US20180318688A1

US20180318688A1 - Golf instruction method, apparatus and analytics platform

Info

Publication number: US20180318688A1
Application number: US15/969,617
Authority: US
Inventors: Mark Colangelo
Original assignee: Individual
Current assignee: Individual
Priority date: 2017-05-03
Filing date: 2018-05-02
Publication date: 2018-11-08
Also published as: US11801433B2; US20200391094A1; US20230264092A9; US20210252367A1

Abstract

A golf instruction method, apparatus and analytics platform for determining the state of a putted golf ball relative to a target using information about the golf ball and external variables and remotely displaying the same comprising a base unit, a mobile device or terminal, and an associated computer program including an analytics engine for analyzing a user's performance. The computer program determines whether and how a putted golf ball has missed said target and then stores, aggregates and displays that information for a user. The analytics engine is operable to aggregate data, analyze data and then correlate a user's actual skill proficiency to ball performance.

Description

CROSS REFERENCE TO RELATED APPLICATION

This application is related to U.S. Provisional Patent Application Ser. No. 62/501,015 filed May 3, 2017 entitled METHOD AND APPARATUS FOR DETERMINING STATE OF A PUTTED GOLF BALL RELATIVE TO A TARGET AND REMOTELY DISPLAYING SAME and U.S. Provisional Patent Application Ser. No. 62/635,229 filed Feb. 26, 2018 entitled GOLF INSTRUCTION METHOD, APPARATUS AND ANALYTICS PLATFORM.

TECHNICAL FIELD

The present invention relates to sporting equipment and includes a combination of a base unit, including a mobile device or terminal, a mobile application within a mobile device or terminal having a graphical user interface (GUI) for displaying formatted data, a data collection and storage component, and an analysis engine for capturing, aggregating, synthesizing, analyzing and processing data.

BACKGROUND OF THE INVENTION

“Putting” in golf refers to the practice of hitting a golf ball along the ground or “green” into a hole, an activity requiring practice to develop the feel for line, speed and accuracy. Putting is the most utilized aspect of the game of golf, considering it must be done on every hole to complete the hole. As such, about 45% of an average golfer's scores are putting strokes. Current products aim to assist golfers in alignment, best in practice stroke techniques, building muscle memory, distance control, reading greens, among other areas, all of which are primarily involved in the measure of and influence on the swing of the putter.
There are a variety of passive products aimed at enhancing a user's golf putting practice, among them floor mats and ramp constructions with actual or simulated holes. While ramp constructions with actual holes can provide a semi-realistic experience, they generally require a large space to set up and use. Floor mats which are flush with the ground can be more compact but rely on flat or shallow simulated target holes, making it sometimes difficult to tell if a shot was made or missed. In addition, because these are passive apparatuses, a user desiring some record of their putting practice must observe and record each shot manually. At best, there are products that simply count the number of strokes you make, but not in relation to the total number of attempts. Further, such conventional products do not have a means to collect, process and analyze putting statistical records so as to provide the user with actionable feedback for improving the user's performance. There exists no platform or solution in this domain that uses multivariate putting performance data or analysis which is based directly on data-based proficiency of putting a ball to a target; nor is there computer-driven modeling that correlates this performance data and analysis to produce instructional guidance specific to each user so as to inform a user to their specific low-performance areas to focus on improving their putting.
An active device could make use of electrical sensors to accurately and automatically determine the position and/or velocity of a putted golf ball with reference to a virtual hole. There are a variety of mechanical sensors which can be used to sense impact force and/or location, among them accelerometers which detect movement, vibration or impact; and piezoelectric sensors which produce an electric charge when deformed by a force. In addition, a variety of remote sensors exist which are able to determine the position and velocity of an object over a distance. These are broadly referred to as motion detectors. They include passive infrared sensors, which measure infrared light radiating from objects, microwave and ultrasonic motion detectors which emit microwave radiation and high-frequency sound waves, respectively, and then detect resulting reflections, and digital video cameras combined with a computer program which applies motion analysis to the generated videos, allowing for object tracking. This type of analysis is stateless in that once the data about the immediate putt has been analyzed and displayed, it is not retained in memory for later analysis with additional data.
Mobile devices or terminals such as smart phones, which are now widely-used, combine a microprocessor and computer memory with several methods of remote communication (e.g. a network connection, Wi-Fi, Bluetooth®, cellular or other wireless communication channel), a camera, and a user-interface which accepts and displays information. This makes them an excellent platform for interfacing with a base unit and objects. In addition, they provide storage of and access to information in a local, convenient location for a user.

