WO2024058322A1

WO2024058322A1 - Method and device for providing putting information

Info

Publication number: WO2024058322A1
Application number: PCT/KR2023/000622
Authority: WO
Inventors: 신동진; 한혁; 김유승; 이광준
Original assignee: 주식회사 카카오브이엑스
Priority date: 2022-09-13
Filing date: 2023-01-13
Publication date: 2024-03-21
Also published as: KR20240036741A

Abstract

The present disclosure is to provide a golf putting assistance method for transmitting putting information. A golf putting assistance method may comprise the steps of: receiving, from a user terminal, a putting information request including location coordinates of the user terminal; on the basis of the putting information request, transmitting, to an inference server, an inference request message including location information of a user determined on the basis of the location coordinates of the user terminal; on the basis of the inference request message, receiving an inference result including putting direction information and putting strength information from the inference server; and transmitting putting information including the putting direction information and the putting strength information to the user terminal, wherein the inference result including the putting direction information and the putting strength information is generated using a learning model learned using an artificial intelligence algorithm by the inference server on the basis of the location information of the user and terrain information of a green area.

Description

Method and device for providing putting information

This disclosure relates to a method and device for providing putting information. More specifically, the present disclosure relates to a method and device for generating and providing information to assist a user's putting by considering the user's location, terrain information, etc.

Golf is a ball sport in which players hit a ball at rest on a course with a club and put it into a designated hole, and compete based on the number of strokes taken up to that point. In other words, the person who hits the fewest balls wins the game. Golfers use different shaped golf clubs to hit golf balls at longer distances and putt on greens with various slopes. In particular, in the case of putting on the green, you must putt in the correct direction and with force considering the lie, slope, break on the green, presence of mountains or water, grass texture around the hole, etc., so you can play golf. It is also a very tricky and difficult part to do.

Accordingly, golfers must perform putting on the green based on experience, but inexperienced golfers find it difficult to determine the direction and strength of putting on the green or rely on their caddies to perform putting. However, when the dependence on the caddy is high, it is difficult for golfers to play golf on a no-caddie golf course without a caddy, and there is a problem in that the number of strokes to the hole is determined by the caddy's ability.

As such, there was a demand for a method and device for a golfer to determine the exact direction and strength of putting on the green during a golf round, but there was a problem that this could not be provided according to the conventional technology, and the present disclosure is intended to solve this problem. will be.

One embodiment of the present disclosure is to provide a method and device for providing putting information including putting direction and putting strength to a golfer.

Additionally, an embodiment of the present disclosure is intended to provide a method and device for collecting and utilizing topographical data of the green area of a golf course.

Additionally, an embodiment of the present disclosure is intended to provide a method and device for collecting the current location and status of a golfer.

Additionally, the present disclosure is intended to provide a method and device for providing putting information using artificial intelligence (AI) technology based on machine learning algorithms such as deep learning.

One embodiment of the present disclosure seeks to provide a method and device for providing putting assistance information including putting direction and putting strength to a golfer.

A golf putting assistance method according to an embodiment of the present disclosure includes receiving a putting information request including location coordinates of the user terminal from a user terminal; Based on the putting information request, transmitting an inference request message including user location information determined based on the location coordinates of the user terminal to an inference server; Based on the inference request message, receiving an inference result including putting direction information and putting strength information from the inference server; It may include transmitting putting information including the putting direction information and the putting intensity information to the user terminal. In addition, the inference result including the putting direction information and the putting intensity information is learned by the inference server using an artificial intelligence algorithm based on the user's location information and topographic information of the green area where the user is located. It can be created using a learning model.

In one embodiment, the terrain information of the green area includes information generated by mapping coordinate information corresponding to the green area and height information of the green area, and the height information of the green area includes a first sensor. It can be obtained using a first device installed on the ground and a second device including a flying vehicle including a second sensor.

In one embodiment, the learning model includes a reinforcement learning model including an action policy model and a valuation model, and the action policy model includes location information of the target point, putting start location information, and topographic information of the green area. , generating behavioral information regarding the putting direction and putting strength based on environmental information including at least one of ball characteristic information, user status information, or weather condition information, and the valuation model is configured to include the environmental information and the putting strength. Rewards can be determined based on behavioral information.

In one embodiment, transmitting the inference request message includes determining whether the user is located in the green area based on a putting information request including location coordinates of the user; And when it is determined that the user is located in the green area, it may include transmitting the inference request message including the location information of the user to the inference server.

In one embodiment, the golf putting assistance method may further include generating location information of the user by correcting the location coordinates of the user terminal using a Real-Time Kinematic (RTK) global positioning system (GPS). You can.

In one embodiment, the putting information including the putting direction information and the putting intensity information may include information to be displayed by the user terminal through AR (Augmented Reality).

A method of operating an inference server according to an embodiment of the present disclosure includes receiving an inference request message including user location information from a golf course management server; Obtaining topographic information of a green area where the user is located; Generating putting direction information and putting strength information using a learning model learned using an artificial intelligence algorithm, based on the user's location information and green area topography information; And it may include transmitting an inference result including the putting direction information and the putting intensity information to the golf course management server.

In one embodiment, the step of obtaining topographic information of the green area includes requesting topographic information of the green area from the golf course management server; And receiving topographic information of the green area that maps coordinate information corresponding to the green area and height information of the green area from the golf course management server, wherein the height information of the green area is obtained by using a first sensor. It can be obtained using a second aircraft device including a first device installed on the ground and a second sensor.

In one embodiment, the obtaining the topographic information of the green area includes transmitting a request for topographic information of the green area, including at least one of golf course information or location information of the user; And based on the request for terrain information of the green area, receiving terrain information of the green area generated by mapping coordinate information corresponding to the green area and height information of the green area, Height information may be obtained using a first device including a first sensor and a second device including a second sensor.

In one embodiment, the learning model includes a reinforcement learning model including a behavioral policy model and a valuation model, and the step of generating the putting direction information and the putting strength information includes putting by the behavioral policy model. Generating behavioral information regarding the putting direction and putting strength based on environmental information including at least one of starting position information, topography information of the green area, ball characteristic information, user status information, and weather condition information; And it may include generating a reward based on the environmental information and the behavioral information using the value evaluation model.

In one embodiment, the inference request message may be transmitted to the inference server when the golf course management server determines that the user is located in the green area.

In one embodiment, the user's location information may be generated by correcting the location coordinates of the user terminal using a Real-Time Kinematic (RTK) global positioning system (GPS).

In one embodiment, the putting direction information and the putting intensity information may be transmitted to a user terminal including an Augmented Reality (AR) device by the golf course management server.

One embodiment of the present disclosure includes: a memory storing one or more instructions; and at least one processor executing the one or more instructions stored in the memory, wherein the at least one processor executes the one or more instructions to request putting information including location coordinates of the user terminal from the user terminal. Receive and, based on the putting information request, transmit to the inference server an inference request message containing the user's location information determined based on the location coordinates of the user terminal, and based on the inference request message, from the inference server Receives an inference result including putting direction information and putting strength information, transmits putting information including the putting direction information and the putting strength information to the user terminal, and includes the putting direction information and the putting strength information. The inference result is generated using a learning model learned using an artificial intelligence algorithm by the inference server based on the user's location information and topographic information of the green area where the user is located. You can.

