Systems and Methods for Generating Shot Strategies for Golfers

This application claims priority under 35 U.S.C. § 119 (e) to U.S. Provisional Application No. 63/569,663, filed Mar. 25, 2024, and titled “SYSTEMS AND METHODS FOR GENERATING SHOT STRATEGIES FOR GOLFERS,” which is hereby incorporated by reference herein in its entirety.

The present disclosure relates to systems and methods for generating shot strategies to provide guidance for golfers.

Golf is a sport in which players use clubs to hit balls into a series of pin cups on a golf course in as few strokes as possible. Golf is played by a golfer on a golf course made up of multiple course segments, commonly called “holes.” A hole typically includes one or more tee locations, a green with a pin cup, an elongate fairway generally between the tee locations and the green, one or more hazards, and/or rough surrounding at least a portion of the fairway or green. Portions of a hole may also be bounded by out of bounds areas from which a golfer may not play. Generally, the objective of golf is to play a golf ball from an initial ball location (e.g., the tee location) to the pin cup on the green by successive strokes. The golfer uses a variety of clubs with different characteristics, typically categorized into woods, irons, wedges, and putters. A small peg called a tee can be used to lift the golf ball off the ground for the initial shot (“tee shot” or “golf shot”) on each hole should the golfer so desire. Subsequent shots are typically played from where the ball comes to rest from the previous shot, e.g. the fairway, the rough, hazards (such as bunkers, lateral hazards, and water hazards), or the green, which contains the pin cup (which can also be called a “hole,” not to be confused with the course segments or “holes” of the golf course). In some instances, the golfer is presented with options on where to play the next shot should the ball come to rest in an unplayable lie or inside an out of bounds area. A pin with a flag attached is commonly inserted into the pin cup in order to indicate the location of the pin cup to the golfer. Courses are designed with course segments or holes of varying lengths and degrees of difficulty. They are often landscaped with trees and other vegetation to provide a scenic appeal. The pin cup can typically be placed in a variety of positions or regions on the green depending on the course set up desired by the course management.

Golf is scored based on the total number of strokes taken to progress the ball from each tee location to each pin cup, across all holes played on a course. Golf scoring terminology includes par (the expected number of strokes for a skilled golfer to complete the hole), birdie (one stroke under par), eagle (two strokes under par), bogey (one stroke over par) and double bogey (two strokes over par).

Many golfers play the game with little or no expert guidance related to their shot strategies. An unguided golfer may rely on intuition to select a shot strategy that may include, for example, aiming a shot in as direct a path as possible towards the pin cup location. A well-reasoned shot strategy would potentially follow a different path. Casual golfers may lack awareness of better shot strategies or may consider more sophisticated shot selection to be beyond their abilities. Furthermore, even if a golfer is trained to correctly analyze their shot direction and angle, every course varies in topography, elevation, vegetation, and weather patterns—factors that can challenge even an experienced golfer when selecting a preferred shot strategy. Further still, previous methods for recommending shot strategies are limited by their inability to account for successive shots inherent to multi-shot strategies. For example, aiming a golf shot directly towards a pin cup location may not be a preferred shot strategy if the corresponding landing location would cause successive shots to encounter hazards and other conditions which could have been avoided had the factors affecting successive shots been considered while generating the shot strategy.

The systems, methods, and devices of this disclosure each have several innovative aspects, no single one of which is solely responsible for all of the desirable attributes disclosed herein. Details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the description below.