SUMMARY OF THE INVENTION

The invention is an apparatus combining a base unit having a transceiver, the base unit with a user facing camera interfaced via a wireless or wired channel to a standalone mobile device or terminal, the mobile device or terminal operable to (i) receive video data from the base unit, and (ii) store, aggregate, process, format and display the processed, formatted data to the user at the mobile device or terminal. The invention is further operable to analyze the data using a variety of analysis techniques and present the analyzed data to the user in a format so as to allow the user to improve putting performance such as, but not limited to assisting golfers in understanding data and analytics of their performance results to identify specific putting strengths, and more importantly putting weaknesses. From this awareness of weaknesses, instruction is provided that addresses specific tendencies related to alignment, stroke techniques, muscle memory, distance control and reading greens which supports skill improvement in these areas.
The invention further comprises an apparatus having a base unit with transceiver, mobile application including a graphical user interface (GUI) and analysis engine which utilizes the aggregate data from the base unit for providing various analysis, such as long term putting made/miss performance statistics (stratified by distance, percentage putts made, surface slope or gradient, proximity (distance) from target (hole), compared to professional users' statistics, etc.) The analysis platform aspect of the invention is operable to collect, aggregate, synthesize, analyze and process data from the base unit and mobile application so as to identify specific low-performance putting statistics and provide the user with actionable feedback for improving these areas of user's performance. The processing and analysis of the data can occur in a processor in the base unit or in a processor in the mobile device or terminal or between both of them.
The invention uses, among other things, video data plus actual ball performance data. A variety of products exist that employ video alone for subjective assessment, meaning the user is only able to view the video to determine by self-diagnosis any abnormalities which are obvious to the unaided eye to correct. Other products exist that use video data in tandem with sensors that capture data about the user physical mechanics or the mechanics of the sporting equipment, such as the golf putter during user's use. Output data can be specific to the user or the equipment such as swing path, angles, swing speed; analytics regarding the proficiency of the user compared to best practices or professional metrics; feedback that contrasts the user to the best mechanics in the sport. The other products target relevant swing mechanics, but not the specific data of the putted ball or analytics of results of the putted ball.
In the invention, video recording is used to sense motion and cue the video; capture the ball (not swing data) from point of origin to end point; and determine spatial orientation of ball relative to target, including based on coordinates such as Cartesian, polar, or other 2-dimensional plotting system; end point: made/missed; missed: left/right, short/long; speed and velocity of the ball when impacting target compared to an acceptable speed range; spin of the ball; lift of the ball off club face; and a basic visual representation of the putting stroke, without data.
The analytics engine of the invention receives user positioning and motion, ball position, ball velocity, and ball acceleration data, hole location and gradient information, between varying locations, including user location and the hole location and performs a multivariate analysis of all variables captured. The data is, inter alia, conditioned, normalized, optimized and then processed, formatted and made available for storage and display on a mobile device or terminal through an application program, including via a mobile device application.
The invention further is operable to allow correlation of analytics to learning modules or instructional databases, such that it curates or matches instructional resources to users directly based on the user's ball data, performance and skill deficiencies.
The invention is further operable as an entertainment game that enables comparison of data for multiple users, such as for a golf team of a plurality of players engaging in a practice game; head-to-head competition either live on one device, or virtually across local or wide area wireless or wired networks and simulated game play against professional golf statistics or against a user's historical stat averages.
The invention includes a base unit, or a base unit comprising, a protective housing, an actual or simulated target golf hole, a means for sensing the position and velocity of a golf ball in proximity to said base unit relative to said target, a microprocessor and associated signal conditioning electronics, a computer program being executed by said microprocessor, means for communicating data to and from a standalone interfacing device if applicable (e.g. a network connection, Bluetooth®, Wi-Fi, cellular or other wired or wireless communication channel), and a power source; and a standalone interfacing device (e.g. a personal computer, tablet, smart phone or user equipment or user terminal) comprising a microprocessor, a computer program comprising instructions for being executed by said microprocessor, a means for accepting user input, a means for storing data, a means for communicating data to and from said base unit, a means for displaying information, and a power source; and an analytics engine comprised of software running on a processor for synthesizing and analyzing data from the base unit and mobile application so as to provide the user with actionable suggestions for improving the user's performance. The collection, storage, processing and analysis of the data can occur in a processor in the base unit or in a processor in the mobile device or terminal. Alternatively, the data can be sent via the mobile device or terminal to a “cloud” provider or hosted server on a wide area network for storage, processing and analysis, the analyzed data then formatted for display back at the mobile device or terminal.
To those skilled in the art to which this invention relates, many changes in construction and widely differing embodiments and applications of the invention will suggest themselves without departing from the scope of the invention as defined herein. The disclosures and the descriptions herein are purely illustrative and are not intended to be in any sense limiting.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the present invention including the features, advantages and embodiments, reference is made to the following detailed description along with accompanying Figures, in which:

FIG. 1 is an illustration of a first form factor of a base unit used in the invention;

FIG. 2 is an additional view of the form factor of a base unit of FIG. 1;

FIG. 3 is a second illustration of a form factor of a base unit used in the invention;

FIG. 4 is an additional view of the form factor of a base unit of FIG. 3;

FIG. 5 is an illustration explaining data analysis and the practical value of analyzing an increasing number (1, 2, 3) of variables used in the analysis engine, the data correlated to the tables of FIGS. 6A to 6F;

FIGS. 6A to 6F are tables with example data and results outputted by the invention, specifically a subset of statistical data for putts missed short, short-left, and short-right;

FIG. 7 is a flow chart of the steps of the invention showing the ball results based on unaided video;

FIG. 8A is a flow chart of the steps of the invention showing the ball results based on the unaided video as informed by an analysis by the analytics engine; and

FIG. 8B is a table with example data and results showing a subset of statistical data for putts recorded for a right-to-left slope following the flow chart of FIG. 8A.