In one embodiment, the at least one processor obtains coordinate information corresponding to the green area and determines the height of the green area using a first device including a first sensor and a second device including a second sensor. Obtain information, wherein the first device comprises a ground-mounted device and the second device comprises an air vehicle; Topographic information of the green area can be generated by mapping coordinate information corresponding to the green area and height information of the green area, and the topographic information of the green area can be stored in a database.

In one embodiment, the at least one processor determines whether the user is located in the green area, based on a putting information request including location coordinates of the user, and determines that the user is located in the green area. If determined, the inference request message including the user's location information may be transmitted to the inference server.

In one embodiment, the at least one processor may generate location information about the user by correcting the location coordinates of the user terminal using a real-time kinematic (RTK) global positioning system (GPS).

One embodiment of the present disclosure includes: a memory storing one or more instructions; and at least one processor executing the one or more instructions stored in the memory, wherein the at least one processor receives an inference request message including the user's location information from the golf course management server by executing the one or more instructions. Obtain topographic information of the green area where the user is located, and obtain putting direction information and putting strength using a learning model learned using an artificial intelligence algorithm based on the user's location information and topographic information of the green area. An inference server may be provided that generates information and transmits an inference result including the putting direction information and the putting intensity information to the golf course management server.

In one embodiment, the at least one processor requests topographic information of the green area from the golf course management server, and maps coordinate information corresponding to the green area and height information of the green area from the golf course management server. Topographic information of the green area is received, and height information of the green area can be obtained using a first device installed on the ground including a first sensor and a second aircraft device including a second sensor.

In one embodiment, the at least one processor transmits a request for terrain information of the green area, including at least one of golf course information or location information of the user, and based on the request for terrain information of the green area, Receive topographic information of the green area generated by mapping coordinate information corresponding to the green area and height information of the green area, and the height information of the green area is received from a first device including a first sensor and a second sensor. It can be obtained using a second device including.

In one embodiment, the learning model includes a reinforcement learning model including a behavioral policy model and a valuation model, and the at least one processor, by the behavioral policy model, puts starting position information, the green area Generate behavioral information regarding putting direction and putting strength based on environmental information including at least one of terrain information, ball characteristic information, user status information, or weather condition information, and generate behavioral information regarding putting direction and putting strength by the valuation model, based on the environment. Rewards can be created based on information and the behavioral information.

One embodiment of the present disclosure includes a program stored in a recording medium to execute the method according to the embodiment of the present disclosure on a computer.

An embodiment of the present disclosure includes a computer-readable recording medium on which a program for executing a method according to an embodiment of the present disclosure on a computer is recorded.

An embodiment of the present disclosure includes a computer-readable recording medium that records a database used in an embodiment of the present disclosure.

According to one embodiment of the present disclosure, there is an effect of efficiently assisting the user's putting by providing putting information to the user terminal based on the location of the user terminal.

According to an embodiment of the present disclosure, there is an effect of effectively inferring putting information based on the location of the user terminal using artificial intelligence technology.

Additionally, according to an embodiment of the present disclosure, topographical information of the green area may be collected and stored in a database.

Additionally, an embodiment of the present disclosure can improve the accuracy of location information of user terminals, target points, etc.

1 is a diagram illustrating a method of providing inference results to a user terminal according to an embodiment of the present disclosure.

FIG. 2 is a diagram illustrating a method of providing inference results to a user terminal according to an embodiment of the present disclosure.

Figure 3 is a flowchart illustrating a method of transmitting putting information to a user terminal according to an embodiment of the present disclosure.

FIG. 4 is a flowchart illustrating a method of determining whether to perform inference based on whether a user terminal is located in a green area according to an embodiment of the present disclosure.

Figures 5(a) and 5(b) are diagrams illustrating a method of obtaining coordinate information of a green area according to an embodiment of the present disclosure.

FIG. 6 is a diagram illustrating a method of obtaining topographic information of a green area using a plurality of sensors according to an embodiment of the present disclosure.

FIG. 7 is a diagram illustrating a height map of a green area according to an embodiment of the present disclosure.

FIG. 8 is a diagram illustrating a method of determining putting information using an artificial intelligence algorithm according to an embodiment of the present disclosure.

Figure 9 is a diagram illustrating a method of training a learning model using an artificial intelligence algorithm according to an embodiment of the present disclosure.

FIG. 10 is a diagram illustrating how a learning model performs reinforcement learning according to an embodiment of the present disclosure.

Figure 11 is a diagram showing the input and output structure of a reinforcement learning model according to an embodiment of the present disclosure.

FIG. 12 is a diagram illustrating a method for obtaining the location of a user terminal according to an embodiment of the present disclosure.

FIG. 13 is a diagram illustrating a method of displaying putting information using an augmented reality (AR) device according to an embodiment of the present disclosure.

Figure 14 is a block diagram of a golf course management server according to an embodiment of the present disclosure.

Figure 15 is a block diagram of an inference server according to an embodiment of the present disclosure.

In order to clarify the technical idea of the present disclosure, embodiments of the present disclosure will be described in detail with reference to the attached drawings. In describing the present disclosure, if it is determined that a detailed description of a related known function or component may unnecessarily obscure the gist of the present disclosure, the detailed description will be omitted. Components having substantially the same functional configuration among the drawings are given the same reference numbers and symbols as much as possible, even if they are shown in different drawings. For convenience of explanation, if necessary, the device and method should be described together. Each operation of the present disclosure does not necessarily have to be performed in the order described, and may be performed in parallel, selectively, or individually.

The terms used in the embodiments of the present disclosure have selected general terms that are currently widely used as much as possible while considering the function of the present disclosure, but this may vary depending on the intention or precedent of a person working in the art, the emergence of new technology, etc. . In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the relevant embodiment. Therefore, the terms used in this specification should not be defined simply as the names of the terms, but should be defined based on the meaning of the term and the overall content of the present disclosure.

Throughout this disclosure, singular expressions may include plural expressions, unless the context clearly dictates otherwise. Terms such as "include" or "have" are intended to designate the presence of a feature, number, step, operation, component, part, or combination thereof, but not one or more other features, numbers, steps, operations, or composition. It should be understood that this does not exclude in advance the possibility of the presence or addition of elements, parts, or combinations thereof. In other words, when it is said that a part "includes" a certain element throughout the present disclosure, this means that other elements may be further included rather than excluding other elements, unless specifically stated to the contrary.

An expression such as "at least one" modifies the entire list of elements, not the elements of the list individually. For example, “at least one of A, B, and C” and “at least one of A, B, or C” means only A, only B, only C, both A and B, both B and C, and A and C All refers to all of A, B, and C, or a combination thereof.

In addition, terms such as "...unit" and "...module" described in the present disclosure mean a unit that processes at least one function or operation, which is implemented in hardware or software or by a combination of hardware and software. It can be implemented.

Throughout the present disclosure, when a part is said to be “connected” to another part, this includes not only the case where it is “directly connected,” but also the case where it is “electrically connected” with another element in between. do. Additionally, when a part "includes" a certain component, this means that it may further include other components rather than excluding other components, unless specifically stated to the contrary.

The expression “configured to” used throughout the present disclosure may be used, depending on the context, for example, “suitable for,” “having the capacity to.” )", "designed to", "adapted to", "made to", or "capable of". . The term “configured (or set to)” may not necessarily mean “specifically designed to” in hardware. Instead, in some contexts, the expression “system configured to” may mean that the system is “capable of” in conjunction with other devices or components. For example, the phrase "processor configured (or set) to perform A, B, and C" refers to a processor dedicated to performing the operations (e.g., an embedded processor), or by executing one or more software programs stored in memory. It may refer to a general-purpose processor (e.g., CPU or application processor) that can perform the corresponding operations.

