may have different reference values for weight transfer compared to the general reference swing information. In field lessons, it is possible to determine whether this rule is well observed through the pressure (p, t) data transmitted from the shoe sensor (20).

Meanwhile, in a shot situation like this, the smart device (30) can also suggest an appropriate strategy (hereinafter, for convenience, 'golf strategy information') for the golfer (s185).

For example, in a difficult lie situation based on the golfer's skill level, advice could be given to hold the club shorter and focus on hitting the ball lightly.

That is, the smart device (30) can recognize to some extent various situations that a golfer may encounter in various places such as the fairway or rough rather than the tee shot on the golf course through sensors (10,20) and the smart device (30), and this can help in making correct decisions such as strategy and club selection in addition to simple swing lessons for the golfer.

Here's an example of a golf lesson using shot target information.

Depending on the golfer's skill level, when he or she approaches the green after completing a second or third shot, he or she may be in a situation where he or she must make a control shot, such as an approach shot. Unlike a full shot, a control shot is a shot that requires the swing size and tempo to control distance.

In putting on the green, it is very important to adjust the swing size and tempo to match the distance. To this end, the smart device (30) can calculate the remaining distance based on the current location of the golfer and map data provided by the golf course, and suggest an appropriate swing size and tempo to the golfer.

That is, the smart device (30) can identify not only the swing trajectory but also the size and tempo of the swing during a practice swing by using (x, y, z, t) and (p, t) data. That is, it can identify situations such as when a short distance is left but the backswing size is too large (judged using x, y, z, t), or when the weight transfer is excessive even though it is not a strong shot (judged using p, t).

By identifying this, feedback can be given to the golfer through vibrations, etc., to the problematic area, and the golfer can obtain improved results through actual shot swings after sufficiently practicing this. Therefore, it is desirable for the smart device (30) to be able to distinguish between practice swings and shot swings. The method of distinguishing between practice swings and shot swings will be described later.

Meanwhile, the golf lesson method of the present embodiment can obtain and utilize other useful information through the golfer's movements within the golf course in addition to coaching on swings (s190). In addition, all information related to a golf round can be stored in a smart device and/or server (s195).

The golfer and the smart device (30) repeat the above steps (s120 to s195) until the game of one hole is completed, including a tee shot, a second shot (or a third shot depending on skill level), an approach shot, and putting.

The step of obtaining other useful information related to golfers (s190) is explained in detail.

Other useful information for a golfer may include useful information that can be obtained through the golfer's movements on the golf course other than information about the swing.

This useful information can be generated using the golfer's location information and the golfer's swing information described above. In addition, golf course information can be used.

For example, smart devices can infer a golfer's overall exercise volume, such as the number of strokes, steps, and slope movement, through the golfer's movements on the golf course.

In other words, the time the golfer rides the cart, the general walking time and number of steps, the time spent going up and down the slope, and the number of steps are added up to infer the calories consumed during the play. In conjunction with the golfer's weight, the amount of calories consumed is predicted using common knowledge, and through general analysis, it is used as health information that golfers obtain through a round.

In addition, the smart device (30) continuously monitors the posture of the golfer, and when the golfer falls (determined by an abnormality in the (p, t) signal) or shows abnormal signs of not moving for a considerable period of time (determined by an abnormality in the (x, y, z, t) signal), it sends emergency feedback, such as strong vibration, to the sensors (10, 20) attached to the entire body so that the golfer can make a first effort to escape from the abnormal situation, and if such an attempt is not possible, a message can be sent to a preset contact.

Referring to FIG. 8, an embodiment of a method for distinguishing between a practice swing and a shot swing is described.

It is desirable for the smart device (30) to be able to distinguish between a practice swing and a shot swing. This is because, although a golfer welcomes feedback from the smart device (30) on a practice swing during a practice swing, he or she may not want it during an actual shot swing because it may be bothersome.

The golfer moves to the ball and stops to make a shot (s301).