In some aspects, the techniques described herein relate to a method for generating shot strategies in a golf game, the method, as implemented by computer program instructions executed by one or more computer processors, can include: receiving a Golf Performance Prediction Grid (GPP Grid) of a golf hole, wherein the GPP Grid includes a positional grid including a plurality of grid positions, wherein each of the plurality of grid positions has a grid value, and wherein each of the plurality of grid positions is associated with a geospatial location within a layout of the golf hole; selecting a preferred shot strategy by a shot strategy selection process including: (a) receiving a ball location within the layout of the golf hole; (b) determining a landing location associated with a shot vector and the ball location; (c) determining a golf performance value based at least on the grid value associated with the landing location from the GPP Grid; (d) determining a carry risk vector associated with the shot vector, wherein the carry risk vector corresponds to a risk of a ball path associated with the shot vector intersecting an obstacle along the ball path; (c) modifying the golf performance value based at least in part on the carry risk vector to generate a modified golf performance value; (f) repeating steps (a) through (c) for a plurality of shot vectors; and (g) selecting the shot vector of the plurality of shot vectors that satisfies a preferred shot selection criterion as at least a portion of the preferred shot strategy; and transmitting the preferred shot strategy; wherein the preferred shot strategy includes one or more shots, corresponding to one or more shot vectors, for a golfer to advance a golf ball from the ball location to a pin cup location.

In some aspects, the preferred shot selection criterion can be a lowest modified golf performance value. In some aspects, the shot strategy selection process can include: determining a shot roll and a roll risk vector associated with the shot vector, wherein the roll risk vector has a magnitude corresponding to a risk of the shot roll encountering a water feature; and modifying the golf performance value based at least in part on the roll risk vector to generate the modified golf performance value.

In some aspects, the plurality of shot vectors includes a shot sequence associated with a plurality of sequential shot vectors, wherein each ball location of each of the plurality of sequential shot vectors is a shot total location of a previous shot vector, and wherein the shot total location is a grid position where the golf ball comes to rest after traversing a distance corresponding to a shot carry and a shot roll. In some aspects, the shot strategy selection process includes aggregating modified golf performance values of each shot vector in the shot sequence to generate a shot sequence score.

In some aspects, the shot strategy selection process includes selecting the shot sequence that satisfies the preferred shot selection criterion as at least a portion of the preferred shot strategy. In some aspects, selecting the shot sequence includes selecting the shot sequence associated with the shot sequence score having a minimum value.

In some aspects, the shot strategy selection process further includes: determining an averaged grid value associated with the landing location or a shot total location, wherein the averaged grid value is determined by averaging grid values within a shot dispersion area surrounding the landing location; and determining the grid value at the ball location. In some aspects, determining the golf performance value is additionally based on the averaged grid value and the grid value at the ball location. In some aspects, the grid value probabilistically represents golf performance metrics in terms of a distance between a grid position and the pin cup location, geospatial information corresponding to the grid position, and a short side area.

In some aspects, the techniques described herein relate to a method, further including identifying the golf hole by obtaining geolocation information and associating the geolocation information with a golf course and a specific course segment of the golf course associated with the golf hole. In some aspects, the geolocation information includes at least one of a latitudinal location, a longitudinal location, or an elevation.

In some aspects, the techniques described herein relate to a method, further including: receiving shot dispersion data associated with the golfer and at least one golf club; and selecting at least one of the plurality of shot vectors by searching for shot vectors within an analysis region located within a range of shot distances from the ball location, wherein the range of shot distances is derived from the shot dispersion data.

In some aspects, the shot dispersion data includes a list of clubs, a shot carry, a shot roll, a shot total, and dimensions of a shot dispersion area. In some aspects, receiving the shot dispersion data includes prompting a predictive model with golfer data and receiving a result from the predictive model. In some aspects, the predictive model is sequentially trained using a plurality of machine learning models to improve a predictive accuracy of prior machine learning models through training sequential machine learning models to predict each prior machine learning model's incorrect predictions.

In some aspects, the shot dispersion data includes: an identifier for the at least one golf club; a shot carry corresponding to a horizontal distance the golf ball travels through air when the golf ball is hit with the at least one golf club; a shot roll corresponding to a distance the golf ball travels on ground when the golf ball is hit with the at least one golf club; a shot total including the shot carry and the shot roll; a shot shape value corresponding to a curvature of a golf shot; and a shot dispersion area including a length, a width, and an angle of rotation, wherein the shot dispersion area corresponds to a probabilistic region in which the golf ball contacts the ground after being struck by the at least one golf club.

