GB2597336A

GB2597336A - System and method for coaching a swimmer

Info

Publication number: GB2597336A
Application number: GB2014106.5A
Authority: GB
Inventors: Young Adam; Newsome Paul
Original assignee: Swim Smooth Ltd
Current assignee: Swim Smooth Ltd
Priority date: 2020-09-08
Filing date: 2020-09-08
Publication date: 2022-01-26
Anticipated expiration: 2040-09-08
Also published as: WO2022053943A1; GB202014106D0; GB2597336B

Abstract

A system (100, fig 3A) for coaching a swimmer comprises a sensing unit (102, fig 3A), a memory unit (104, fig 3A), a processor (106, fig 3A), and a display unit (108, fig 3A). The sensing unit (102) is positioned on a wrist of the swimmer (200) and senses movement of the wrist during a swimming session; the processor (106) analyses the sensed movement of the wrist to determine stroke related data which is compared to stroke related parameters stored in memory (104) to derive at least one insight 406, and retrieve a set of instructions 408 from the memory unit 104 based on the derived at least one insight 406. The display unit 108 is arranged to display a graphical representation of the stroke related data and the retrieved set of instructions 408 in which the retrieved set of instructions 408 are configured to coach the swimmer 200. The sensor may be a tri-axis accelerometer or magnetic angular rate gravity (MARG) sensor.

Description

SYSTEM AND METHOD FOR COACHING A SWIMMER TECHNICAL FIELD

The present disclosure relates generally to swimming technique analysis; and more specifically, to a system and method for coaching a swimmer 5 for improving a stroke of the swimmer.

BACKGROUND

Swimming is a technique-oriented activity. Swimming in various environments such as competitive sport environments for e.g., in triathlons, or to complete fastest laps possible in a pool of water, or in non-competitive environments such as open water swimming requires different skillsets in terms of stroke technique as they present varying types and degrees of challenges to swimmers. In many cases, a swimmer experienced with swimming in one type of environment may not be aptly, or optimally, skilled and suited to swim in another type of environment.

It is hard for a swimmer to be aware of their stroke and what they are doing wrong.

Receiving lessons or coaching from a swim teacher or coach may be an option, but it is expensive, and the quality of that coaching varies greatly. Moreover, it is sometimes unfeasible to have a physical coach available at all times for correction of strokes made by the swimmer. For example, a coach for providing coaching for a particular type of swim stroke as desired to be learned, may not always be feasible. Regardless of the environment, it may be tedious for the coach as well to efficiently monitor and track the performance of the swimmer from one swimming session to another, as coaches themselves can also struggle to identify stroke faults, prioritise and correct them effectively with their swimmers. Furthermore, even with a coach, it may difficult for the swimmer to access any playback of his or her strokes in order to track and subsequently improve their performance over multiple swimming sessions, especially since filming of video is not allowed in many pools for privacy reasons.

In light of the aforementioned drawbacks, there exists a need of some means for coaching swimmers efficiently and in a cost-effective manner 5 for improving swimming techniques.

SUMMARY

The present disclosure seeks to provide a system for coaching a swimmer. The present disclosure also seeks to provide a method for coaching a swimmer. The present disclosure seeks to provide a solution io to the existing problem of providing a physical coach to a swimmer as it may become cost-intensive yet remain inefficient in that it may be tedious and cumbersome for the coach to monitor the performance of the swimmer during one, or even over multiple, swimming sessions. An aim of the present disclosure is to provide a solution that overcomes, at least partially, the aforementioned problems encountered in prior art, and to provide a system that is cost-effective yet efficient in coaching a swimmer.

In one aspect, the present disclosure provides a system for coaching a swimmer, wherein the system comprises: -a sensing unit positioned on a wrist of the swimmer, the sensing unit configured to sense and measure movement of the wrist during a swimming session; - a memory unit comprising stroke related parameters stored therein; - a processor in communication with the sensing unit and the memory 25 unit, the processor configured to: - analyse the sensed and measured movement of the wrist to determine stroke related data for the swimmer; - compare the determined stroke related data with the stroke related parameters to derive at least one insight; - retrieve a set of instructions from the memory unit based on the derived at least one insight; and -a display unit in communication with the processor, the display unit arranged to display a graphical representation of the stroke related data 5 and the retrieved set of instructions, wherein the retrieved set of instructions are configured to coach the swimmer.

Optionally, the retrieved set of instructions is configured to improve a stroke of the swimmer.

io Additionally, in the system for coaching a swimmer, the processor may be configured, for analysing the sensed and measured movement of the wrist to determine stroke related data for the swimmer, to: - log data related to movement of the wrist during the swimming session from the sensing unit; -identify start and end of stroke from the logged data using pitch rate criteria; - determine orientations from the logged data and the identified start and end of stroke; - calculate direction of average acceleration over stroke cycle using zo orientations; - determine positions in space from the logged data, based on the identified start and end of stroke and re-determine orientations based on the calculated direction of average acceleration; and - adjust initial velocity of stroke such that the start and end of 25 stroke is in same position.