Throughout the present disclosure, a terminal or terminal may include a UE (User Equipment), MS (Mobile Station), a cellular phone, a smartphone, a computer, a smart watch, a Bluetooth earphone, or a multimedia system capable of performing a communication function. Of course, it is not limited to the above example.

An embodiment of the present disclosure seeks to provide a golf course management server, an inference server, a user terminal, and a method of operating these for providing putting information such as putting strength and putting direction to a user terminal. In one embodiment, the golf course management server may be a golf course ERP (Enterprise Resource Planning) server. In another embodiment, the golf course management server may be implemented as a separate server. Additionally, the inference server may be a server that determines putting information using an artificial intelligence model. In the method of operating the inference server according to the present disclosure, a method for inferring or predicting putting information including putting direction and putting strength, includes location information of the user terminal, topography information of the green area, location information of the ball, and user information. An artificial intelligence model can be used to recommend putting direction and putting strength using status information, type of ball, weather, etc. The processor of the inference server can perform a preprocessing process on the data and convert it into a form suitable for use as input to an artificial intelligence model. Artificial intelligence models can be created through learning. Here, being created through learning means that the basic artificial intelligence model is learned using a large number of learning data by a learning algorithm, thereby creating a predefined operation rule or artificial intelligence model set to perform the desired characteristics (or purpose). It means burden. Inference prediction is a technology that judges information and makes logical inferences and predictions, including knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Includes.

Referring to FIG. 1, the user terminal 103 may transmit a putting information request 120 to the golf course management server 107. In one embodiment, the putting information request 120 may include the location coordinates of the user terminal. The user terminal 103 includes a Global Positioning System (GPS) device, and the user's location coordinates may include coordinate information of the user terminal 103 collected through the GPS device. Additionally, the user terminal 103 may include a barometric pressure sensor, an acceleration sensor, a gyro sensor, a geomagnetic sensor, etc. Accordingly, the putting information request 120 may further include the degree to which the user terminal 103 is tilted, the amount of change in the tilt of the user terminal 103, and compass information. For example, when the putting information request signal includes GPS information acquired by a GPS device and compass information acquired by a geomagnetic sensor, the golf course management server 107 provides latitude and longitude information where the user terminal is located and the user terminal. Direction information for (103) can be obtained. Additionally, the putting information request signal may further include ball position information, ball characteristic information, etc. The user terminal 103 may receive ball location information from an external device and transmit it to the golf course management server 107.

In one embodiment, the golf course management server 107 may determine the user's location information based on the user's location coordinates included in the putting information request 120. For example, the golf course management server 107 may determine the user's location information by performing correction using RTK GPS on the user's location coordinates. Additionally, the user's location information may include orientation information of the user terminal 103.

In one embodiment, the golf course management server 107 may identify whether the user is located in the green area 101 based on the user's location information or the location coordinates of the user terminal. If the golf course management server 107 identifies that the user is located in the green area 101, it transmits an inference request message 130, and if it identifies that the user is not located in the green area 101, the golf course management server 107 performs the following operation. A message may not be performed or a message that putting information cannot be provided may be transmitted to the user terminal 103. For example, the message that putting information cannot be provided may include information that it is not located in the area drawn by the user. Details regarding this will be described later with reference to FIG. 4 .

The golf course management server 107 may transmit an inference request message 130 to the inference server 109. In one embodiment, the inference request message 130 may include the user's location information. In addition, the golf course management server 107 obtains ball location information, golf course current weather condition information, ball characteristic information, and hole cup 105 location information, and sends the inference request message 130 among these information. It may be transmitted including at least part of it. For example, current weather condition information on a golf course may include current weather, current temperature, humidity, wind strength, etc. Additionally, the user terminal 103 may receive user input about a ball profile, such as the type of ball, brand, etc., and transmit it to the golf course management server 107. The golf course management server 107 identifies the characteristics of the ball, including the surface shell material of the golf ball and the repulsion force of the ball, from the profile of the ball, and provides ball characteristic information to the inference server 109 based on the identified ball characteristics. can be transmitted.

Based on the inference request message 130, the inference server 109 sends a request for topographic information of the green area including at least one of golf course information 140 or user location information to the device including the topographic database 111. Can be transmitted. Golf course information 140 may be determined based on the location of the user terminal 103 or the identifier of the golf course management server 107. In one embodiment, the golf course information 140 includes an identifier of the golf course management server 107, an identifier for the golf course where the user terminal 103 is located, location information about the golf course where the user terminal 103 is located, and the user terminal 103. It may include at least one of an identifier for the green area 101 where the user terminal 103 is located, or location information about the green area 101 where the user terminal 103 is located.

The inference server 109 may obtain terrain information 150 of the green area from the terrain database 111. In one embodiment, the device including the terrain database 111 may transmit the terrain information 150 of the corresponding green area to the inference server 109 based on the received green area terrain information request. In one embodiment, the topographic information of the green area may be a height map generated by mapping coordinate information corresponding to the green area and height information of the green area.

The inference server 109 produces an inference result 160 including putting direction information, putting strength information, etc. based on the user's location information, green area topographic information 150, and factors that may affect the trajectory of the ball. It can be obtained. In one embodiment, the inference server 109 may obtain the inference result 160 using a learning model learned using an artificial intelligence algorithm. Learning models may include reinforcement learning models, including action policy models and valuation models.

In one embodiment, the inference server 109 includes location information of the target point including the location of the hole cup 105, location information of the user, topography information of the green area, ball characteristic information, user status information, and weather conditions. By inputting information, etc., an inference result 160 for hole-in determined by the learning model can be obtained. In one embodiment, the terrain information of the green area may include slope information of the green area, friction information of the green area, grass condition information, information about whether mountains or water exist in or around the green area, grass texture information, etc. You can. Additionally, the ball characteristic information may include ball friction information, ball surface material information, etc., and the user's condition information may include the type of golf club used by the user, user ability information, user's condition information, etc. You can.

The inference server 109 may transmit the inference result 160 including putting direction information and putting strength information to the golf course management server 107.

The golf course management server 107 may transmit putting information 170 to the user terminal 103. In one embodiment, the putting information may include information about the recommended putting direction, information about the recommended putting strength, a recommended putting path, a recommended putting angle, etc.

Based on the received putting information 170, the user terminal 103 outputs putting information including putting direction, putting strength, etc. through an audio device such as a speaker, in-ear, etc., or installed on the user's mobile phone, tablet PC, etc. Putting information can be displayed using an application, or putting information can be displayed through an AR device such as AR glasses. However, this is only an example and is not limited to this, and the user terminal 103 may provide putting information to the user through various devices.

In one embodiment, the user terminal 103 includes a smartphone, a smartwatch, an AR device, etc., and the user terminal requesting putting information and the user terminal receiving the putting information may be the same terminal or different terminals. For example, the user terminal requesting putting information may be a smartwatch, and the user terminal receiving putting information may be an AR device.

Meanwhile, although FIG. 1 illustrates a method of determining and transmitting putting information according to an embodiment of the present disclosure, various modifications may be made to FIG. 1 of course. As an example, although successive operations are shown in FIG. 1, it goes without saying that the various operations in FIG. 1 may overlap, occur in parallel, occur in a different order, or occur multiple times.