In some aspects, receiving the shot dispersion data includes receiving data entered via user interaction with a manual dispersion data entry interface.

In some aspects, the GPP Grid is generated from a plurality of inputs including a polygon representation of the golf hole and short sides data associated with the golf hole. In some aspects, the polygon representation includes geospatial information of the golf hole, wherein the geospatial information includes: topographical characteristics of the golf hole, wherein the topographical characteristics include locations of one or more water features and locations of one or more bunker regions; boundary characteristics of the golf hole, wherein the boundary characteristics include out-of-bounds regions near the golf hole; and obstacle characteristics of the golf hole, wherein the obstacle characteristics include vegetation type and vegetation location, and water features. In some aspects, the polygon representation includes coded geometric shapes overlaying a representation of the golf hole. In some aspects, the coded geometric shapes are color-coded according to corresponding geospatial information of the golf hole.

In some aspects, the positional grid uses a coordinate system that is two-dimensional or three-dimensional. In some aspects, the pin cup location is a default pin cup location, wherein the default pin cup location is centrally located on a green of the golf hole. In some aspects, the default pin cup location is selected if the pin cup location is not selected by a user, received from a network computing system, retrieved from a course data store, or received from an end user device.

In some aspects, the plurality of shot vectors are generated from a data set including: a plurality of target shot distances derived from shot dispersion data for the golfer and the layout of the golf hole; a plurality of target landing locations derived from shot dispersion areas associated with the plurality of target shot distances; and environmental factors corresponding to the golf hole. In some aspects, the environmental factors include at least one of an elevation, an elevation change (e.g., “slope”), a direction of elevation change (e.g., “aspect”), a wind velocity, a temperature, or a relative humidity.

In some aspects, the shot strategy selection process includes: generating a remaining route value (RRV) for each of the plurality of grid positions in the GPP grid by adjusting the grid value at a grid position to account for golfer data, risk factors, and environmental factors associated with advancing the golf ball from the grid position towards the pin cup location or another landing location; generating an average RRV for the grid position, wherein the average RRV is generated by determining the average RRV associated with a shot dispersion area corresponding to the shot vector; selecting a preferred shot vector associated with a minimum value of the average RRV; and transmitting the preferred shot vector.

In some aspects, the shot strategy selection process is repeated to generate a shot sequence including a plurality of sequential preferred shot vectors. In some aspects, sequential shot vectors are generated until a selected grid position corresponds to a tee shot location.

In some aspects, the techniques described herein relate to a method for generating a Golf Performance Prediction Grid (GPP Grid) associated with a golf course segment. The method, as implemented by computer program instructions executed by one or more computer processors, can include: receiving a polygon representation of the golf course segment; receiving a pin cup location corresponding to a geospatial location on a green of the golf course segment; receiving short side data including an indication of a short side area off the green between a short side boundary of the green and an outer boundary of the short side area; generating a positional grid with dimensions corresponding to the golf course segment, the positional grid including a plurality of grid positions; generating position scores for each grid position, wherein each position score corresponds to a distance between each grid position and the pin cup location; correlating geospatial information to the plurality of grid positions, wherein the geospatial information corresponds to the polygon representation; and generating grid values based at least on a transformation associated with a particular course feature present at each of the plurality of grid positions.

In some aspects, generating the grid values is additionally based on the short side data, the position scores, and the geospatial information. In some aspects, the polygon representation includes an edited polygon representation that is modifiable using direct or indirect manipulation of geometric shapes.

In some aspects, the techniques described herein relate to a method for generating shot strategies for golf instruction. The method, as implemented by computer program instructions executed by one or more computer processors, can include: receiving a ball location on a golf hole; generating a preferred shot strategy for advancing a golf ball from the ball location to a pin cup location on a green of the golf hole; generating data to display a graphical user interface showing the golf hole and the ball location; receiving, via user interaction with a training shot selection interface, a training shot for advancing the golf ball to a subsequent ball location; determining a score based on an analytical comparison of the training shot and the preferred shot strategy; and transmitting the score.