Typically, "the same position" is considered relative to the position of the swimmer's body.

Optionally, the sensing unit comprises at least one of: a tri-axial accelerometer, a tri-axial gyroscope, and a magnetic angular rate gravity (MARG) sensor. Further optionally, the sensing unit comprises a tri-axial accelerometer and a tri-axial gyroscope.

Optionally, the processor is configured to determine orientations from the logged data obtained from the tri-axial gyroscope.

Optionally, the processor is configured to determine positions in space from the logged data obtained from the tri-axial accelerometer.

Optionally, the processor is configured to determine positions and orientations in space throughout the stroke.

Additionally, the display unit is configured to display: -a spatial distribution of the determined stroke related data using one of: a heat map or a tabulation; and -the retrieved set of instructions using at least one of: text, images and video.

Optionally, the determined stroke related data comprises of a calculated 15 angle and position of at least one arm of the swimmer.

Optionally, the derived at least one insight comprises two or more insights therein, and in which the processor is configured to determine a rank of each insight from the two or more insights using an insight hierarchy determination algorithm and retrieve the set of instructions to provide the retrieved set of instructions in an order of development to the swimmer based on the ranking of each insight from the two or more insights provided by the insight hierarchy determination algorithm.

Optionally, the memory unit is a remotely located server. Alternatively, the memory unit is a random-access memory unit located together with 25 at least one of: the sensing unit, the processor, or the display unit.

Optionally, the memory unit is a random-access memory unit located together with at least one of: the sensing unit, the processor, or the display unit.

In another aspect, the present disclosure provides a method for coaching 5 a swimmer. The method comprises: - positioning a sensing unit on a wrist of the swimmer; - sensing and measuring, using the sensing unit, movement of the wrist during a swimming session; - analysing, using a processor, the sensed and measured movement of 10 the wrist to determine stroke related data for the swimmer; - comparing, using the processor, the determined stroke related data with stroke related parameters to derive at least one insight; - retrieving, by the processor, a set of instructions from a memory unit based on the derived at least one insight; and -displaying, using a display unit, a graphical representation of the stroke related data and the retrieved set of instructions to coach the swimmer.

Additionally, for analysing the sensed and measured movement of the wrist to determine stroke related data for the swimmer, the method further comprises: -logging data related to movement of the wrist during the swimming session from the sensing unit; - identifying start and end of stroke from the logged data using pitch rate criteria; - determining orientations from the logged data and the identified 25 start and end of stroke; - calculating direction of average acceleration over stroke cycle using orientations; - determining positions in space from the logged data, based on the identified start and end of stroke and re-determining orientations 30 based on the calculated direction of average acceleration; and - adjusting initial velocity of stroke such that the start and end of stroke is in same position.

Additionally, the method comprises displaying: - a spatial distribution of the determined stroke related data (414) using one of: a heat map (404) or a tabulation; and - the retrieved set of instructions (408) using at least one of: text, 10 images and video.

Optionally, the determined stroke related data (414) comprises of a calculated angle and position of at least one arm (204) of the swimmer (200).

Optionally, the derived at least one insight (406) comprises two or more insights therein, and wherein the processor (106) is configured to determine a rank of each insight from the two or more insights using an insight hierarchy determination algorithm and retrieve the set of instructions (408) to provide the retrieved set of instructions in an order of development to the swimmer (200) based on the ranking of each insight from the two or more insights provided by the insight hierarchy determination algorithm.

Optionally, the sensing unit comprises at least one of: a tri-axial accelerometer, a tri-axial gyroscope, and a magnetic angular rate gravity (MARG) sensor. Further optionally, the sensing unit comprises a tri-axial 25 accelerometer and a tri-axial gyroscope.

Further optionally, the sensing unit comprises a tri-axial accelerometer and a tri-axial gyroscope.

Optionally, the processor determines orientations from the logged data obtained from the tri-axial gyroscope.

Optionally, the processor determines positions in space from the logged data obtained from the tri-axial accelerometer.

Optionally, the processor determines positions and orientations in space throughout the stroke.

Optionally, the method is implemented in a wrist-worn device.