Referring to FIG. 2, the user terminal 103 may transmit a putting information request 220 to the golf course management server 107. In one embodiment, the putting information request 220 may correspond to the putting information request 120 described above with reference to FIG. 1 .

In one embodiment, the golf course management server 107 may identify whether the user is located in the green area 101 based on the user's location information included in the putting information request 220. If the golf course management server 107 identifies that the user is located in the green area 101, it transmits golf course information 230, and if it identifies that the user is not located in the green area 101, it performs the following operation. Alternatively, a message that putting information cannot be provided may be transmitted to the user terminal 103. For example, the message that putting information cannot be provided may include information that it is not located in the area drawn by the user. Details regarding this will be described later with reference to FIG. 4 .

The golf course management server 107 may obtain topographic information 240 of the green area corresponding to the golf course information 230 from the topographic database 111 based on the putting information request 220. In one embodiment, the topographic information 240 of the green area corresponding to the golf course information 230 may be information collected by the golf course management server 107 and stored in the topographic database 111. That is, in the terrain database 111, the golf course management server 107 of each golf course generates terrain information of the green area and delivers it to a device including the terrain database 111, and the device controls the green area corresponding to each golf course. The terrain information of the area can be stored in the terrain database 111. In one embodiment, the terrain database 111 may be included in the golf course management server 107, an external server, or in the inference server 109. The golf course information 230 and the green area terrain information 240 may respectively correspond to the golf course information 140 and the green area terrain information 150 described above with reference to FIG. 1 .

The golf course management server 107 sends an inference request message 250 containing the user's location information included in the putting information request 220 and the topographic information 240 of the green area obtained from the topographic database 111 to the inference server. It can be sent to (109). The inference server 109 may obtain an inference result 260 including putting direction information, putting strength information, etc. based on the user's location information, green area topography information 240, etc. In one embodiment, the inference server 109 may obtain the inference result 260 using a learning model learned using an artificial intelligence algorithm. Learning models may include reinforcement learning models, including action policy models and valuation models.

In one embodiment, the inference server 109 includes location information of the target point including the location of the hole 105, location information of the user, topography information of the green area, ball characteristic information, user status information, and weather condition information. The inference result 260 determined by the learning model can be obtained by using the etc. as input. In one embodiment, the topographic information of the green area may include slope information of the green area, friction information of the green area, grass condition information, information about whether mountains or water exist in or around the green area, grass texture information, etc. You can. Additionally, the ball characteristic information may include ball friction information, ball surface material information, etc., and the user's condition information may include the type of golf club used by the user, user ability information, user's condition information, etc. You can.

The inference server 109 may transmit the inference result 260 including putting direction information and putting strength information to the golf course management server 107. The inference result 260 may correspond to the inference result 160 described above with reference to FIG. 1 .

The golf course management server 107 may transmit putting information 270 to the user terminal 103. In one embodiment, the putting information 270 may correspond to the putting information 170 described above with reference to FIG. 1 .

Meanwhile, although FIG. 2 illustrates a method for determining and transmitting putting information according to an embodiment of the present disclosure, various modifications may be made to FIG. 2 of course. As an example, although successive operations are shown in FIG. 2, it goes without saying that the various operations in FIG. 2 may overlap, occur in parallel, occur in a different order, or occur multiple times.

Referring to FIG. 3, in step 310, the golf course management server may receive a putting information request including the user's location information from the user terminal. In one embodiment, the user's location information may be obtained using a Real-Time Kinematic (RTK) global positioning system (GPS). In other words, the user terminal identifies the user's location based on at least two reference coordinates around the green area within the golf course, and sends a putting information request including user location information to the golf course management server based on the identified user location. Can be sent. The method of acquiring coordinates using RTK GPS will be described in more detail later in the description related to FIG. 12.

In one embodiment, the user terminal may include a smartphone equipped with a GPS function. Additionally, the user terminal may include a wearable device such as a smart watch or earphones equipped with a GPS function. For example, wearable devices can be implemented in the form of wristwatches, glasses, earrings, necklaces, earphones, earring-type accessories, shoes, rings, clothes, belts and helmets. However, the present invention is not limited to this, and the wearable device may be implemented in a form that is directly attachable and detachable from the user's body. For example, a wearable device may be implemented in the form of a patch and may be attached or detached to the user's body in an adhesive or non-adhesive manner. Additionally, a wearable device may be implemented in a form that is inserted inside the user's body. For example, wearable devices can be implemented in the form of epidermal electronics (E-Skin) or electronic tattoos (E-Tattoo) and inserted into the epidermis or interior of the body through medical procedures.

In one embodiment, the wearable device identifies the user's location using the GPS function and directly transmits the user's location information to the golf course management server based on the identified user's location, or transmits the user's location based on the identified location information. Information may be transmitted to a user terminal such as a smartphone, and the user terminal may transmit the user's location information to the golf course management server. The user's location information may include latitude and longitude information and direction information of the user terminal.

In one embodiment, the putting information request may include user's status information, ball profile information, etc. along with the user's location information. The user's status information may include, for example, gender, age, height, weight, whether the user is right-handed or left-handed, usual putting habits, putting ability, etc. Additionally, the ball profile information may include the ball brand name, ball identification information, ball material information, etc.

In step 330, the golf course management server may transmit an inference request message including the user's location information to the inference server based on the putting information request. In one embodiment, the golf course management server may determine whether the user is located in the green area based on the user's location information included in the putting information request. If the golf course management server determines that the user is located in the green area, an inference request message containing the user's location information may be transmitted to the inference server. Alternatively, if the golf course management server determines that the user is not located in the green area, the golf course management server may transmit information that putting information cannot be provided to the user terminal.

In one embodiment, the inference request message may include location information of the target point, terrain information of the green area, ball characteristic information, user status information, weather condition information, etc. along with the user's location information. The user's location information and the target point's location information may be information obtained using RTK GPS to identify the exact location.

In step 350, the golf course management server may receive an inference result including putting direction information and putting strength information from the inference server based on the inference request message. In one embodiment, the inference server may generate an inference result using a learning model learned using an artificial intelligence algorithm based on the user's location information and the topographic information of the green area.

In one embodiment, the terrain information of the green area may include information generated by mapping coordinate information corresponding to the green area and height information of the green area. Here, the height information of the green area may include a height map of the green area obtained using a device installed on the ground including a first sensor, a flying vehicle including a second sensor, etc. For example, the first sensor may be a stand-type LiDAR (Light Detection and Ranging) sensor, and the second sensor may be a LiDAR sensor mounted on a drone.

In one embodiment, the learning model of the inference server may include a reinforcement learning model including a behavioral policy model and a valuation model. In addition, the action policy model includes at least one of location information of the target point (e.g., the location of the hole cup), location information of the user, topography information of the green area, ball characteristic information, user status information, or weather condition information. Based on the state information, an inference result including putting direction information, putting strength information, etc. may be output. Additionally, the valuation model can determine rewards based on state information and inference results.

In step 370, the golf course management server may transmit putting information including putting direction information and putting strength information to the user terminal. In one embodiment, based on the received putting information, the user terminal provides putting direction information (e.g., a club to the left of the hole cup) and putting strength information (e.g., a force aimed at 8.5 m) through sound devices such as speakers and earphones. ) can be output. Alternatively, based on the received putting information, the user terminal may output putting direction information and putting strength information using augmented reality through a wearable device such as AR glasses.