DETAILED DESCRIPTION

While the making and using of the disclosed embodiments of the present invention is discussed in detail below, it should be appreciated that the present invention provides many applicable inventive concepts which can be embodied in a wide variety of contexts. Some features of the preferred embodiments shown and discussed may be simplified or exaggerated for illustrating the principles of the invention.

Form Factors and Functionalities of the Invention

The invention, as more fully described herein, is a golf instruction apparatus and analytics platform embodied in one of a base unit alone, a base unit having certain functionalities in combination with a mobile device or terminal, a base unit having certain functionalities in combination with modules in a remote server; a base unit having certain functionalities in combination with modules in a mobile device or terminal and a remote server, a mobile device or terminal alone; or a mobile device or terminal having certain functionalities in combination with modules in a remote server.
In certain such combinations above, the base unit housing comprises a form factor, two examples of which are seen in FIGS. 1-2 and 3-4 respectively. Referring to FIGS. 1 and 3 respectively, the base unit 100, 300 incorporates an optical sensing means 101, 301, a microprocessor and a wired or wireless transceiver. The interfacing device to the base unit 100, 300 is preferably a mobile device or terminal such as a smart phone, running an associated computer program.
The base unit 100 comprises a housing unit 102 being made of a rigid material, preferably a hard plastic such as polyvinyl chloride. Within the housing unit is a base unit microprocessor (S-MPC) running camera driver firmware, means for remotely communicating video data to and from a standalone interfacing device, and an infrared or optical sensing means such as a camera or recording video for playback in mobile application, and a power source or sources (e.g. a battery).
The base unit 100, 300 incorporates an optical sensing means 101, 301, a microprocessor and a wired or wireless transceiver. The interfacing device to the base unit 100, 300 is preferably a mobile device or terminal such as a smart phone, running an associated computer program. As seen in FIG. 2, a retractable support 200 is provided for holding a mobile device or terminal that interfaces to the base unit. As seen in FIG. 4, a folding support 400 is provided for supporting a mobile device or terminal that interfaces with the base unit. In other embodiments, all functionalities of the invention are included in a mobile device or terminal and the base unit is a passive platform for supporting the mobile device or terminal.
The invention further comprises a mobile application running on a mobile device or terminal, such as a smartphone, the mobile device or terminal comprising a device microprocessor (D-MPC), a device computer program being executed by said D-MPC, a means for accepting user input, a means for storing data, a means for communicating video data to and from said base unit (e.g. a wired or wireless network connection, Bluetooth®, cellular or other wired or wireless communication channel), a means for displaying information including results of the putt and the video playback, and a power source(s). The device computer program comprises instructions being stored on a computer readable medium to be executed by a microprocessor to perform the following functions: querying for and accepting user input (e.g. putt length, planned number of shots in current session), communicating with the D-MPC in said base unit to trigger certain events, manipulating video data, and formatting said video data for output via said means for displaying information.
In an embodiment, said means for sensing positioning and velocity of a putted golf ball in close proximity to user-facing side of said housing comprises one or more remote motion detectors, e.g. an optical sensor such as a passive infrared detector or video camera, or an active radiating sensor such as an ultrasonic motion detector or a microwave motion detector. In one aspect of the invention, the camera on a mobile device or terminal, such as a smart phone, is used to capture the video, the camera being positioned on the base unit and in another aspect of the invention, the camera is integrated into the base unit and the video is communicated to from the base unit, along with sensor data, to the mobile device or terminal.
In a further embodiment, said means for sensing positioning and velocity of a putted golf ball in close proximity to user-facing side of said housing comprises 3 remote motion detectors in the base unit located to the left, right, and rear respectively of a simulated golf hole. The invention further comprises a display for displaying the output results.
The golf instruction apparatus and analytics platform can include a base unit coupled to a mobile device or terminal over a wired or WiFi, Bluetooth®, cellular or similar radio spectrum link. The base unit further comprises a housing being made of a rigid material, preferably a hard plastic such as polyvinyl chloride, a sensing circuit for sensing position and velocity of a putted golf ball in close proximity to a user-facing side of said housing, a microprocessor running base unit firmware for driving camera input/output (I/O), a signal conditioning circuit for conditioning sensor data such that it is readable by said microprocessor, a transceiver for remotely communicating video data to a mobile terminal or device, a recording module for video recording user's putting stroke from the vantage point of simulated target hole; and a power source.