A golfer may take practice swings near the ball or at a distance (s302). The practice swings may be repeated multiple times.

In order for a golfer to actually hit the ball, the golf club must be positioned behind the ball (aka sole). To do this, the feet are moved from the position they were in during the practice swing (s303).

In this state, unlike the practice swing, the entire body stops moving for a relatively long time (s304). This is because most golfers are more careful with the shot swing that actually hits the ball than with the light practice swing. After that, the backswing of the shot swing is also usually done more slowly than the practice swing (s305). After that, the actual shot swing is performed (s306). After that, since the ball has been hit, the movement is moved to another location (s307).

The (x, y, z, t) and (p, t) data sent by the sensors (10, 20) for the above movements can be grouped by movement, and this can be used to distinguish between a practice swing and a shot swing. In particular, what is important when distinguishing between a practice swing and a shot swing is the stop phase (s304) and the slow backswing (s305).

Even if the golfer makes an OB or receives a mulligan and takes another shot from the same spot, the golfer is more careful about the shot swing, so the smart device (30) can infer that the golfer is taking another shot from the same spot by checking the movement phase (s307) and observing the entire body stationary phase (s304) and the backswing phase (s305).

Referring to Fig. 9, a method for restoring a golf round is described.

The method of replaying a golf round is to graphically reproduce the content of play on a golf course through the display of a smart device (30) to perform the function of replaying a golf round.

The smart device (30) contains data on the movements performed by the golfer throughout the entire golf round (walking, preparatory movements, setup, practice swing, shot swing, and movements after the shot swing), so by reproducing this graphically, it is possible to review reflection points on shots or strategic reflection points during the golf round and promote future development.

When a golfer plays a round of golf using the golf lesson method of the field mode described above on a golf course, the content of the play performed by the golfer on the golf course can be recorded on a smart device (30) and/or a server (50). In addition, map data of the golf course can be stored on the smart device (30).

Therefore, if a golfer wants, he or she can use this information to replay his or her play in graphic form. In other words, a golfer can replay his or her play using a smart device after or during a round of golf on the golf course.

Here, play may include swing motions, movement motions, etc. performed by a golfer on a golf course. In addition, one's play may be reproduced using the skeletal model or avatar described above.

Explains in detail how to review a golf round.

The golf round related information described above is stored in the smart device (30) (s205). Although the expression 'storage' is used here for convenience of explanation, it also includes receiving and using information in real time from wearable sensors, servers, etc. when executing the golf round review method.

The smart device (30) can select a golf round that the golfer wishes to repeat. (s210)

The smart device (30) generates graphic data using various information stored in the selected golf round. (s220)

Various information may include data on the golfer's play generated during the corresponding golf round, such as swing data, swing information, and movement information. In addition, various information may include data on the golf course where the corresponding golf round was performed, such as map data.

Smart devices can use information about a golfer's play to create graphic data about the golfer's play. Smart devices can also use information about a golf course to create graphic data about the golf course. Graphic data about a golfer's play can be expressed in the form of a skeletal model or avatar.

The smart device (30) can output graphic data generated about the golf course and the golfer's play (s230). (Details will be described later)

Meanwhile, the golf round replay method may additionally include executing a virtual golf round (s240). The virtual golf round is not something that the golfer actually played, but rather something in which the golfer can change his swing information to create a virtual play and output it. (Details will be described later.)

Each step is explained in detail.

First, the step of generating graphic data (s220) and the step of outputting graphic data (s230) are described. For convenience of explanation, the step of generating graphic data (s220) and the step of outputting graphic data (s230) are expressed separately, but the two steps are not necessarily separated and may be performed simultaneously.

A golfer's graphic data is data used to reproduce the golfer's play.

The most important part of a golfer's play is the practice swing and shot swing. The practice swing and shot swing are saved as swing data or swing information.

As described above, the swing data is in the form of (x,y,z,t) and (p,t). The swing data can be stored in a sensor, a smart device, or a server. In order to reproduce the swing, the data must be quickly transferred to the graphics engine (hardware and software for creating graphics) in the smart device (30). Therefore, it is preferable that the swing data be stored in the memory of the smart device.