Referring to FIG. 4, in step 410, the golf course management server may receive a putting information request message from the user terminal. In one embodiment, the putting information request message may include the user's location information.

In step 430, the golf course management server may identify whether the user's location is on the green based on the received user's location information. If the golf course management server identifies that the user's location is on the green, step 450 may be performed, and if not, step 490 may be performed.

That is, if the golf course management server identifies that the user's location is on the green, in step 450, the golf course management server sends an inference request to the inference server, and the inference server uses an AI (Artificial Intelligence) model to determine the optimal direction and optimal You can decide the century. Additionally, in step 470, the golf course management server may receive information about the optimal direction and optimal intensity from the inference server and transmit it to the user terminal.

Alternatively, if the golf course management server identifies that the user's location is not on the green, the golf course management server may transmit an inference impossible message to the user terminal in step 490. In one embodiment, when the user's location is not on the green, the golf course management server may transmit a message guiding other services that can be provided according to the user's current location. For example, if the user's location is a fairway, the remaining distance and direction from the current location to the hole cup of the corresponding hole can be provided.

Topographic information of the green area may be generated by a topographic data collection server. In one embodiment, the terrain data collection server may be included in the golf course management server or inference server, or may be a third server.

Referring to FIG. 5, the terrain data collection server may obtain coordinate information corresponding to the green area to generate terrain information of the green area. In one embodiment, the terrain data collection server may obtain the latitude and longitude of a point on the boundary of the green area, as shown in FIG. 5(a), to specify the green area. In addition, the terrain data collection server repeats the process of FIG. 5(a) for obtaining latitude and longitude for other points on the boundary of the green area, and the processor of FIG. 5(b) obtains a plurality of latitude and longitude corresponding to the boundary. can be performed. The topographic data collection server generates coordinate information of the green area using the plurality of latitudes and longitudes obtained in this way, and maps the coordinate information of the green area with the height map of the green area, which will be described later with reference to FIG. 6, to generate topographic information of the green area. You can.

Referring to FIG. 6, the terrain data collection server may obtain height information of the green area to generate terrain information of the green area. In one embodiment, the terrain data collection server may obtain height information of the green area using at least one of a first device including a first sensor and a second device including a second sensor. The first device may include a stand-type device installed on the ground, and the second device may include a flying vehicle device. For example, the first device including the first sensor may be a stand-type device including a LiDAR sensor, and the second sensor may be a drone equipped with a LiDAR sensor. In one embodiment, before the inference server trains the learning model, the terrain data collection server may obtain height information of the drawn area using at least one of the first device and the second device. Additionally, when the terrain data collection server obtains the height information of the green area, the first device can be removed from the ground and the flight of the second device can be terminated. Additionally, the terrain data collection server may update the height information of the green area when a specific event occurs or periodically using at least one of the first device and the second device.

In one embodiment, the terrain data collection server uses at least one of green area terrain data acquired by a first device including a first sensor and green area terrain data acquired by a second device including a second sensor. This allows you to obtain height information of the drawn area. In addition, the terrain data collection server may generate terrain information of the green area by mapping the coordinate information of the green area described above with reference to FIG. 5 with the height information of the green area, and store the terrain information of the green area in a terrain database. . In one embodiment, the topographic information of the green area may include a height map of the green area.

In one embodiment of the present disclosure, a method of generating height information of the green area using two sensors has been described, but this is not limited and the height information of the green area can also be generated using one sensor or three or more sensors. there is.

Referring to FIG. 7, the terrain data collection server generates a height map (700) of the area drawn based on terrain data collected using at least one of a first device including a first sensor and a second device including a second sensor. ) can be created. In one embodiment, the terrain data collection server may generate a height map 700 of the green area by focusing on the height of the green area. Additionally, the terrain data collection server may store the height map 700 of the green area in a database in the form of an image or the like. For example, referring to FIG. 7 , the yellow color of the height map 700 of the green area may indicate higher terrain, and the darker green color may indicate lower terrain.

In one embodiment, once the height map of the green area is created and stored in the database, a golf course management server, inference server, etc. can later obtain the height map of the green area from the database and use it whenever necessary. However, the terrain data collection server may update and store the height map of the green area at regular intervals, or update and store the height map of the green area when a specific event occurs.

According to one embodiment, by storing the height map of the green area or the topographic information of the green area in the form of an image, etc., the topographic data collection server can store the topographic information of the green area in the database using few resources. Additionally, the golf course management server, inference server, etc. can be used to obtain inference results by obtaining topographic information of the green area from the database whenever necessary.

Referring to FIG. 8, the inference server uses the AI model 820 to determine topographic information of the drawn area (e.g., height map of the drawn area), location of the target point (e.g., location of hole cup), and location of the user terminal (e.g. : Putting information 830 can be generated by inputting information including golfer's position), green area state information (e.g. green friction force), ball state information (e.g. ball repulsion force), etc. as input 810. there is. In one embodiment, putting information 830 may include optimal putting distance, optimal putting angle, etc. However, this is only an example and the input and output information of the AI model is not limited to this.

Referring to FIG. 9, in step 905, the inference server may configure a simulation environment for training a reinforcement learning model. In one embodiment, the simulation environment may include a screen golf-based simulation environment, a 3D (three-dimensional) physics engine-based simulation environment, etc. The simulation environment may be configured to be the same or similar to putting on a green. The inference server can set and measure the situation when putting is performed in a simulation environment. For example, the inference server can set the golfer's position, hole cup position, green friction coefficient, ball restitution coefficient, wind strength, wind direction, etc. in the simulation environment.

According to one embodiment, the inference server can easily train the learning model up to a target number of times by training the learning model in a simulation environment. In addition, the inference server can have the effect of providing a screen golf service and learning a learning model by using screen golf in which topographical information of the green area is implemented to learn a learning model. Additionally, the inference server can be used to learn a reinforcement learning model using a 3D physics engine to respond to complex terrain or environments that cannot be accessed through screen golf alone.

In step 910, the inference server may initialize the reinforcement learning model and evaluation function and set the target learning number. Additionally, the inference server can determine a policy. For example, the inference server can determine a policy by taking the action with the highest evaluation score in a specific state.

At step 915, the inference server may initialize the scenario and set the user's location. In one embodiment, the inference server may obtain topographic information of the green area, the location of the target point (e.g., the location of the hole cup), and determine the user's location (e.g., the golfer's location). Additionally, the inference server can set the friction force of the green area, the ball's repulsion force, wind strength, wind direction, etc. The scenario may mean that during the simulation process, topographic information of the green area, location at the time of putting, inferred putting strength and putting direction, and evaluation scores for the results of putting in the inferred putting strength and putting direction are recorded. .

In step 920, the inference server may obtain the optimal direction and optimal intensity using a reinforcement learning model. In one embodiment, the inference server uses a reinforcement learning model to determine the optimal direction by using the terrain information of the green area, the location of the target point, the user's location, the friction force of the green area, the ball's repulsion force, wind strength, and wind direction as input. And the optimal intensity can be derived.

In step 925, the inference server may control the simulation environment so that putting is performed with the optimal direction and optimal strength derived from the simulation environment.