Analytics Engine

The invention includes an analytics engine as more fully described below. As used herein, “multivariate” generally means the use, analysis and correlation of 2 or more different data points in combination. “Multivariate putting performance data and analysis” means the numerical and mathematical measures of a golf ball putted towards a golf hole or simulated target, so that any combination of variables collected may be cross-analyzed with any others to provide more granular feedback on specific skill areas needing improvement. For example, multivariate analysis may determine a user's worst performance skill segment is “when putting from 6′ from target with Left to Right slope, user misses right 70% and short 83%.” Contrast this with a single variable that shows a user on average “misses right 70%” over all putts recorded. The more meaningful and accurate feedback is the multivariate analysis because in that example, the Left to Right slope variable imparts an otherwise non-obvious influence on putting performance since more balls on right sloping terrain that are putted short (83%) will fall right of the hole due to gravity. So, the multivariate analysis points to the high miss percentage most likely being a speed/distance issue since 83% were short of the target. But if only the single variable were reviewed, in this case (70% right misses), it can mislead the user about the cause of the missed putts—for example that perhaps they are pushing the ball right.
The analytics engine of the invention includes application software executed by at least one processor that enables the analysis of both individual data points and multivariate analysis of any number of variables. Such data analysis provides broader and more practical feedback since there are inherent interdependencies and influences across each of the different variables. The breadth of the invention's analysis operations can be best understood this way: there are at least 3 primary categories of variables for “made” putts, and at least 8 for “missed” putts, with over 70 total subcategories. This enables thousands of combinations of multivariate analysis of factual ball-related performance data.
The analytics engine of the invention collects video data either in-device or via customized mobile phone camera/software and integrates it with aggregate user performance data and performs a multivariate analysis thereof so as to provide actionable feedback to a user. More specifically, the invention is a method and apparatus employing video capture to record users putting stroke and determine the state of a putted golf ball relative to a target, and remotely displaying the same comprising a camera(s), base unit, a standalone interfacing device, and an associated computer program.
In an embodiment, the golf instruction apparatus and analytics platform invention herein comprises a base unit or mobile device or terminal operable to receive stimuli related to the state of a golf ball, a processor coupled to the base unit operable to digitize said stimuli into input data; and an analytics engine comprising software in the form of computer instructions stored on a computer readable medium and executed by a processor, the analytics engine operable to (i) correlate the input data with at least one external variable, (ii) using an algorithm, process the correlated input data and at least one external variable to obtain a result and (iii) output the result.
The invention further includes a data analytics engine that comprises software in the form of instructions to be executed by a processor. The invention, in the form of a system comprising the base unit, including at least one camera and/or sensor, and the mobile application executable on a mobile device or terminal, is operable to capture, aggregate, analyze, and then correlate a user's actual skill proficiency and ball performance during use. The invention is operable to capture video data from the base unit or mobile device or terminal, save the data in a local memory unit therein or in a central server via a network connection, the data segregated by user for individual user access and analyze user data to provide the user practical information.
Referring now to FIG. 5 and FIGS. 6A to 6F, an example of how data is useful for self-guided instruction. Referring to FIG. 5, a single variable analysis 501 shows percentages based on the number of putts taken and made and number of putts taken and missed. The percentages correlate to cells 605 and 606 in FIG. 6D.
A two variable analysis 502 shows percentages based on the number of putts taken and missed left side versus number of putts taken and missed right side. The percentages correlate to cells 607 and 608 in FIG. 6D.
A two variable analysis 503 shows percentages based on the number of putts taken and missed short versus number of putts taken and missed long. The percentages correlate to cells 610 in FIG. 6E, and 612 in FIG. 6F.
A three plus variable analysis 504 shows percentages based on the number of putts taken and missed short and left. The percentage correlates to cell 611 in FIG. 6E.
A three plus variable analysis 505 shows percentages based on the number of putts missed left and short for left-to-right sloping attempts. The percentage correlates to cell 604 in FIG. 6B.
A three plus variable analysis 506 shows percentages based on the number of putts missed left and short for left-to-right sloping attempts at 2-4 feet, for example, from target or hole. The percentage correlates to cell 602 in FIG. 6B.
A three plus variable analysis 507 shows percentages based on the number of putts missed left and short for left-to-right sloping attempts at 2-4 feet in the most recent 30 days.
FIG. 5 further demonstrates how the multivariate analysis provides practical value to the user in the form of feedback 508. In this example, the data from FIG. 6D (609), FIG. 6A (601) and FIG. 6B (603) is fed into the analysis
The location of a missed putt is determined by orientation to the hole by coordinates wherein the coordinates are Cartesian, polar, or some other coordinate system that plots the ball's location in proximity to the target. Statistical percentages can also be provided based on the user's proficiency with “2-putting”, defined as putting the ball in the hole on the next (or 2^nd) putt attempt immediately after a missed putt.
Permutations of all of the above, further modified by the length of the putt, timespan of practice (i.e., last 30 days, last 120 days), gradient of slope between the beginning of the putt and the hole (e.g., level, left-to-right (LR), or right-to-left (RL)). Each set of variables related to a putt is referred to as a putt context. Similar putting contexts can be grouped and analyzed together by the analysis engine.
The analysis engine is operable to calculate a user's lowest performance segments or contexts as focus areas for practice. The analysis engine is further operable to provide trend data for each data class or context. For example, trend data may show that a user's putts made from less than 10 feet improved from 43 percent to 45 percent over last 90 days and 1,373 attempts.