In some aspects, the ball location is selected by a user at any location on the golf hole. In some aspects, the pin cup location is generated by selecting a random location within a boundary of the green on the golf hole. In some aspects, generating the preferred shot strategy includes determining a golf performance value associated with a shot total location associated with a shot vector, wherein the golf performance value probabilistically represents golf performance metrics in terms of (i) estimated strokes to par or (ii) an estimated number of strokes to reach the pin cup location from the shot total location.

In some aspects, the techniques described herein relate to a method of generating an edited polygon representation of a golf hole. The method, as implemented by computer program instructions executed by one or more computer processors, can include: receiving a polygon representation of the golf hole; receiving an indication of an intent to modify the polygon representation; receiving, via user interaction with a polygon manipulation interface, proposed modifications to the polygon representation, wherein the proposed modifications to the polygon representation include addition, deletion, or resizing of one or more geometric shapes; generating data to display the proposed modifications to the polygon representation; receiving, via user interaction with a modification confirmation interface, confirmation that the proposed modifications are to be committed; in response to receiving the confirmation that the proposed modifications are to be committed, generating the edited polygon representation based on the polygon representation and the proposed modifications; and transmitting the edited polygon representation.

In some aspects, the techniques described herein relate to a method of generating short sides data for a golf hole. The method, as implemented by computer program instructions executed by one or more computer processors, can include: receiving a pin cup location on a green of the golf hole; determining a short side boundary including a portion of a perimeter of the green within a predefined distance from the pin cup location; determining a short side area of the golf hole including an area between the short side boundary and the predefined distance from the pin cup location; determining a short side penalty value to discount a grid value associated with a shot vector having a landing location inside of the short side area; and transmitting the short sides data including the short side area and the short side penalty value.

In some aspects, the predefined distance is a predetermined value, such as about 50 yards. In some aspects, such as where the predefined distance is a predetermined value, the predefined distance may be a fixed distance that does not vary. In some aspects, the predefined distance varies according to geospatial features in a vicinity of the short side boundary. In some aspects, at least a portion of the short side area extends to an edge of the golf hole.

In some aspects, the techniques described herein relate to a system configured to facilitate generating shot strategies in a golf game. The system can include one or more computer processors that are configured to perform a method including: receiving a Golf Performance Prediction Grid (GPP Grid) of a golf hole, wherein the GPP Grid includes a positional grid including a plurality of grid positions, wherein each of the plurality of grid positions has a grid value, and wherein each of the plurality of grid positions is associated with a geospatial location within a layout of the golf hole; selecting a preferred shot strategy by a shot strategy selection process including: (a) receiving a ball location within the layout of the golf hole; (b) determining a landing location associated with a shot vector and the ball location; (c) determining a golf performance value based at least on the grid value associated with the landing location from the GPP Grid; (d) determining a risk vector associated with the shot vector, wherein the risk vector corresponds to a risk of a ball path associated with the shot vector intersecting an obstacle along the ball path; (c) modifying the golf performance value based at least in part on the risk vector to generate a modified golf performance value; (f) repeating steps (a) through (c) for a plurality of shot vectors; and (g) selecting the shot vector of the plurality of shot vectors that satisfies a preferred shot selection criterion as at least a portion of the preferred shot strategy; and transmitting the preferred shot strategy; wherein the preferred shot strategy includes one or more shots, corresponding to one or more shot vectors, for a golfer to advance a golf ball from the ball location to a pin cup location.