The graphics engine uses (x,y,z,t) data to generate the movement trajectory and velocity of the corresponding body part in space. In other words, the graphics engine uses (x,y,z,t) to graphically create the process of how the points of the head, shoulders, hips, hands, and both feet change over time in three-dimensional space to complete the swing.

The smart device (30) can output the golfer's swing with changes in these points. That is, the swing can be output using a skeletal model. However, the skeletal model may be insufficient as a form to be shown to a person. Therefore, the graphic engine can render the body shape including these points and display it in the form of an avatar.

That is, one point of the head is replaced with the face of an avatar that resembles oneself, one point of the shoulder and pelvis is created by analogy with another point of the shoulder and pelvis at a symmetrical location with the head as the central axis, and a body connecting them is rendered.

As the shoulders and pelvis rotate or move with the swing, another shoulder and pelvis are created at a symmetrical location based on the waist center line connecting the centers of the head and both feet, and each of the two points above can move.

For right-handed people, gloves are usually worn on the left hand, so the left arm is rendered according to the movement of the left hand, and the right arm and right hand are rendered differently depending on the position of the left hand.

That is, during the setup, both arms are extended and the left and right hands are overlapped and rendered, but when going to the top of the backswing, the right arm should be naturally bent. When going to the top of the forward swing, the right arm should be naturally extended. The legs are rendered naturally by connecting the pelvis and both feet. When the pelvis rotates, one of the legs naturally bends or extends, so this is also included in the rendering.

As described above, the head, shoulders, pelvis, and hands are exemplified as the basic attachment sites of the body sensor (10). Therefore, since there is no separate sensor attached to the legs, it is not possible to render accurately in this case, but even if it is rendered to the extent of completing the general form of the swing, the swing can be replayed and the play content can be reviewed after a golf round.

Of course, golfers for whom leg movement is important can install a separate body sensor (10) on a part of the leg, such as the knee, and play. In this case, the graphic engine can also accurately render leg movement.

In other words, the wearable sensor (10, 20) can be flexibly installed according to the golfer's purpose.

For example, in a situation where one wants to play on a golf course by reducing the number of sensors as much as possible to make the game easier, or in a case where the only problem is the golfer's hand movement during the backswing and correction of this is sufficient, one body sensor (10) worn on the hand or wrist (including GPS) can be used to record the swing trajectory and tempo, and after playing, the avatar's swing can be replayed using only this data.

However, if you are a considerably careful golfer and want to analyze your swing form in detail, you can install multiple sensors (10, 20) in addition to the basic attachment points to reproduce your swing form in more detail. The graphic engine can be implemented to flexibly render the avatar's swing according to the number of sensors (10, 20) to suit the golfer's purpose.

Meanwhile, the graphics engine can determine where the avatar is located from GPS data (X, Y) and render the golf course scenery using the map data of the golf course.

Afterwards, by combining and rendering the avatar's swing described above, you can complete the reconstruction of the shot you took at that location on the golf course.

The rendering of a swing basically includes all three-dimensional information about the space and the swing performed in that space, so the observer can change the perspective and review the swing he performed in detail and the goal at that time, and analyze the strategy and result of the shot at that time.

Meanwhile, the information secured by the smart device (30) is information on body movements including swings and does not include information on the actual flight of the ball. However, the shape in which the ball was hit and moved can be inferred by using changes in the shot swing and the golfer's position.

For example, if the location of the second shot after the tee shot is located on the fairway at a considerable distance from the tee shot point, it can be inferred that the ball was hit well without side spin and was a nice shot. However, if the location of the second shot is located at a very short distance from the tee shot point, it is highly likely that it was a ground ball or a high-flying ball.

Also, considering the change in position between shots and the type of club that was probably used at the time, the actual ball speed can be inferred. The inference of the ball speed is only an inference and does not exactly reproduce the shot that was seen at the time, but can be a means of adding a kind of fun element.