Machine Learning Module

A further embodiment of the invention incorporates a machine learning module. The stimuli (referred to as a “feature” in machine learning as described herein) is at least one selected from the group consisting of location of the golf ball relative to a start position and an actual or virtual target or target hole, speed, velocity, acceleration and deceleration of the golf ball over a selected length of travel, golf ball force at any time prior to, and upon impacting or reaching the actual or virtual target or target hole, spin of the golf ball at any time prior to, and upon reaching or impacting the actual or virtual target or target hole and lift of the golf ball off a club face. The at least one external variable (also a “feature”) is one selected from the group consisting of user position, user motion, club position, club motion, actual or virtual target location, actual or virtual target hole location, gradient between selected locations, wind velocity, wind acceleration, characteristics of a putting surface, including attributed speed based on the measurement of static and dynamic co-efficient of friction and obstacles. At least one of the at least one external variables are previously stored in a memory accessible by the processor. Alternatively, or in addition, at least one of the at least one external variables are sensed, collected and stored in a memory accessible by the processor contemporaneously with the receipt of stimuli related to the state of the golf ball. The external variables comprise spatial orientation of ball relative to a target, speed and velocity of the ball, when impacting target compared to an acceptable speed range, spin of the ball, lift of the ball off club face, and a visual representation of the putting stroke.
According to Wikipedia, machine learning uses statistical techniques to give computer systems the ability to “learn” (i.e., progressively improve performance on a specific task) with data, without being explicitly programmed. Machine learning is closely related to (and often overlaps with) computational statistics, which also focuses on prediction-making through the use of computers. It has strong ties to mathematical optimization, which delivers methods, theory and application domains to the field. Within the field of data analytics, machine learning is a method used to devise complex models and algorithms that lend themselves to prediction; in commercial use, this is known as predictive analytics. These analytical models allow researchers, data scientists, engineers, and analysts to “produce reliable, repeatable decisions and results” and uncover “hidden insights” through learning from historical relationships and trends in the data.
An algorithm which can be used with the invention to improve its performance and user suggestions is TensorFlow, a machine learning open source application. Using a neural network algorithm, over time the invention builds data sets of stimuli related to the state of a golf ball and external variables that act upon the golf ball and maps them to the final result of the golf ball. The data sets are thus training data sets that are improved each time a user uses the invention.
The data sets are stored as tabular data formatted as comma-separated values (CSV). The first line of a CSV is a header containing information about the dataset, including features, which are float numbers to hold information about the stimuli and the external variables and a label, which is a final outcome, result, prediction or suggestions. These data sets are parsed and combined to provide a single tensor, which is the label, using a model coded in Python. A model is the relationship between “features” and the “label”. Traditional programming techniques (for example, many conditional statements) can be used to create a model. However, this requires an analysis of datasets over a long period of time to determine the relationships between golf ball measurements, external variables and the final result. The machine learning approach of this invention determines the model. Neural networks can find complex relationships between features and the label. It is a highly-structured graph, organized into one or more hidden layers.
Each hidden layer consists of one or more neurons. There are several categories of neural networks and an embodiment of the invention uses a dense, or fully-connected neural network: the neurons in one layer receive input connections from every neuron in the previous layer. When the model is trained and then receives stimuli and external variables data, it yields predictions and suggestions to the user.
Training is the stage of machine learning when the model is gradually optimized, or the model learns the dataset. The invention learns enough about the structure of the training dataset to make predictions about unseen data. The invention can implement either supervised machine learning where the model is trained from examples that contain labels or unsupervised machine learning where the examples don't contain labels. Instead, the model typically finds patterns among the features (stimuli and external factors).
The invention uses a training loop to feed dataset examples into the model to help it make better predictions. Using TensorFlow, the invention implements a code block to set up the training steps:
(1) Iterate each epoch. An epoch is one pass through the dataset.
(2) Within an epoch, iterate over each example in the training Dataset grabbing its features (x) and label (y).
(3) Using the example's features, make a prediction and compare it with the label. Measure the inaccuracy of the prediction and use that to calculate the model's loss and gradients.
(4) Use an optimizer to update the model's variables.
(5) Keep track of statistics for visualization.
(6) Repeat for each epoch.
Over time, the invention is trained so that when it receives certain combinations of stimuli related to the state of a golf ball and external variables, it provides a certain output or result comprising user performance.
The output result of the machine learning module comprises a user's putting performance relating to user and club alignment, user and club position, stroke technique, muscle memory and distance control (each being a “label”) as further described herein.
Image Capture. Processing and Analysis Referring to FIG. 8A, the platform (base unit and/or mobile device or terminal) is powered on 801, receives user preference inputs 802, mode inputs 803, such as practice mode 804 If game mode 805 is selected, user selections 806 are prompted related to the date, putting length, number of putts and putting green slope. These are stored in the database 813. Once a put 807 is made, the platform receives video data 808.