In some aspects, the techniques described herein relate to a system configured to facilitate generating shot strategies in a golf game, the system including one or more computer processors that are configured to perform a method including: as implemented by computer program instructions executed by the one or more computer processors: receiving a Golf Performance Prediction Grid (GPP Grid) of a golf hole, wherein the GPP Grid includes a positional grid including a plurality of grid positions, wherein each of the plurality of grid positions has a grid value, and wherein each of the plurality of grid positions is associated with a geospatial location within a layout of the golf hole; selecting a preferred shot strategy by a shot strategy selection process including: (a) receiving a ball location within the layout of the golf hole; (b) determining a landing location associated with a shot vector and the ball location; (c) determining a golf performance value based at least on the grid value associated with the landing location from the GPP Grid; (d) determining a risk vector associated with the shot vector, wherein the risk vector corresponds to a risk of a ball path associated with the shot vector intersecting an obstacle along the ball path; (c) modifying the golf performance value based at least in part on the risk vector to generate a modified golf performance value; (f) repeating steps (a) through (c) for a plurality of shot vectors; and (g) selecting the shot vector of the plurality of shot vectors that satisfies a preferred shot selection criterion as at least a portion of the preferred shot strategy; and transmitting the preferred shot strategy; wherein the preferred shot strategy includes one or more shots, corresponding to one or more shot vectors, for a golfer to advance a golf ball from the ball location to a pin cup location.

In some aspects, the techniques described herein relate to a method for generating shot strategies in a golf game, the method including, as implemented by computer program instructions executed by one or more computer processors: receiving a ball location within a layout of a golf hole; receiving a Golf Performance Prediction Grid (GPP Grid) of the golf hole, wherein the GPP Grid includes a positional grid including a plurality of grid positions, wherein each of the plurality of grid positions has a grid value, wherein each of the plurality of grid positions is associated with a geospatial location within the layout of the golf hole, and wherein each grid position has a midpoint reference comprised of a midpoint, wherein the midpoint is a location at the center of a grid position and is used as a reference location for the grid position on the GPP grid; receiving, and updating the GPP Grid to account for a short side area; selecting a preferred shot strategy by a shot strategy selection process including: (a) determining a landing location associated with a shot vector and the ball location; (b) determining the grid value associated with the landing location from the GPP Grid; (c) determining an averaged grid value within a shot dispersion area corresponding to the landing location; (d) determining a risk vector associated with the existence of vegetation along the shot vector; (e) performing a roll-out check to identify water features along the shot vector or at the landing location and, if a water feature is identified, removing from consideration the corresponding shot vector; (f) applying the averaged grid value, the risk vector to the golf performance value associated with the landing location, and the golf performance value of the initial ball location to produce a modified golf performance value associated with the landing location; (g) repeating steps (a) through (f) for a plurality of shot vectors; (h) treating each landing location associated with the plurality of shot vectors as initial ball locations and repeating steps (a) through (g), for a plurality of shot vector sequences, until each shot vector sequence reaches a green, until each sequence contains a predefined number of shot vectors, or another metric; and (i) selecting the shot vector sequence of the plurality of shot vector sequences that satisfied a preferred shot vector selection criterion; and transmitting the preferred shot strategy; wherein the preferred shot strategy includes one or more shots for a golfer to advance a golf ball from the ball location to a pin cup location in a statistically preferential path.

The headings provided herein, if any, are for convenience only and do not necessarily affect the scope or meaning of the claimed invention. Although certain preferred implementations, embodiments, and examples are disclosed below, the inventive subject matter extends beyond the specifically disclosed implementations to other alternative implementations and/or uses and to modifications and equivalents thereof. Thus, the scope of the claims appended hereto is not limited by any of the particular implementations described below. For example, in any method or process disclosed herein, the acts or operations of the method or process may be performed in any suitable sequence and are not necessarily limited to any particular disclosed sequence. Various operations may be described as multiple discrete operations in turn, in a manner that may be helpful in understanding certain implementations; however, the order of description should not be construed to imply that these operations are order dependent. Additionally, the structures, systems, and/or devices described herein may be embodied as integrated components or as separate components. For purposes of comparing various implementations, certain aspects and advantages of these implementations are described. Not necessarily all such aspects or advantages are achieved by any particular implementation. Thus, for example, various implementations may be carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other aspects or advantages as may also be taught or suggested herein.