Embodiments of the present disclosure substantially eliminate or at least partially address the aforementioned drawbacks typically encountered with use of a physical coach, and provide a cost-effective and efficient system for coaching the swimmer. The present disclosure solves the problems most swimmers have of not having a swimming coach. The present disclosure further solves problem of privacy and privacy regulations which in most cases forbid filming and film analysis, video sharing of swimmers between swimmers and coaches, and overcomes complexity, regulations time and resources needed to comply with regulation of video acquisition particularly for under-aged swimmers, which in most countries mean that videoing is not allowed for swimmers aged below 16 to 18 years old.

zo Additional aspects, advantages, features and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative embodiments construed in conjunction with the appended claims that follow.

It will be appreciated that features of the present disclosure are 25 susceptible to being combined in various combinations without departing from the scope of the present disclosure as defined by the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those skilled in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.

Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein: FIG. 1 is a schematic illustration of a system for coaching a swimmer, in accordance with an embodiment of the present disclosure; FIG. 2 is a diagrammatic illustration of implementation of the 15 system, by means of a watch worn on a wrist, by the swimmer for sensing and measuring movements of the wrist during a swimming session, in accordance with an embodiment of the present disclosure; FIGs. 3A-3C are diagrammatic illustrations of the system being implemented in the form of an exemplary watch and its implementation 20 for sensing and measuring movements of the wrist during a swimming session, in accordance with an embodiment of the present disclosure; FIG. 4A is an exemplary diagrammatic view of a display unit of the system displaying spatial distribution of a determined stroke related data using a heat map obtained from an analysis of one of the sensed and 25 measured movements shown as crossover, an insight derived from the determined stroke related data, and an instruction for improvement of the stroke for the swimmer upon completion of the swimming session, in accordance with an embodiment of the present disclosure; FIG. 4B is another exemplary diagrammatic view of the display unit 30 displaying a spatial distribution of another determined stroke related data, using another heat map, obtained from an analysis of another one of the sensed and measured movements shown as pull-through, another insight derived from the determined stroke related data, and another instruction for improvement of the stroke for the swimmer upon completion of the swimming session, in accordance with an embodiment of the present disclosure; FIG. 5 is another exemplary diagrammatic view of the display unit displaying a severity of improvement needed based on the historical stroke related data of the swimmer, in accordance with another lo embodiment of the present disclosure; FIG. 6A is a flowchart listing steps of a method for coaching a swimmer, in accordance with another embodiment of the present disclosure; FIG. 66 is a flowchart listing steps of a process for determining is stroke related data for the swimmer, in accordance with another embodiment of the present disclosure; FIGs. 7A-7C are illustrations of a wearable device with different graphical user interfaces (GUIs), in accordance with an exemplary implementation of the present disclosure; FIG. 8 is an illustration of a graph plotting pitch rate versus time for watch worn on left wrist, in accordance with an exemplary implementation of the present disclosure; FIGs. 9A-9E are exemplary diagrammatic views of the system device for providing various insights, in accordance with an embodiment 25 of the present disclosure; and FIGs. 10A and 10b provide exemplary positional model outputs and corresponding extrapolated positions for various exemplary strokes, in accordance with an embodiment of the present disclosure.

In the accompanying drawings, an underlined number is employed to 30 represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.

DETAILED DESCRIPTION OF EMBODIMENTS

The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.

The present disclosure is concerned about movement detection, transmission, calculation and correction of swimming techniques that are typical and necessary for a swimming stroke to make a body float and move in water. The disclosed system and method help a swimmer to improve his/her swimming techniques for enhancing health benefits, preventing injuries, all while saving time that may be lost by depending only on a physical swimming coach.

In overview, embodiments of the present disclosure are concerned with a system for coaching a swimmer. The embodiments of the present disclosure are also concerned with a method for coaching the swimmer. The present disclosure relates to a system and a method that can be used to coach a swimmer by analysing movements made by the swimmer during a swimming session for determining stroke related data therefrom and for deriving at least one insight from the analysed movements so that a set of instructions can be provided based on the derived at least one insight for coaching and hence, improving a stroke of the swimmer.

Prior to operation, the system, particularly -a memory unit of the system, is provided with one or more stroke related parameters. The stroke related parameters may comprise position and orientation data for the wrist of an exemplary swimmer at sequential times through a full stroke. They may further comprise corresponding calculated angles and positions of the arm of the exemplary swimmer. The stroke related data may comprise position and orientation data for the wrist of the swimmer being coached at sequential times through a full stroke. They may further comprise corresponding calculated angles and positions of the arm of the swimmer being coached. These stroke-related parameters may be known before-hand e.g., through previously held trial-runs or from an expert swimmer's stroke data. The system, when in operation, performs a plurality of functions: sensing and measuring movement of a wrist of the swimmer during a swimming session, analysing the sensed and measured movement of the wrist to determine stroke related data for the swimmer, comparing the determined stroke related data with the stroke related parameters to derive at least one insight, and retrieving a set of instructions from the memory unit based on the derived at least one insight. Moreover, the system also displays a graphical representation of the stroke related data and the retrieved set of instructions in which the retrieved set of instructions are configured to coach the swimmer i.e., to improve a stroke of the swimmer. In this manner, the swimmer can access a playback of his or her strokes from the system, in particular, a display unit of the system, at any time after a swimming session (e.g., after a lap or a desired block of swim, a single swimming session, or multiple swimming sessions) in order to track their performance, receive the set of instructions and accordingly improve their performance over time.