In one aspect of the invention, the camera on a mobile device or terminal, such as a smart phone, is used to capture the video data 808, the camera being positioned on the base unit and in another aspect of the invention, the camera is integrated into the base unit and the video 808 is communicated from the base unit, along with sensor data, to the mobile device or terminal.
Another embodiment utilizes a mobile device alone without a separate base unit, and such mobile device is placed in a stand or platform so as to both capture, process and analyze video data with statistical data in the multivariate analysis, or to capture, partially process and communicate data to the cloud (a distributed or remote processor or server networked to the mobile device or terminal), whereupon such data is further processed and made available (pushed or pulled) to the user locally or remotely. In this example embodiment, the video data is used to determine which putts are made 809 and which are missed left 810A, and of those, which are short 811A and long 811B and those which are missed right 810B, and of those, which are short 812A and which are long 812B.
The components of the invention include several video aspects. In one aspect, an unaided video camera records the user's putting stroke for subjective self-assessment by user or sharing externally through any conventional means of communication such as email, text, and social media. Video review is available in real-time and without tethered/coupled motion capture elements either on the putter or user, sensor data, other inputs, or in-device validations.
A further video aspect of the invention comprises image/motion capture. This aspect measures factual user performance of the putted golf ball in relation to a target during use; records result and metrics such as ball accuracy, path, spin, lift, speed, spatial orientation. None of above metrics are necessarily based on calculated or predicted outcomes nor based on any sensor-based measures of equipment or user motion. Video analysis and/or image processing is used to determine metrics and analyze actual ball data from which to draw meaningful conclusions that a user can use to improve putting skills.
A further video aspect of the invention is the ability to auto-trim video using camera 701. In this aspect, a video recording start 702 is initiated by either a recognized vocal/audible cue 711, visual motion detection within the frame of the camera 710, vibration, or mechanical trigger. The length of the video to be saved is automatically edited by starting 703 and stopping 704 to cover the cue plus 5 to 8 seconds, depending on the length of the putt input by the user.
As noted in FIG. 8A, prior to starting a batch of putts, a user inputs the length, slope and number of putts in the batch they desire to take, shown as step 806. Predetermined time periods are then coded into a processor such that at shorter putt lengths, the video clips will be shorter than longer lengths in order to capture the full period of interest while avoiding the recording of an excessive amount of video data. For example, a user that inputs a batch of 3 foot putts might require video lengths of cue plus 3 seconds, whereas a batch of 15 foot putts might require video lengths of cue plus 6 seconds. The invention performs video trimming using triggering method(s) for start-recording 702 and stop-recording 704. In contrast, conventional techniques use other data, such as non-audio, and then correlate it with audio data to confirm an event, whereas the invention initially recognizes the prospective event using the above audio or other triggers.
Further, the invention uses video clips only for subjective self-analysis and “tagged” to allow the user to review strokes based on an understanding of the data. For example, assume a user performs 20 putts of which the user makes 14 and misses 6, 5 of which are left misses. The user may wish to review the 5 left miss video clips to observe issues that caused the misses so as to inform what aspects are in need of correction. To do this, the videos are simply tagged with the related result, “made” “missed left” The invention does not formulate comparison data, rather it uniquely captures outcome data that is analyzed as seen in steps 706, 707, 708 and 814 by the analytics engine for immediate performance feedback as seen in FIGS. 5, 6A-6F, and long-term trending statistics that demonstrate the true impact of the training device. This is in contrast to conventional systems which use data and video to compare a trajectory referenced against digital markers to the desired trajectory.
The invention, in operation, is as follows according to an embodiment seen in FIGS. 7 and 8A-8B: The user putts 807 standard, non-sensor golf ball towards a golf hole or simulated golf hole target. The base unit or mobile device or terminal is positioned such that the target is in the front thereof, in the anticipated path of the golf ball (behind the target or hole). The base unit (on which a mobile device or terminal can be temporarily integrated) or the mobile device or terminal alone, detects motion (702, 710 or 711) and initiates the video recording 703 of user's stroke, as well as registers various aspects related to, among other things, the path, speed and ending location of the ball in proximity to the target. The target can be represented by an actual or virtual hole or marker. The video recording whether from a camera in the base unit or standalone mobile device or terminal, may be motion-activated by the video camera or by the user remotely using sound or vibration sensors, other sensor types, a detachable mechanical switch, or verbal command. In one aspect of the invention, the camera on a mobile device or terminal, such as a smart phone, is used to capture the video, the camera being positioned on the base unit or independent stand, and in another aspect of the invention, the camera is integrated into the base unit and the video is communicated to from the base unit, along with sensor data, to the mobile device or terminal or in some aspects, directly to the cloud via a Wi-Fi or cellular link.
The invention is operable to store results of user putts in batches, the quantity chosen by the user before each use. So, all putts taken in a batch are stored, analyzed and displayed to the user on the mobile device or terminal in near-real time. For example, as seen in FIG. 8B (815) for a batch of 50 putts, a user makes 45 and misses 5, 2 of which are missed left and 3 right. The mobile device or terminal would then display (816) the percentages of those Made, Missed; Missed Right, and Missed Left.
The invention then aggregates factual data and then calculates the user's actual performance. Data is captured locally based on the actual movement of the ball and is not extrapolated by a) utilizing a remote database of archived reference data, or b) predictive calculation based on stroke technique or characteristics other than those related directly to the ball.