Generally described, aspects of the present disclosure relate to systems and methods that use one or more computing systems to offer constructive guidance related to a game of golf and/or golf training. The disclosed embodiments address technical problems inherent within automated or semi-automated methods for generating golf shot strategies including, but not limited to, generating preferred shot strategies which account for course data (e.g., geospatial information), golfer data (e.g., personalized demographic information, club dispersion, etc.), and other data (e.g., weather information and other environmental factors). These technical problems are addressed by the various technical solutions described herein, including generation of shot strategies, accounting for the short side areas, use of golf performance prediction (GPP) grids, such as estimated shots to finish (ESTF) grids, generation of shot vectors to account for hazardous vegetation, and use of editable polygon representations of a golf course or of one or more golf hole(s) within the golf course. Thus, the present disclosure represents an improvement to computing systems used to offer guidance related to a game of golf and/or to golf training.

In some embodiments, a shot strategy is generated by using a 2-dimensional or 3-dimensional grid of probabilistic values that represent the distances between each position on the grid and a pin cup location on a green, geospatial features which can describe the hole of golf, and the existence of a short side area on the golf hole. The purpose of the grid is to probabilistically evaluate golf performance (e.g., in terms of estimated strokes to advance a gold ball to the pin cup location) at different locations within the boundaries of a golf hole. The structured data in the grid is organized in a way that allows for analysis and visualization of potential across the course or course segments. Such a grid of values can be called various names, including, for example, a “Golf Performance Prediction Grid” (or “GPP Grid”), or an “Estimated Strokes to Finish Grid” (or “ESTF Grid”). In this disclosure, the grid of probabilistic values will generally be referred to as a GPP Grid, and it is understood that such a grid could encompass any of the implementations described in this paragraph or elsewhere in this disclosure.

In some implementations, a GPP Grid includes a coordinate system where each grid position corresponds to grid values. These grid values probabilistically represents golf performance metrics which account for the distances between each position on the grid and a pin cup location on a green, geospatial features which can describe the hole of golf, and the existence of a short side area on the golf hole. Each position within the hole's layout is linked to a distinct grid position on the GPP Grid, facilitating a strategic analysis of potential outcomes based on ball location and pin cup location. The dimensions of the GPP Grid can be correlated to the dimensions of the hole or the course segment. The grid values in each grid position account for a distance between the grid position and the pin cup and course data associated with the hole. The pin cup location can refer to a point (or region) on the selected hole and within the green where the actual hole or cup is located.

In some embodiments, geospatial information can correspond to geospatial features, which can include topographical characteristics of the golf hole, wherein the topographical characteristics include locations of one or more water features and locations of one or more bunker regions; boundary characteristics of the golf hole, wherein the boundary characteristics include out-of-bounds regions near the golf hole; location and dimensions of fairway, rough, a tee box, hazard, and green; and obstacle characteristics of the golf hole, wherein the obstacle characteristics include vegetation type and vegetation location. Regions of the polygon representation that include obstacle characteristics can be referred to as obstacle regions.

A shot strategy generation system can use a GPP Grid in combination with shot dispersion data, other golfer-specific data, other course-specific data, elevation characteristics, and other types of data such as weather information to determine a preferred shot strategy. A golfer using the disclosed method and/or system can receive the preferred shot strategy representing a preferred path for advancing a golf ball from an initial ball location to the pin cup location.

depicts an example computing environmentin which embodiments of the present disclosure can be implemented by a shot strategy generation systemto generate shot strategies for use by golfers in a game of golf, in training, while previewing a golf hole, or in other situations. The computing environmentmay include the shot strategy generation system, a network, any number of course data store(s), any number of golfer data store(s), a network computing system, and end user devices. The shot strategy generation systemcan be accessed by the end user devicesthrough the network. In other embodiments, the shot strategy generation systemcan be integrated with one or more end user devices. The network computing systemmay host one or more websites, application programing interfaces (APIs), or information services that can be accessed by the end user devicesthrough the network. The shot strategy generation systemcan access the course data store(s), golfer data store(s), and the network computing systemthrough the network.