The important thing in the present invention is to provide a method to accurately manage and teach the entire swing, including the practice swing and the shot swing, so as to produce a good shot as a result. In other words, in the review mode, the swing performed by the golfer is accurately reviewed and used as the basis for learning. Therefore, the shape of the ball that moved between swings is not very important, but is inserted as a graphic to add a fun element in the review mode, and may differ from the actual one.

By replaying the play, golfers can reflect on the swings and strategies they have implemented or improve their skills with the help of smart devices.

Next, the virtual golf round execution step (s240) is described.

Running a virtual golf round allows golfers to change their swing information to create a virtual play and output it.

In other words, you can virtually change the content of a golf round at a certain point by changing the shot at that point in time to a different swing than the one you actually performed.

For example, you may have had a hard time during an actual golf round due to a bad trouble shot. In this case, you can have fun and learn by correcting the swing taught by the smart device (30) and then swinging to see the different results. You can then try the changed swing through your avatar in a virtual space and mix it with the graphics of the golf course.

For example, if all sensors (10, 20) are attached to the body during, immediately after, or even after a round of golf, the content of the golf round can be changed by substituting one's current swing (hereinafter, for convenience, 'virtual swing') for a specific shot at a place and time of one's choice during the round of golf.

In other words, you can effectively have fun and learn at the same time by changing your current swing to the swing of your avatar in the virtual golf course space, similar to the experience of the metaverse.

Replacement with a virtual swing is done by replacing the swing data in the existing memory, i.e., (x,y,z,t) data and (p,t) data, with (x,y,z,t) and (p,t) of the virtual swing data and changing them in the graphics engine as described above.

In other words, the swing performed by the avatar is changed to a virtual swing, for example, the latest swing data obtained after correcting the swing, or the swing data performed while wearing the sensor (which may be different from the corrected swing). By doing so, you can have fun and learn at the same time by creating situations that can be different.

In addition, the smart device (30) can periodically (for example, when one practice session is over or one golf round is over) store data on the server of the company providing the service via the network. The period at which the smart device (30) transmits data to the server (50) may vary depending on mechanical circumstances such as the storage capacity or backup of the smart device (30) or external circumstances such as the service policy between the company and the golfer.

The server (50) accumulates and stores almost all data related to the play of a specific golfer, including swing data of a specific golfer and score data recorded on a golf course, so it knows the changes in the swing of the golfer best. Based on this, in addition to swing lessons by a smart device (30), offline services such as recommendations of the most suitable professional offline or recommendations of golf clubs suitable for the current situation when the golfer gets older are possible.

In addition, the server (50) stores and analyzes data transmitted from a number of golfer members who have registered for this service in the same way, so that it can utilize these big data to derive similar data by group of golfers and provide better services. For example, information extracted from big data, such as commonalities of female golfers or commonalities of swings that change with age, can be used for the development of the golf industry, such as research on golf clubs or golf equipment.

Meanwhile, in each of the embodiments described above, at least one of the parts described in the other embodiments may be applied equally to the parts that are not described in each embodiment. In addition, unless they are conflicting with each other in the embodiments described above, even if there is no special mention, technical details described in one embodiment may be applied equally to the other embodiments.

As described above, the present invention has been described by way of examples and drawings, but these have been used to help understanding of the present invention. Therefore, the present invention is not limited to the examples described above. Those with ordinary knowledge in the technical field to which the present invention pertains can make various modifications and variations from this description, and such modifications and variations are also within the scope of the present invention.

In this specification, body sensors, shoe sensors, central processing units, inertial sensors, alarm units, pressure sensors, communication units, etc. are conveniently classified by functional aspects for convenience of explanation, and are not necessarily separate concepts in terms of hardware and software. They can be implemented as one software or as multiple modules in one software, and can also be implemented as an appropriate combination of hardware or software.