Further, as movements can be accurately and efficiently sensed and measured in comparison to a physical coach, the system allows the swimmer to take full advantage of the high accuracy and efficiency with which the system operates. Moreover, even if used in conjunction with a physical coach, the system requires minimal to no intervention (e.g., supervision) by the physical coach for coaching the swimmer. Therefore, the system is cost effective as it mitigates the need for a physical coach.

Referring now to FIG. 1, illustrated is a schematic or block representation of a system 100 for coaching a swimmer 200 (shown in FIG. 2), in accordance with an embodiment of the present disclosure. As shown, the system comprises a sensing unit 102, a memory unit 104, a processor 106, and a display unit 108. Explanation to the components of the system of FIG. 1 will hereinafter be made in conjunction with FIGs. 2 through 6.

As shown in FIG. 2, the sensing unit 102 is positioned on a wrist 202 of the swimmer 200. The sensing unit 102 is configured to sense and measure movement of the wrist 202 during a swimming session. As shown in FIG. 2, the swimmer 200 typically moves his or her wrist 202 through a sequence of positions (hereinafter also interchangeably referred to as 'modes') during a stroke (hereinafter also interchangeable referred to as 'stroke cycle'). For example, the wrist 202 may be sequentially moved from e.g., a catch position 206 towards a pull mode 208, a push mode 210, a finish mode 212 before transidoning into a recovery mode 214 (in which the wrist 202 is moved to a position above the water line), an entry mode 216, and a press mode 218 to return to the catch position 206 and repeat the stroke cycle.

FIGs. 3A-3C depicts a diagrammatic illustration of an exemplary implementation of the system 100. As shown in FIG. 3A, the system 100, and in particular at least the sensing unit 102 of the system 100, is implemented by way of a wrist-worn device 300 having a wrist band 302 that facilitates the swimmer 200 to wear the device onto his or her wrist 202 (refer to FIG. 2). Additionally, optionally, the sensing unit 102 may comprise at least one of: a tri-axial accelerometer, a tri-axial gyroscope, and a magnetic angular rate gravity (MARG) sensor. During operation, the sensing unit 102 can sense any axial movement of the wrist 202 in X, Y, and Z-axes. Additionally, the sensing unit 102 can also sense any rotational movement i.e., a pitch 304, a roll 306, and a yaw 308 of the wrist 202 about the Y, X, and Z-axes. As shown in FIG. 33, the wearable device 300 is worn by the swimmer 200 on wrist which moves along with the corresponding wrist (and the corresponding arm) as the swimmer 200 may perform a stroke to carry out the swimming activity. As may be seen from FIG. 3C, With continued reference to FIGs. 1, 2 and 3A-3C, the sensing unit 102, the memory unit 104, the processor 106, and the display unit 108 may altogether reside on the wrist-worn device 300 of FIGs. 3A-3C. That is, the memory unit 104 may be a random-access memory unit residing on the wrist-worn device 300. In other words, the memory unit 104 may be located together with the sensing unit 102, the processor 106, and the display unit 108. Optionally, the memory unit 104 may be implemented by way of a remotely located server e.g. a cloud-based storage system.

The memory unit 104 comprises stroke related parameters (such as, stroke related parameters 410 as will be discussed in reference to FIGs. 4A and 43 later in the description) stored therein. These stroke-related parameters 410 includes e.g., an expert swimmer's stroke data, or data from previous trial runs of another swimmer that are known before-hand and stored within the memory unit 104.

The processor 106 disclosed herein may embody a single processor or comprise of multiple microprocessors that are configured to operatively perform functions consistent with the present disclosure (as explained later herein). Numerous commercially available microprocessors can be configured to perform the functions of the processor 106 disclosed herein. Optionally, the processor 106 could readily be embodied in a general machine microprocessor capable of controlling numerous other functions of a general-purpose machine. That is, the wrist-worn device 300 comprising the processor 106 may also be used as e.g. a watch or a health tracking device. Some examples of commercially known wrist-worn devices used for such purposes may include fitness trackers such as Fitbit®, or watches sold under the commercial trademarks Apple® or Garmin®. Further, various routines, algorithms, and/or programs can be programmed within the processor 106 for execution thereof. For example, the processor 106 may include a power supply circuitry, a signal conditioning circuitry, and other types of circuitry for performing functions that are consistent with the present disclosure.