Generally described, the shot strategy generation systemcan generate a preferred shot strategy from a large number (e.g., hundreds of thousands, millions, or more) of potential shot strategies through automated or semi-automated shot strategy selection processes. For example, the shot strategy generation systemmay analyze shot strategies in combination with course data stored in the course data store(s)and golfer data stored in the golfer data store(s). The shot strategy generation systemmay perform a shot strategy selection procedure in response to receiving request(s) to generate a preferred shot strategy from one or more end user devices. Alternatively, or in addition, the shot strategy generation systemmay generate preferred shot strategies based on a triggering event, such as the opening of an application on an end user device, a change in course data or user data, a change in location data associated with the end user device, user interaction with a user interface on the end user device, and the like.

The shot strategy generation systemmay be implemented in one or more computing devices for automatically processing course and user data and executing a shot strategy selection procedure. The shot strategy generation system(or individual components thereof not shown in) may be implemented on one or more physical server computing devices and/or on one or more physical client or end user computing devices. In some implementations, the shot strategy generation system(or individual components thereof) may be implemented on one or more host devices, such as blade servers, midrange computing devices, mainframe computers, desktop computers, mobile computers, tablet computers, smartphones, wearable computers, or any other computing device configured to provide computing services and resources, such as obtaining, storing, evaluating, selecting, and displaying shot strategies.

In some implementations, the features and services provided by the shot strategy generation systemmay be implemented as web services consumable via one or more communication networks (e.g., the network). In further implementations, the shot strategy generation system(or individual components thereof) is provided by one or more virtual machines implemented in a hosted computing environment. The hosted computing environment may include one or more rapidly provisioned and released computing resources, such as computing devices, networking devices, and/or storage devices. In additional implementations, the shot strategy generation system(or individual components thereof) is provided by one or more client or end user computing devices.

In some implementations, the shot strategy generation systemmay be a part of a cloud provider network (e.g., a “cloud”), which may correspond to a pool of network-accessible computing resources (such as compute, storage, and networking resources, applications, and services), which may be virtualized or bare-metal. The cloud can provide convenient, on-demand network access to a shared pool of configurable computing resources that can be programmatically provisioned and released in response to customer commands. These resources can be dynamically provisioned and reconfigured to adjust to provide various services, such as automatically evaluating a large number of shot strategies relative to course data and golfer data associated with a particular game of golf or training session. The resources can be provisioned and configured to execute one or more machine learning model(s) and shot strategy selection techniques as disclosed in the present disclosure. The computing services provided by the cloud that may include the shot strategy generation systemcan thus be considered as both the applications delivered as services over a publicly accessible network (e.g., the Internet, a cellular communication network) and the hardware and software in cloud provider data centers that provide those services.

End user devicesmay communicate with or interact with the shot strategy generation systemvia various interfaces such as application programming interfaces (APIs). These APIs can be accessed locally and/or as part of cloud-based services. In some implementations, the shot strategy generation systemmay interact with the end user devicesthrough one or more user interfaces, command-line interfaces (CLI), application programing interfaces (API), and/or other programmatic interfaces for requesting actions or services, such as receiving a request to select a preferred shot strategy from the end user devicesor presenting results of the shot strategy selection procedure to the end user devices.

Various examples of end user devicesare shown in, including a desktop computer, laptop, and a mobile phone, each provided by way of illustration. In general, the end user devicescan be any computing device such as a desktop, laptop or tablet computer, personal computer, wearable computer, server, personal digital assistant (PDA), hybrid PDA/mobile phone, mobile phone, electronic book reader, set-top box, voice command device, camera, digital media player, speaker, range finder, virtual reality headset, mixed reality headset, and the like.