Additional Aspect of the Invention

The analytics engine of the invention is further operable to correlate performance results to instructional resources so as to provide curated instructional resources to user for purchase and compare user performance data against professional golf putting statistics.
The invention provides actionable feedback to the user to enable self-diagnosis and device-generated guidance on areas of practice by providing actual performance data along with video footage of stroke at ground level to observe set up or stroke flaws.
In addition to being a training system and platform, the invention is also an entertainment platform as it is operable to permit multiple players to compete against each other either in alternating player format or virtually across the internet using network connectivity. The invention is further operable as a standalone, unaided video analysis system.
The invention provides results based on ball movement, regardless of the stroke or swing used. As a result, the invention provides user feedback that is based on performance whereas conventional training systems provide feedback based on technique. Hence, a user with an incorrect stroke will benefit from the invention's accurate measurement of factual results. In contrast, conventional systems incorporate sensor data and video to provide feedback to the user specific to aspects of their technique even if improper technique may provide better actual performance results for the user. The invention does not provide data related to appropriate or desired techniques for using equipment. The invention is further differentiable from conventional systems in that the input and output to the invention is based on image and data capture of actual ball movement after being engaged by the user whereas conventional systems use calculated or other predictive methods based on sensors attached to equipment or user.
The invention further incorporates an analytics engine operable to perform multivariate analysis of actual performance results for instructional purposes, whereas conventional systems provide predictive analysis of stroke characteristics. In contrast, the invention compares actual putting performance against the actual putting performance of professional golfers' average putting performance, without regard for technical data comparison.
The invention does not require any visual markers, sensors, other aids or objects on the equipment or user in order to recognize and process video to record the ball or user. Video is taken from the vantage point of the base unit or stand-alone mobile device facing toward user.
The recording is transferred from the base unit over a wireless protocol such as Wi-Fi or Bluetooth to the mobile device or terminal and is accessed 816 at the mobile device or terminal via a mobile application.
The invention is operable to transmit data over a wired or wireless network to the cloud comprising public and/or private server(s), and all historical performance data and video data is accessible from such server(s) in the mobile application via the mobile device or terminal. In an aspect of the invention, the user can log into the mobile application to review their performance statistics, analysis, and compare such statistics against the average statistics of professional golfers. The invention is further operable to communicate data to the mobile application via the Bluetooth wireless protocol, the Wi-Fi protocol or by plugging the mobile device or terminal using a USB cord into a computer that is communicates to the network. In operation, a user can review the video on their mobile application to determine issues in set-up or stroke. By using the personal statistics and the video review, users can identify which corrections are needed to improve putting performance.
The invention provides the following advantages. It is operable as a factual assessment tool that reflects putting skill performance based on actual results of putting a ball in relation to a target; it is operable as a game that can be played by multiple players against each other either in alternating player format or virtually across the internet assuming players have connectivity. The video component enables visual self-learning of stroke and ball characteristics. The longitudinal data analysis enables device-driven learning and curated instructional feedback. The use of analytics by the invention is operable to generate and serve alerts to the user of the primary areas where practice is required. The invention enhances the user experience compared to conventional systems that only allow user to see success on a stroke by stroke basis. The application embodied in the invention can include different levels of instruction as the user reaches skill plateau and the invention provides 24/7 access via the mobile application to video and performance data.
A method and apparatus in accordance with the present invention for determining the state of a putted golf ball relative to a target and remotely displaying the same generally comprises a base unit and a standalone interfacing device and an analysis engine operable to provide actionable feedback to the user.
In an embodiment, at least one of the input data, external variables, machine learning algorithm and output are transmitted and stored in a remote or distributed server. In a further embodiment, the input data, external variables, algorithm and output are collected, transmitted and stored in a remote or distributed server and machine learning and artificial intelligence techniques are applied thereto to provide insights into the user's past, current and predicted performance.
The data generated and stored by the firmware and device computer program are maintained to be acted upon by analytics programs to enable factual conclusions to be drawn about the users putting skills, both historically as compared to the user's prior putts (all or a subset thereof) and as compared to putts made by others, such as by professional golfers.
In an embodiment, said means for sensing position and velocity of a putted golf ball in close proximity to user-facing side of said base unit housing comprises one or more optical sensors.
The analytics engine of the invention is further operable with a virtual reality headset. In such embodiment, the only physical device may be a sensor handle that mimics a putter or golf club. An extension to the golf head, the golf ball, the target hole and the putting green are all or a portion thereof, are virtual. The firmware and device computer program are configured to execute algorithms to detect the velocity of the virtual ball to make a determination if the virtual golf ball would have entered a virtual target hole. If it determines that a golf ball had too much velocity, it can further determine how far past the virtual target hole the virtual golf ball progressed. Further, the user can input virtual parameters into the platform, such as the elevation of the hole, obstacles between the user and the hole and wind speed and direction and take such factors into account when determining if the golf ball went into or to the left or right of the hole.
To those skilled in the art to which this invention relates, many changes in construction and widely differing embodiments and applications of the invention will suggest themselves without departing from the scope of the invention as defined herein. The disclosures and the descriptions herein are purely illustrative and are not intended to be in any sense limiting. cm I claim:

Claims

1. A golf instruction apparatus and analytics platform, comprising:

a base unit operable to receive stimuli related to the state of a golf ball;

a processor coupled to the base unit operable to digitize said stimuli into input data;

an analytics engine comprising software in the form of computer instructions stored on a computer readable medium and executed by a processor, the analytics engine operable to (i) correlate the input data with at least one external variable, (ii) using an algorithm, process the correlated input data and at least one external variable to obtain a result and (iii) output the result.

2. The golf instruction apparatus and analytics platform of claim 1, wherein the stimuli is at least one selected from the group consisting of location of the golf ball relative to a start position and an actual or virtual target or target hole, speed, velocity, acceleration and deceleration of the golf ball over a selected length of travel, golf ball force at any time prior to, and upon impacting or reaching the actual or virtual target or target hole, spin of the golf ball at any time prior to, and upon reaching or impacting the actual or virtual target or target hole and lift of the golf ball off a club face.