The processor 106 is configured to analyse the sensed and measured movement of the wrist 202 to determine stroke related data, which may be represented as graphic 414 (refer to FIGs. 4A and 4B) for the swimmer 200. Optionally, the processor 106 is configured to analyse the sensed and measured movement using one or more mathematical algorithms e.g., correlation algorithms or recursive algorithms that are beneficially used for predictive analysis in the field of biomechanics. The determined stroke related data comprises a calculated angle and position of at least one arm 204 of the swimmer 200. Such angle and position may be calculated by simple mechanics from the calculated positions and orientations of the wrist, based on ergonomic information about the human arm and dimensional information on the swimmer. Additionally, or optionally, the determined stroke related data may be combined (i.e. integrally used) with a timing of the movements executed by the wrist 202 of the swimmer 200 during the stroke cycle for analysing the sensed and measured movement and determining stages of motion of the swimmer's arm therefrom. Further, the processor 106 is also configured to compare a graphic of the determined stroke related data 414 with a graphic of the stroke related parameters 410 to illustrate at least one insight (refer to FIGs. 4A and 4B), and retrieve a set of instructions 408 and observation of progress 406 from the memory unit 104 based on the derived at least one insight. The processor 106 compares the determined stroke related data, e.g., a position and angle of the swimmer's arm during a cross-over movement with the stroke related parameters e.g. a target position and angle of an expert swimmer's arm during the crossover movement. The processor 106 compares the determined stroke related data, e.g., a pull-through movement of the swimmer's arm with the stroke related parameters, e.g. a target position and angle of an expert swimmer's arm during the pull-through movement.

Referring to FIGs. 1, 4A and 43, the display unit 108 is arranged to display a graphical representation of the stroke related data 414 and the retrieved set of instructions 408 in which the retrieved set of instructions 408 are configured to coach the swimmer 200. As shown in the view of FIGs. 4A and 43, the display unit 108 comprises of a Graphical User Interface (GUI) 402 for displaying the graphical representation of the stroke related data 414 and graphical representation of stroke related parameters 410, and the retrieved set of instructions 408 and observation of progress 406. Optionally, the display unit 108 is configured to display is a spatial distribution 414 of the determined stroke related data using one of a heat map or tabulation. As shown exemplarily in the view of FIG. 4A, the display unit 108 is configured to display the spatial distribution of the determined stroke related data 414 e.g., the cross-over movement from several stroke cycles using the heat map 404. Further, the display unit 108 is also configured to display the retrieved set of instructions using 408 at least one of text, images and video. As shown exemplarily in the view of FIGs. 4A and 43, the display unit 108 is configured to display the retrieved set of instructions 408 using text. Furthermore, the processor 106 is also configured to optionally compute a severity 412 in a deviation of each stroke related data from a corresponding stroke related parameter. The processor 106 may retrieve historical stroke related data of the swimmer 200 over a period of time from the memory unit 104. Referring to the exemplary illustration of FIG. 5, the display unit 108 is configured to display a severity of improvement needed based on the historical stroke related data of the swimmer 200 retrieved by the processor 106 from the memory unit 104.

Optionally, the derived at least one insight comprises two or more insights therein. In such cases, the processor 106 is configured to determine a rank of each insight from the two or more insights using an insight hierarchy determination algorithm and retrieve the set of instructions 408 to provide the retrieved set of instructions 408 in an order of development to the swimmer 200 based on the ranking of each insight from the two or more insights provided by the insight hierarchy determination algorithm. The position in the hierarchy may be influenced by the magnitude of differences between stroke related data 10 and stroke related parameters for a particular insight.

The set of instructions may be selected by the processor from a database of instructions in an extension of the algorithm, where the processing may be based on a look-up table.

Optionally, the method may be implemented using data acquired over an entire swimming session or over a portion of such swimming session. For example, the portion of a swimming session may be defined by a lap, which in turn is defined by the length of a swimming pool. For example, the portion of a swimming session may a given distance swam (for example, 100m or 800m or 1500m) or a given number of strokes as identified by the zero points or a time interval. Such portion may be defined by any factor which is known (namely input or measured) to the system. Optionally, when the method is implemented using data acquired over a portion of a swimming session, alerts may be sent to the swimmer. For example, an alert may be triggered after a number of swimming strokes. For example, such alerts may be sent if the system or the coach identify one or more factors or insights measured or calculated that deviate from a predicted ideal or from the swimmer's average on a particular session or from the swimmer's historical average.

Optionally, the wearable device is equipped with heart rate monitor. 30 Herein, the system 100 may alert the swimmer if the heart rate value drops below a certain level or raises above a defined level. These would help in monitoring the health of the swimmer.