In step 930, the inference server may determine the arrival location of the ball and whether the target point is reached based on the evaluation function. For example, the inference server can determine the arrival location of the ball, whether a hole-in is successful, etc., obtained as a result of putting in a simulation environment with optimal direction and optimal strength.

In step 935, the inference server may train a reinforcement learning model based on the evaluation score of the evaluation function. In one embodiment, the inference server may train a reinforcement learning model based on evaluation scores and policies without knowledge of the correct answer or prior environment. In addition, the inference server can learn a reinforcement learning model by considering not only the evaluation score obtained as a single result of step 930, but also the evaluation score obtained by repeating step 930 so far and the evaluation score that will be obtained by performing step 930 in the future. there is. In one embodiment, the evaluation score may include at least one of a score for the putting result, a score for the number of putts, and a score considering both the putting result and the number of putts.

For example, if there is no hole-in, the inference server may determine the evaluation score (R _i+1 ) as R _i+1 = R _i - 1. Here, R _i may be an evaluation score for the previous putting. In other words, the evaluation score for each number of putts can be calculated as -1, so that as the number of putts until the hole-in increases, the evaluation score is lower, and as the number of putts decreases, the evaluation score is higher.

At step 940, the inference server may identify whether the hole-in was successful in the simulation environment. In one embodiment, the inference server may proceed to step 945 if it identifies that the hole-in has failed, and may proceed to step 950 if it identifies that the hole-in has been successful.

In step 945, the inference server may record the location at the time of putting, topographical information of the green area, inferred putting strength and direction, evaluation score, etc. as a scenario based on the failed hole-in. Additionally, the inference server may perform step 920 again based on the ball's arrival location.

In step 950, the inference server may update the evaluation function and initialize the scenario based on the recorded scenario based on a successful hole-in.

At step 955, the inference server may identify whether the learning number has reached the target. In one embodiment, if the inference server identifies that the learning number has reached the target, learning may be terminated at step 960, and if the inference server identifies that the learning number has not reached the target, step 915 may be performed.

Referring to FIG. 10, the agent (Agent, 1010) is a learning target of reinforcement learning. The agent (1010) does not simply increase the immediate evaluation score (R _i ), but also increases the evaluation score (ΣR _i ) received during one scenario. You can perform actions to maximize it. The state may be the agent 1010's own state. For example, the state includes the user's location (or the virtual user's location in the simulation environment), the location of the ball, the location of the target point, the green area's status information such as the terrain of the green area, weather, and humidity, and whether the target point has been reached. (e.g. whether hole in or not), etc. may be included. Action 1030 may be an action performed by the agent 1010 when given a specific state in the environment 1050. The agent 1010 may determine putting information including putting direction and putting strength based on the status and reward, and take action 1030 based on the determined putting information. In one embodiment, the action 1030 may include a target direction and a target distance from the user's current location based on the state in which the hole cup is viewed from the user's current location or the ball's location. Alternatively, action 1030 may include a putting operation performed based on the target direction and target distance. For example, the action may include putting with force to ensure the putting distance is 8.5m and putting in the direction of 'one club to the left of the hole cup'. In one embodiment, agent 1010 may update state by taking action 1030. Environment 1050 may include a simulation environment for training a reinforcement learning model. In addition, the environment 1050 may include physical laws of the world or various constraints for learning the agent 1010, including the state, which are measurable and expressible in numbers and will affect the results. Expected elements may be included in the state.

Referring to FIG. 11, the reinforcement learning model may include a behavioral policy model (Policy network) and a value evaluation model (Q-value network). In one embodiment, the action policy model includes location information of the green area, location information of the target point (e.g., location of the hole cup), location information of the user (e.g., location of the user terminal, location of the virtual user in the simulation), and green area information. Environmental information including state information (e.g. friction of the green), ball state information (e.g. ball repulsion force, surface material of the ball), weather condition information, user's body information, etc. are inputted to determine putting strength and putting direction. etc. can be decided. However, the information included in the environmental information is not limited to this, and various other information may be included. Additionally, the reinforcement learning model can determine the value based on the above environmental information, putting strength, putting direction, etc. determined by the behavioral policy model.

In one embodiment, the action policy model may correspond to the action 1030 described above with reference to FIG. 10 , and the valuation model may correspond to the state and reward described above with reference to FIG. 10 .

Referring to FIG. 12, the golf course management server may obtain coordinate information of the user terminal from the user terminal. Additionally, the golf course management server can obtain coordinate information of the target point (eg, hole cup). In one embodiment, coordinate information of the user terminal may be received from the user terminal. Additionally, the coordinate information of the target point can be received from the GPS device attached to the flagpole or from the greenkeeper's terminal. For example, the golf course management server can obtain the location of the hole cup using the coordinate information of the target point before the golf course opens every day.

In one embodiment, the golf course management server may correct the coordinate information of the user terminal and the coordinate information of the target point using RTK GPS to generate the user's location information and the location information of the target point. In other words, since errors may exist in GPS coordinates, the golf course management server corrects the GPS coordinates of the user's GPS coordinate target point using GPS with RTK applied to specify a more accurate location and determines the user's location (1210) and target point. The location 1220 can be specified.

In one embodiment, a GPS device may be installed around the green area at a location whose latitude, altitude, and longitude are accurately known. This GPS device will be referred to as a GPS Base device. As a plurality of GPS base devices are installed around the green area, the golf course management server determines the user's location (1210) and the target point's location (1220) based on the GPS coordinates of the user terminal and the GPS coordinates of the hole cup. Errors in GPS coordinates can be corrected by considering the relative positions of multiple GPS base devices.

According to one embodiment, the user's location 1210 and the target point location 1220 can be identified as accurate locations with almost no error using RTK GPS.

Referring to FIG. 13, the user terminal may include an AR device. The user terminal may receive putting information including putting strength and putting direction from the golf course management server, and display the putting information through the AR device. According to one embodiment, as shown in FIG. 13, the AR device may display the distance to the hole cup and the distance to the target point to indicate the strength and direction of the putting. For example, the AR device can display that the putting strength must increase more than when putting in the direction of the hole cup by expressing the distance to the target point as farther than the distance to the hole cup. Additionally, the AR device can display the location of the hole cup and the target point to indicate that putting should be done to the left of the hole cup through AR on a terrain feature. However, this is only an example and the method by which the user terminal displays putting information is not limited to this.

Referring to FIG. 14, the golf course management server 1400 may include a transceiver 1410, a memory 1420, and a processor 1430. Additionally, the processor 1430 may be implemented as a software module. However, not all of the components shown in FIG. 14 are essential components of the golf course management server 1400. The golf course management server 1400 may be implemented with more components than those shown in FIG. 14, or the golf course management server 1400 may be implemented with fewer components than those shown in FIG. 14. In addition, the transceiver 1410, memory 1420, and processor 1430 may be implemented in the form of a single chip.

In one embodiment, the golf course management server 1400 may not be implemented as a separate server, but as a virtual server using a cloud server. In another embodiment, the golf course management server 1400 may be implemented as a function on another server and may operate on the server in the form of a software module. For example, the golf course management server 1400 may be implemented as a single server and an inference server.

In one embodiment, the transceiver unit 1410 may communicate with a terminal, server, or other electronic device connected to the golf course management server 1400 by wire or wirelessly. For example, the transceiver 1410 receives a putting information request including the location coordinates of the user terminal from the user terminal, and based on the putting information request, sends the inference server the user's information determined based on the location coordinates of the user terminal. Transmits an inference request message including location information, receives an inference result including putting direction information and putting strength information from the inference server based on the inference request message, and includes putting direction information and putting strength information to the user terminal. Putting information can be transmitted.