3. The golf instruction apparatus and analytics platform of claim 1, wherein the at least one external variable is one selected from the group consisting of user position, actual or virtual target location, actual or virtual target hole location, gradient between selected locations, wind velocity, wind acceleration, characteristics of a putting surface, including attributed speed based on the measurement of static and dynamic co-efficient of friction and obstacles.

4. The golf instruction apparatus and analytics platform of claim 3, wherein at least one of the at least one external variables are previously stored in a memory accessible by the processor.

5. The golf instruction apparatus and analytics platform of claim 3, wherein at least one of the at least one external variables are sensed, collected and stored in a memory accessible by the processor contemporaneously with the receipt of stimuli related to the state of the golf ball.

6. The golf instruction apparatus and analytics platform of claim 3, wherein the external variables comprise spatial orientation of ball relative to a target, speed and velocity of the ball, when impacting target compared to an acceptable speed range, spin of the ball, lift of the ball off club face, and a visual representation of the putting stroke.

7. The golf instruction apparatus and analytics platform of claim 1, further comprising a display for displaying the output results.

8. The golf instruction apparatus and analytics platform of claim 1, wherein the stimuli is received in the base unit using a sensor selected from the group consisting of a camera, motion detector, infrared sensor, sonar, and acoustic sensor and an at least one external variable is received using a sensor selected from the group consisting of a keypad, mouse, touchscreen, camera, infrared sensor, sonar, acoustic sensor, wind vane, friction detector and motion detector.

9. The golf instruction apparatus and analytics platform of claim 1, wherein the output result comprises a user's putting performance results relating to putting line, speed, accuracy and distance control.

10. The golf instruction apparatus and analytics platform of claim 1, wherein the base unit comprises a mobile device or terminal.

11. The golf instruction apparatus and analytics platform of claim 1, wherein the base unit is coupled to a mobile device or terminal over a wired or WiFi, Bluetooth®, cellular or similar radio spectrum link.

12. The golf instruction apparatus and analytics platform of claim 1, wherein the base unit further comprises:

a housing being made of a rigid material, preferably a hard plastic such as polyvinyl chloride;

a sensing circuit for sensing position and velocity of a putted golf ball in close proximity to a user-facing side of said housing;

a microprocessor running base unit firmware for driving a camera and I/O;

a signal conditioning circuit for conditioning sensor data such that it is readable by said microprocessor;

a transceiver for remotely communicating video data to a mobile terminal or device;

a recording module for video recording user's putting stroke from the vantage point of simulated target hole; and

a power source.

13. The golf instruction apparatus and analytics platform of claim 1, wherein at least one of the input data, external variables, algorithm and output are transmitted and stored in a remote or distributed server.

14. The golf instruction apparatus and analytics platform of claim 1, wherein the input data, external variables, algorithm and output are collected, transmitted and stored in a remote or distributed server and machine learning and artificial intelligence techniques are applied thereto to provide insights into the user's past, current and predicted performance.

15. A mobile device and computer instructions, comprising:

a sensor operable to receive stimuli related to the state of a golf ball;

a microprocessor coupled to the sensor operable to digitize said stimuli into input data;

an analytics engine comprising software in the form of instructions stored on a computer readable medium and executed by a processor, the analytics engine operable to (i) correlate the input data with at least one external variable, (ii) using an algorithm, process the correlated data and at least one variable to obtain a result and (iii) output the result.

16. The mobile device and computer instructions of claim 15, wherein the analytics engine comprises instructions being stored on a computer readable medium to be executed by the microprocessor to query for and accept user input, communicate said user input with firmware in said microprocessor, receiving video data from a camera on the mobile device, correlate the video data with the at least one external variable; operate on such correlated video data and at least one external variable, obtain a result, format the result for output to a display on the mobile device.

17. The mobile device and computer instructions of claim 15, wherein at least one of the input data, external variables, algorithm and output are transmitted and stored in a remote or distributed server.

18. The mobile device and computer instructions of claim 15, wherein the input data, external variables, algorithm and output are collected, transmitted and stored in a remote or distributed server and machine learning and artificial intelligence techniques are applied thereto to provide insights into the user's past, current and predicted performance.

19. The mobile device and computer instructions of claim 15, wherein the stimuli is at least one selected from the group consisting of location of the golf ball relative to a start position and an actual or virtual target or target hole, speed, velocity or acceleration of the golf ball over a selected length of travel, golf ball force at any time prior to, and upon impacting or reaching the actual or virtual target or target hole, spin of the golf ball at any time prior to, and upon reaching or impacting the actual or virtual target or target hole and lift of the golf ball off a club face and wherein the at least one external variable is one selected from the group consisting of actual or virtual target location, actual or virtual target hole location, gradient between selected locations, wind velocity, wind acceleration, characteristics of a putting surface, including attributed speed based on the measurement of static and dynamic co-efficient of friction and obstacles.

20. The mobile device and computer instructions of claim 15, wherein at least one of the at least one external variables are sensed, collected and stored in a memory accessible by the processor contemporaneously with the receipt of stimuli related to the state of the golf ball.