It will be evident that the method relates to coaching the swimmer 200 as described in conjunction with the system 100 herein. The steps of the method herein may be performed by the processor 106 of the system, 100. Therefore, various embodiments and variants disclosed regarding the system 100 above apply mutatis mutandis to the method.

FIG. 6A is a flowchart 600A listing steps of a method for coaching a swimmer, in accordance with an embodiment of the present disclosure.

The method includes, at a step 602, positioning a sensing unit 102 on a wrist 202 of the swimmer 200. Herein, the sensing unit 102 is a wearable device which in this case is a wrist-watch (such as, the wrist-watch 702 as discussed in reference to FIGs. 7A-7C). Herein, the sensing unit 102 incorporates a gyroscopic sensor and an accelerometer (which are well known in the art, and now shown herein). The method includes, at a step 604 sensing and measuring the movement of the wrist 202 during a swimming session using the sensing unit 102. Herein, the gyroscopic sensor and the accelerometer present in the sensing unit 102 measures the movement of the wrist 202 on which the sensing unit 102 is worn.

The method includes, at a step 606 analysing the sensed and measured movement of the wrist 202. Herein a processor 106, is employed to analyse the sensed and measured movement of the wrist 202 in order to determine the stroke related data 414 for the swimmer 200. The method includes, at a step 608 deriving at least one insight. Herein, the processor 106 is employed in order to compare the stroke related data 414 with the known stroke related parameters 410 to derive at least one insight 406. The method includes, at a step 610 retrieving a set of instructions based on the derived at least one insight 406. Herein, the processor 106 retrieves a set of a set of instructions 408 from a memory unit 104 based on the derived at least one insight 406. The method includes, at a step 612 displaying a graphical representation of the stroke related data 414. The graphical representation of the stroke related data 414 and the retrieved set of instructions are displayed to coach the swimmer 200, using the display unit 108. Herein, the displayed instructions help in improving the swimming of the swimmer 200.

FIG. 6B is a flowchart 600B listing steps of a process for analysing the sensed and measured movement of the wrist 202 to derive at least one insight 406, in accordance with an embodiment of the present disclosure. In an embodiment, the wearable device is an Apple® Watch. In another embodiment, the wearable device is configured to be worn on the ankles. Herein, the wearable device includes wireless transmission and network connection capabilities.

The process includes, at a step 614, logging data related to movement of the wrist during the swimming session from the sensing unit 102. Herein, the data is logged form the sensing unit 102 using the accelerometer and gyroscope channels of a wearable device worn on the wrist 202 by the swimmer 200. Herein, the wearable device may be in the form of a watch, a ring or a band comprising the gyroscopic sensor and the accelerometer sensor. The gyroscopic sensor measures the angular velocity and the zo accelerometer sensor measures the acceleration of the arm of the swimmer 200 on which the wearable device is worn. Once the wearable device is worn and logged in, essential information like swimming pool length and personal information such as, height of the swimmer, weight of the swimmer and arm length may optionally be provided as an input to the wearable device. In case the wearable device is a wrist-watch, elbow to wrist distance or any information that enables accurate determination of movement of the wrist and/or the arm may also be provided. In some cases, such information may be read from one or more of a default setting of the wearable device, from a system device and from another connected device. Another connected device may be a personal mobile phone that keeps a record of data that may be used by the wearable device. The present solution uses the sensors signals to measure/calculate the rate of change, with examples here given in radians per second. A typical wearable device, e.g. Apple® Watch or the like, generally sends such signals every 20 milliseconds.

Referring to FIGs. 7A-7C, there are shown illustrations of a wearable device 700 with different graphical user interfaces (GUIs), in accordance with an exemplary implementation of the present disclosure. The wearable device 700, herein, is a wrist-watch which could be worn on the arm by the swimmer. As shown in FIG. 7A, GUI 702 allows the swimmer to start a swim session, and provide inputs related to the swim session including details like "Pool Swim," "Open Water," etc. As shown in FIG. 7B, GUI 704 allows the swimmer to define pool length, and further provide option to start logging in data by pressing on "Start Workout" button. As shown in FIG. 7C, GUI 706 allows the swimmer to check the swimming data, including the total swim time, distance travelled, number of laps, etc. Referring back to FIG. 63, the process includes, at step 616, identifying starting and ending of each stroke from the logged data (in step 614) using pitch rate criteria. Specifically, this involves reading the pitch information from the logged data, and finding the zero points for rate of rotation, as discussed later in reference to FIG. 8. It is to be noted that there are usually two zero points for the pitch in each full stroke. Herein, the stroke, i.e. the arms stroke, involving the movement of the arm from one point back to the same point is considered as one cycle. Hence, the start and end point remain constant for the arm stroke. The algorithm employs a pitch rate to define a full stroke (as discussed later in more detail). The pitch rate may be defined as the rate at which the arm of the swimmer pitches. In order to set the start and end points for each stroke, the algorithm assigns a negative value to the underwater portion of the stroke which is measured by the gyroscope sensor and a positive value to the over water portion of the stroke. Therefore, a full stroke is set to start/end when the pitch rate crosses the zero value. It is to be noted that the given scenario is for when the wearable device is worn on the left wrist by the swimmer. When the wearable device is worn on the right wrist instead, the algorithm assigns a positive pitch rate during the underwater stroke and a negative pitch rate on the recovery over the water surface.