Various types of data, such as programs and files such as applications, can be installed and stored in the memory 1420. The processor 1430 may access and use data stored in the memory 1420, or may store new data in the memory 1420.

The processor 1430 controls the overall operation of the golf course management server 1400 and may include at least one processor such as a CPU or GPU. The processor 1430 may control other components included in the golf course management server 1400 to perform operations for operating the golf course management server 1400. For example, the processor 1430 may execute a program stored in the memory 1420, read a stored file, or save a new file. In one embodiment, the processor 1430 may perform an operation to operate the golf course management server 1400 by executing a program stored in the memory 1420.

Referring to FIG. 15, the inference server 1500 may include a transceiver 1510, a memory 1520, and a processor 1530. In addition, the processor 1530 includes a data acquisition unit 1540, a preprocessor 1550, a data selection unit 1560, a putting direction determination unit 1570, a putting intensity determination unit 1580, and a model evaluation unit 1590. It can be included. In one embodiment, the data acquisition unit 1540, the preprocessor 1550, the data selection unit 1560, the putting direction determination unit 1570, the putting intensity determination unit 1580, and the model evaluation unit 1590 are software modules. It can be implemented as: However, not all of the components shown in FIG. 15 are essential components of the inference server 1500. The inference server 1500 may be implemented with more components than those shown in FIG. 15 , or the inference server 1500 may be implemented with fewer components than the components shown in FIG. 15 . In addition, the transceiver 1510, memory 1520, and processor 1530 may be implemented in the form of a single chip.

In one embodiment, the inference server 1500 may not be implemented as a separate server, but as a virtual server using a cloud server. In another embodiment, the inference server 1500 may be implemented as a function on another server and may operate on the server in the form of a software module. For example, the inference server 1500 may be implemented as one server with the golf course management server 1400.

In one embodiment, the transceiver 1510 may communicate with a terminal, server, or other electronic device connected to the inference server 1500 by wire or wirelessly. For example, the transceiver 1510 may receive an inference request message including user location information from a golf course management server, and transmit an inference result including putting direction information and putting strength information to the golf course management server.

Various types of data, such as programs and files such as applications, can be installed and stored in the memory 1520. The processor 1530 may access and use data stored in the memory 1520, or may store new data in the memory 1520.

The processor 1530 controls the overall operation of the inference server 1500 and may control other components included in the inference server 1500 to perform operations for operating the inference server 1500. For example, the processor 1530 may execute a program stored in the memory 1520, read a stored file, or save a new file. In one embodiment, the processor 1530 may perform an operation to operate the inference server 1500 by executing a program stored in the memory 1520.

The artificial intelligence-related functions of the inference server 1500 according to the present disclosure are operated through a processor and memory. The processor may consist of one or multiple processors. At this time, one or more processors may be a general-purpose processor such as a CPU, AP, or DSP (Digital Signal Processor), a graphics-specific processor such as a GPU or VPU (Vision Processing Unit), or an artificial intelligence-specific processor such as an NPU. One or more processors control input data to be processed according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, when one or more processors are dedicated artificial intelligence processors, the artificial intelligence dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

Predefined operation rules or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that the basic artificial intelligence model is learned using a large number of learning data by a learning algorithm, thereby creating a predefined operation rule or artificial intelligence model set to perform the desired characteristics (or purpose). It means burden. This learning may be accomplished in the device itself that performs the artificial intelligence according to the present disclosure, or may be accomplished through a separate server and/or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

An artificial intelligence model may be composed of multiple neural network layers. Each of the plurality of neural network layers has a plurality of weight values, and neural network calculation is performed through calculation between the calculation result of the previous layer and the plurality of weights. Multiple weights of multiple neural network layers can be optimized by the learning results of the artificial intelligence model. For example, a plurality of weights may be updated so that loss or cost values obtained from the artificial intelligence model are reduced or minimized during the learning process. Artificial neural networks may include deep neural networks (DNN), such as Convolutional Neural Network (CNN), Deep Neural Network (DNN), Recurrent Neural Network (RNN), Restricted Boltzmann Machine (RBM), Deep Belief Network (DBN), Bidirectional Recurrent Deep Neural Network (BRDNN), or Deep Q-Networks, etc., but are not limited to the examples described above.

In one embodiment, the data acquisition unit 1540 collects the user's location information, target point location information, green area terrain information, weather condition information, user's body information, ball characteristic information, green area characteristic information, etc. Data including can be obtained. The data acquisition unit 1540 may acquire data from a terminal, server, database, or other electronic device connected to the inference server 1500 by wire or wirelessly.

In one embodiment, the preprocessor 1550 may preprocess the acquired data so that the data can be used for learning to infer the putting direction and putting strength. For example, the preprocessor 1550 may process the data into a preset format, such as removing duplicate data from the acquired data, removing data with a low possibility, vectorizing metadata corresponding to each data, etc. , but is not limited to this.

In one embodiment, the data selection unit 1560 may select data necessary for learning from preprocessed data. The data selection unit 1560 may select data required for learning from preprocessed data according to preset standards for inferring putting strength and putting direction.

In one embodiment, the putting direction determination unit 1570 and the putting intensity determination unit 1580 may determine the putting direction and putting intensity, respectively, based on data selected by the data selection unit 1560. Additionally, the model evaluation unit 1590 may perform re-learning when evaluation data is input and the inference result output from the evaluation data does not satisfy a predetermined standard. In this case, the evaluation data may be preset data for evaluating the learning model.

An embodiment of the present disclosure may also be implemented in the form of a recording medium containing instructions executable by a computer, such as program modules executed by a computer. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may include both computer storage media and communication media. Computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Communication media typically includes computer-readable instructions, data structures, or program modules and includes any information delivery medium.

The foregoing description of the present disclosure is for illustrative purposes, and a person skilled in the art to which the present disclosure pertains will understand that the present invention can be easily modified into other specific forms without changing its technical idea or essential features. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present disclosure is indicated by the claims described below rather than the detailed description above, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present disclosure. do.

Claims

Receiving a putting information request including location coordinates of the user terminal from a user terminal;

Based on the putting information request, transmitting an inference request message including user location information determined based on the location coordinates of the user terminal to an inference server;

Based on the inference request message, receiving an inference result including putting direction information and putting strength information from the inference server;

Comprising the step of transmitting putting information including the putting direction information and the putting intensity information to the user terminal,

The inference result including the putting direction information and the putting strength information is,

A golf putting assistance method that is generated using a learning model learned using an artificial intelligence algorithm by the inference server based on the user's location information and topographic information of the green area where the user is located.
The method of claim 1, wherein the terrain information of the green area is:

Contains information generated by mapping coordinate information corresponding to the green area and height information of the green area,

The height information of the green area is,

A golf putting assistance method obtained using a first device installed on the ground including a first sensor and a second device including a flying vehicle including a second sensor.
The method of claim 1, wherein the learning model is:

Contains reinforcement learning models, including behavioral policy models and valuation models;

The behavioral policy model is,

Regarding the putting direction and putting strength based on environmental information including at least one of target point location information, putting start location information, topography information of the green area, ball characteristic information, user status information, and weather condition information. generate behavioral information,