In some implementations, the networkmay include any wired network, wireless network, or combination thereof. For example, the networkmay be a personal area network, local area network, wide area network, over-the-air broadcast network (e.g., for radio or television), cable network, satellite network, cellular telephone network, or combination thereof. As a further example, the networkmay be a publicly accessible network of linked networks, possibly operated by various distinct parties, such as the Internet.

In some implementations, at least some parts of the networkmay be a private or semi-private network, such as a corporate or university intranet. The networkmay include one or more wireless networks, such as a Global System for Mobile Communications (GSM) network, a Code Division Multiple Access (CDMA) network, a Long Term Evolution (LTE) network, or any other type of wireless network. The networkcan use protocols and components for communicating via the Internet or any of the other aforementioned types of networks. For example, the protocols used by the networkmay include Hypertext Transfer Protocol (HTTP), HTTP Secure (HTTPS), Message Queue Telemetry Transport (MQTT), Constrained Application Protocol (CoAP), and the like. Protocols and components for communicating via the Internet or any of the other aforementioned types of communication networks are well known to those skilled in the art and, thus, are not described in more detail herein.

In some implementations, the shot strategy generation systemmay access course data stored in the course data store(s)via the network. The course data store(s)may store course data that can include, for example, a polygon representation of one or more holes or course segments, short side data associated with the one or more holes, and/or geospatial information about the one or more holes. Course data may relate to a physical course existing in a physical location, or to a virtual course that may or may not have an associated course in a physical location. The polygon representation can include data about topographical characteristics (e.g., locations and sizes of fairways, rough, hazards, bunker regions, tee shot boxes, and greens), putting green characteristics (e.g., green speed), elevation characteristics (e.g. sidehill, uphill, and downhill), hole boundary data (e.g., out-of-bounds regions), and/or vegetation characteristics (e.g., vegetation types, dimensions, densities, and locations). As illustrated in, end user devicesand/or the network computing systemmay also access the course data store(s)via various interfaces such as application programming interfaces (APIs) and/or cloud-based services.

In some implementations, a polygon representation can correspond to one or more tangible, physical holes or course segments. For example, a polygon representation may have been generated based on geographic, spatial, LIDAR, image, or other information representing an existing course, green, hole, or course segment. The polygon representation may have been modified from an initial polygon representation (e.g., in response to change in the physical characteristics of the course segment represented by the initial polygon representation, to correct a measurement error, etc.)

In some implementations, a polygon representation can correspond to one or more virtual holes or virtual segments. The virtual holes or virtual segments can be wholly digital, with no direct tangible, physical holes or course segments which are being represented. For example, a virtual hole or course segment may be designed in a digital format (e.g., as a three-dimensional model, point cloud mesh, etc.). A polygon representation may then be generated from the virtual hole or course segment in the digital format. In another example, a virtual hole or course segment may have been generated as a polygon representation. A virtual hole or course segment may, for example, be incorporated as part of a golf simulator, video game, virtual reality game, movie, image, or a passive or interactive virtual environment.

In some implementations, the course data store(s)that store course data may be any computer-readable storage medium and/or device (or collection of data storage mediums and/or devices). Course data may be generated by the network computing system, the end user devices, the shot strategy generation system, and/or other computing systems or devices not illustrated in. Examples of the course data store(s)include, but are not limited to, optical disks (e.g., CD-ROM, DVD-ROM, and the like), magnetic disks (e.g., hard disks, floppy disks, and the like), memory circuits (e.g., solid state drives, random-access memory (RAM), and the like), and/or other memory or storage devices. In some examples, the course data store(s)and/or the shot strategy generation systemmay be parts of a hosted storage environment that includes a collection of physical data storage devices that may be remotely accessible and may be rapidly provisioned as needed (commonly referred to as “cloud” storage).