Referring to FIG. 8, there is shown an illustration of a graph 800 for pitch rate versus time, in accordance with an exemplary implementation of the present disclosure. The pitch rate is the rate at which the pitch of the swimmer's wrist changes (see FIG 3c). The pitch rate is measured from the gyroscopic sensor and it is measured in terms of radians per second.

The graph 800 gives information of the peak pitch rate 802 which is the maximum pitch rate for a stroke. The peak pitch rate 802 is identified from the graph (as shown) and occurs during the arm recovery over the surface of the water (represented by the numeral 214 in FIG. 2). Herein, the end of the underwater stroke is represented by the numeral 806 and is the point where pitch rate moves from negative to positive prior to the peak pitch rate 802. The beginning of the stroke is represented by the numeral 804 and is the point prior to the end of the stroke 806 where pitch rate moves from positive to negative, allowing for over sixty percent of total negative pitch to have occurred between. Herein the arrow 808 represents the sixty percent of total negative pitch.

Once the start and end of the stroke has been determined from pitch rate analysis 800 the stroke time can be measured between these two points. However, prior to this determination an estimate of stroke time is required to locate the peak pitch rate 802 for each stroke. It may be understood that the factors such as, the amount of force applied and technique/position may change, but the stroke time is constrained within known upper and lower bounds. Too long a stroke time and the swimmer starts to drop low in the water, increasing drag and removing their ability to swim effectively. Too short a time and the swimmer cannot complete strokes given the resistance of the water. Hence the stroke time is always within a range that allows the algorithm to accurately identify peak pitch rate 802. An average stroke time across swimmers is approximately 2 seconds which may be used as a default value.

Since swimming strokes perform the same set of actions of going in and out of water and finally arriving at the same position, the variation in stroke time is generally not large enough to cause inaccuracies in the model. It may be understood that the factors such as, the amount of force applied and technique/position may change, but the stroke time is approximately constant. In fact, in order to swim efficiently, the stroke time should be maintained as a constant; otherwise, the body would not be able to move and glide, and would drag and may sink to some extent. Hence, the swimming of the swimmer may be affected if the stroke time is not maintained constant. In the present embodiments, the stroke time is measured by the wearable device or any another device specifically meant for measuring the stroke time.

As discussed, there are usually two zero points for the pitch in each full arm stroke cycle. It may be appreciated that the peak and the zero point proceeding the same could be easily be easily identified from the graph 800. It may be noted that the second zero point 806 can be confused by movements during the stroke, as in the third cycle shown. From the pitch definition, the negative values correspond to over movements of the wrist, i.e. when the swimmer is pulling on the water. The observed coincidence of pitch movements allows a second criterion to be selected, i.e. that greater than 60% of the negative pitch has occurred between the two significant zero pitch rate measurements. That is, the front of the stroke is considered to occur after 60 percent negative pitch has occurred. Herein, the arrow 808 represents the sixty percent of total negative pitch. It is to be appreciated that the second such zero pitch rate point 804, working backwards along the time axis, is the critical start of the stroke cycle (and end of the proceeding stroke cycle). It is also referred to as the front of the stroke; whereas the rear of the stroke could be thought of as the part of the stroke that the swimmer ceases to apply positive traction against the water. The arm may or may not be still in the water at the rear of the stroke. In order to differentiate the points in circle 810 (as shown in FIG. 8) from zero points 804, given that positions and angles are known, the points inside the circle 810 would always fall less than 60 percent negative pitch, so those are identified by not being greater than 60 percent negative pitch.

Again, referring back to FIG. 6B, the process includes, at step 618, determining orientations from the log data and the identified start and end of stroke. Herein, a base set of orientations are calculated from the log data, in particular gyroscope data. For this purpose, the strokes are integrated over time for the determination of the accurate position of the wearable device. For instance, referring to FIG. 10A-10B, provided are positional model output (represented by numeral 1000A) for various strokes (Strokes:1-6) from the wearable device (as shown in FIG. 10A) and the extrapolated positions (represented by numeral 10003) of elbow, hand, etc. (as shown in FIG. 103) for the same strokes from the positional model output of FIG. 10A.