10: Body sensor 20: Shoe sensor
30: Smart Device 40: Server

Claims (19)

A first step of obtaining first swing data including at least one of spatial coordinates of the points and changes (x, y, z, t) of the spatial coordinates on a time axis and second swing data including changes (p, t) of pressure on the time axis of the points from wearable sensors attached to a plurality of points of the golfer's body;
A second step of generating swing information including at least one of the first swing data and the second swing data;
A third step of comparing the swing information with the predefined reference swing information corresponding to the skeletal model; and
A golf lesson method using a wearable device, comprising a fourth step of transmitting a signal to the wearable sensor to generate a notification at a corresponding point when the above swing information deviates from the above reference swing information by a predetermined range. In the first paragraph,
The wearable sensor includes an inertial sensor and a pressure sensor, and the first swing data and the second swing data are obtained from the inertial sensor and the pressure sensor.
A golf lesson method in which the above skeleton model is defined using nodes which are locations where the above inertial sensors are attached. In the second paragraph,
The above skeletal model includes at least one of a static skeletal model and a dynamic skeletal model,
A golf lesson method that uses at least one of the upright posture, setup posture, and backswing top posture of the golfer as the origin coordinate to determine the change in the coordinates of the nodes of the above skeletal model. In any one of claims 1 to 3,
A golf lesson method comprising a fifth step of obtaining lie information of a point where the golfer is located, and defining the reference swing information by further reflecting the lie information. In paragraph 4,
A golf lesson method in which the above-mentioned lie information is determined using map data of the golf course on which the golfer is playing and location information of the golfer. In paragraph 4,
A golf lesson method in which the above-mentioned lie information is determined using the first swing data and the second swing data. In paragraph 4,
A golf lesson method wherein the target point information intended by the golfer is obtained, and the reference swing information is defined to further reflect the target point information. In Article 7,
A golf lesson method in which the above target point is determined using map data of a golf course on which the golfer is playing and location information of the golfer. In Article 7,
A golf lesson method including a sixth step of obtaining motion information on various motions of the golfer by using the location information of the golfer, the first swing data, and the second swing data. In Article 9,
A golf lesson method for selectively providing a golf lesson to a golfer by judging the golfer's practice swing motion and shot swing motion from the above motion information. In Article 9,
A golf lesson method using the above motion information to perform at least one of transmitting a signal to the wearable sensor to generate a notification and transmitting a message to a preset contact when the golfer is determined to be motionless for a certain period of time. In any one of claims 1 to 3,
A golf lesson method including a seventh step in which the reference swing information is changed by the golfer to generate individual reference swing information. In Article 12,
A golf lesson method in which the above-mentioned individual reference swing information is generated by changing the first swing data and the second swing data in the above-mentioned skeletal model. A first step in which information obtained from any one of the golf lesson methods of clauses 1 to 13 is stored as golf round information;
Step 2: Select the golf round you wish to play;
A third step of generating golf course graphic data and golfer graphic data using golf round information corresponding to the selected golf round;
A golf round reconstruction method comprising a fourth step of outputting the golf course graphic data and the golfer graphic data. In Article 14,
A method for reconstructing a golf round, wherein the above golf round information includes golfer's location information, first swing data, second swing data, and golf course information. In Article 15,
A method for reconstructing a golf round, wherein in the third step, the golfer graphic data is generated as one of a golfer's skeletal model and a golfer's avatar. In any one of Articles 14 to 16,
A fourth step is included in which the first swing data and the second swing data are changed by the golfer, thereby generating the changed first swing data and the changed second swing data.
A golf round reconstruction method in which the third step and the fourth step are performed using the changed first swing data and the changed second swing data. In Article 17,
A golf round reconstruction method in which, in the fourth step, the changed first swing data and the changed second swing data are generated using the skeleton model. In Article 17,
A golf round reconstruction method in which, in the fourth step, the first swing data and the second swing data obtained as a result of the golfer swinging while wearing the wearable device are used to generate the changed first swing data and the changed second swing data.

Images