The valuation model above is,

A golf putting assistance method that determines a reward based on the environmental information and the behavioral information.
The method of claim 1, wherein transmitting the inference request message comprises:

Based on a putting information request including location coordinates of the user, determining whether the user is located in the green area; and

When determining that the user is located in the green area, transmitting the inference request message including location information of the user to the inference server.
The method of claim 1, wherein the golf putting assistance method includes:

A golf putting assistance method further comprising generating location information of the user by correcting the location coordinates of the user terminal using a Real-Time Kinematic (RTK) global positioning system (GPS).
The method of claim 1, wherein the putting information including the putting direction information and the putting intensity information,

A golf putting assistance method including information to be displayed through AR (Augmented Reality) by the user terminal.
In the method of operating the inference server,

Receiving an inference request message including user location information from a golf course management server;

Obtaining topographic information of a green area where the user is located;

Generating putting direction information and putting strength information using a learning model learned using an artificial intelligence algorithm, based on the user's location information and green area topography information; and

A method of operating an inference server, comprising transmitting an inference result including the putting direction information and the putting intensity information to the golf course management server.
The method of claim 7, wherein the step of acquiring topographic information of the green area includes:

requesting topographic information of the green area from the golf course management server; and

Receiving topographic information of the green area that maps coordinate information corresponding to the green area and height information of the green area from the golf course management server,

The height information of the green area is,

A method of operating an inference server, obtained using a first device installed on the ground including a first sensor and a second air vehicle device including a second sensor.
The method of claim 7, wherein the step of acquiring topographic information of the green area includes:

Transmitting a request for topographic information of the green area, including at least one of golf course information or location information of the user; and

Based on a request for terrain information of the green area, receiving terrain information of the green area generated by mapping coordinate information corresponding to the green area and height information of the green area,

The height information of the green area is,

A method of operating an inference server, obtained using a first device including a first sensor and a second device including a second sensor.
The method of claim 7, wherein the learning model is:

Contains reinforcement learning models, including behavioral policy models and valuation models;

The step of generating the putting direction information and the putting intensity information is,

By the behavioral policy model, the putting direction and putting intensity are determined based on environmental information including at least one of putting start position information, topography information of the green area, ball characteristic information, user status information, or weather condition information. generating behavioral information about; and

A method of operating an inference server, comprising generating a reward based on the environmental information and the behavioral information by the value evaluation model.
The method of claim 7, wherein the inference request message is:

A method of operating an inference server, transmitted to the inference server when the golf course management server determines that the user is located in the green area.
The method of claim 7, wherein the user's location information is:

A method of operating an inference server that is generated by correcting the location coordinates of a user terminal using RTK (Real-Time Kinematic) GPS (global positioning system).
The method of claim 7, wherein the putting direction information and the putting intensity information are:

A method of operating an inference server, which is delivered to a user terminal including an AR (Augmented Reality) device by the golf course management server.
a memory storing one or more instructions; and

At least one processor executing the one or more instructions stored in the memory,

The at least one processor executes the one or more instructions,

Receiving a putting information request including the location coordinates of the user terminal from a user terminal,

Based on the putting information request, transmitting an inference request message containing user location information determined based on the location coordinates of the user terminal to the inference server,

Based on the inference request message, receive an inference result including putting direction information and putting strength information from the inference server,

Transmitting putting information including the putting direction information and the putting intensity information to the user terminal,

The inference result including the putting direction information and the putting strength information is,

A golf course management server generated using a learning model learned using an artificial intelligence algorithm by the inference server based on the user's location information and topographic information of the green area where the user is located.
15. The method of claim 14, wherein the at least one processor:

Obtain coordinate information corresponding to the drawn area,

Obtain height information of the green area using a first device including a first sensor and a second device including a second sensor, wherein the first device includes a device installed on the ground and the second device includes a flying vehicle. Including;

Generate topographic information of the green area by mapping coordinate information corresponding to the green area and height information of the green area,

A golf course management server that stores terrain information of the green area in a database.
The method of claim 14, wherein the terrain information of the green area is:

Contains information generated by mapping coordinate information corresponding to the green area and height information of the green area,

The height information of the green area is,

A golf course management server obtained using a first device installed on the ground including a first sensor and a second device including a flying vehicle including a second sensor.
The method of claim 14, wherein the learning model is:

Contains reinforcement learning models, including behavioral policy models and valuation models;

The behavioral policy model is,

Regarding the putting direction and putting strength based on environmental information including at least one of target point location information, putting start location information, topography information of the green area, ball characteristic information, user status information, and weather condition information. generate behavioral information,

The valuation model above is,

A golf course management server that determines a reward based on the environmental information and the behavioral information.
15. The method of claim 14, wherein the at least one processor:

Based on a putting information request including the location coordinates of the user, determine whether the user is located in the green area,

When determining that the user is located in the green area, the golf course management server transmits the inference request message including the location information of the user to the inference server.
15. The method of claim 14, wherein the at least one processor:

A golf course management server that generates location information of the user by correcting the location coordinates of the user terminal using RTK (Real-Time Kinematic) GPS (global positioning system).
The method of claim 14, wherein the putting information including the putting direction information and the putting intensity information,

A golf course management server that includes information to be displayed through AR (Augmented Reality) by the user terminal.
a memory storing one or more instructions; and

At least one processor executing the one or more instructions stored in the memory,

The at least one processor executes the one or more instructions,

Receive an inference request message containing the user's location information from the golf course management server,

Obtain topographic information of the green area where the user is located,

Based on the user's location information and the topographical information of the green area, putting direction information and putting strength information are generated using a learning model learned using an artificial intelligence algorithm,

An inference server that transmits an inference result including the putting direction information and the putting intensity information to the golf course management server.
22. The method of claim 21, wherein the at least one processor:

Request topographic information of the green area from the golf course management server,

Receive topographic information of the green area that maps coordinate information corresponding to the green area and height information of the green area from the golf course management server,

The height information of the green area is,

An inference server obtained using a first device installed on the ground including a first sensor and a second air vehicle device including a second sensor.
22. The method of claim 21, wherein the at least one processor:

Transmitting a request for topographic information of the green area, including at least one of golf course information or location information of the user,

Based on a request for terrain information of the green area, receive terrain information of the green area generated by mapping coordinate information corresponding to the green area and height information of the green area,

The height information of the green area is,

An inference server, obtained using a first device including a first sensor and a second device including a second sensor.
The method of claim 21, wherein the learning model is:

Contains reinforcement learning models, including behavioral policy models and valuation models;

The at least one processor,

By the behavioral policy model, the putting direction and putting intensity are determined based on environmental information including at least one of putting start position information, topography information of the green area, ball characteristic information, user status information, or weather condition information. generate behavioral information about

An inference server that generates a reward based on the environmental information and the behavioral information by the value evaluation model.
The method of claim 21, wherein the inference request message is:

An inference server transmitted to the inference server when the golf course management server determines that the user is located in the green area.
The method of claim 21, wherein the user's location information is:

An inference server created by correcting the location coordinates of the user terminal using RTK (Real-Time Kinematic) GPS (global positioning system).
The method of claim 21, wherein the putting direction information and the putting intensity information are:

An inference server transmitted by the golf course management server to a user terminal including an AR (Augmented Reality) device.
A program stored in a computer-readable recording medium to execute the method of any one of claims 1 to 13 on a computer.