A base set of orientations is determined between the start and end of the stroke cycle identified in step 616, using quaternions to represent the orientation vectors. The gyroscope rates of 'pitch', 'roll' and 'yaw' are integrated over time for the determination of the accurate orientation of the wearable device/arm. Thus, the calculations define vectors to represent each direction through each complete arm stroke cycle.

It may be understood that such approach may lead to a phenomenon known as "Gimbal lock." Herein, the gimbal lock is a mathematical problem that leads to a loss of one degree of freedom in a three-dimensional and three-gimbal mechanism. The problems occur when two axes of the three-gimbal mechanisms are parallel to each other making the system neglect one axis and thus turn into a two-dimensional one. Since, the action of swimming is highly repetitive, accumulation of errors may often lead to the gimbal lock. In order to get rid of the gimbal lock, some corrections mechanisms are employed. The gimbal lock problem is prevented by using the quaternion mathematical model on the gyroscope data. The quaternion mathematical model uses numbers that extends the complex numbers. The model is used in particular for calculations involving three-dimensional rotations such as, for multiplication of two quaternions which are parameters of the model. Herein, the two quaternions would be vector quantities as those define the orientation of gyroscope sensor. The two quaternions may also be any directional or positional parameters. Since the quaternions are non-cumulative, hence the gimbal lock type problem is prevented.

The process further includes, at step 620, calculating direction of average acceleration over stroke cycle using orientations. It may be appreciated that the acceleration is quite complex to calculate because the orientation of the wearable device is continually changing. So, the orientation data has to be applied with the accelerometer data to track the magnitude and direction of the acceleration over time. As may be understood, a time-weighted average of the accelerations results in an acceleration vector, which represents the change in velocity between the initial velocity at the start of the stroke to the end velocity at the end of the stroke. Herein, the calculations are performed using gravity force as the known force (under which only small deviations have been found, but even those deviations are inconsistent). The gravitational force is chosen, since a swim stroke requires an arm to move through a complete stroke cycle, ending at the exact same place and velocity relative to the swimmer's body, the only resulting force would be gravity. Other forces may be assumed to be not present because their presence would deviate the start and end of the stroke. As may be understood, the components of 5 acceleration applied to the watch by the swimmer in vertical directions would not necessarily average out to zero; but if they did, the net acceleration applied to the watch would be the acceleration due to gravity. The other force vectors related to the action of the water and variations by the swimmer, average out very closely to gravity. Thus, the 10 determination of the direction of gravity provides an orientation fix.

Further, in the process of step 620, the orientations are rotated so those are facing straight down. Also, the correction of the zero points (i.e., points 806 in FIG. 8) need to be done via adjusting velocity vectors rather than position. This information is then used as the key directional is information for calculating accurate orientation and position. It may be appreciated that such correction techniques are used in applications such as, video games and aircrafts; however, the determination of the accurate position of the wearable device of the present disclosure may not be replicated in the above-mentioned applications. For instance, referring to FIGs. 10A-10B, in exemplary "Stroke:1," the velocity North (vN) changes from 0.13 to 0.20 meters per second between the beginning and end of the stroke cycle, and thus other positive changes to velocity can be determined for the East and Down directions using the techniques described above.

The process further includes, at step 622, determining positions and orientations in space based on the identified start and end of stroke and the calculated direction of average acceleration. This could be calculated from the data obtained by the accelerometer sensor. It may be appreciated that, if the corrections in steps 618 and 620 are not implemented, the accurate position will not be obtained. This is equivalent to collection of different, inaccurate sensor data. In step 622, the accelerometer data is first converted to velocity data and then converted to position data, such that the stroke always starts and ends at the same place. It may be appreciated that with such data, the positions and orientations in space could be calculated at any given instant of the swimming session.

The difference in position between the start and the end of a stroke (presuming that the forward motion of the swimmer is subtracted) should be in the same direction as the presumed initial velocity. Thus, changing 10 the initial velocity brings the start and end of the stroke into alignment.

The process further includes, at step 624, adjusting initial velocity of stroke so that the start and end of stroke is in same place. The step 624 provides the correction of the distances, thus complementing the correction of the orientations as performed in step 620. It may be appreciated that these corrections are necessary because the velocity at the start of the stroke (initial velocity) is unknown, and therefore it is determined to allow the corrections that result in the stroke track resolving back where it started. Potentially, depending on the accuracy of the measurement device, the corrections may also overcome a potential accumulation of measurement errors inherent in the use of the sensors of the wearable device The process may further include extrapolating the elbow and/or the hand positions (as shown in FIGs. 10A and 10B). The extrapolation helps in calculation of the elbow and hand positions and accordingly, the improvements on position of the stroke could be recommended. Such extrapolation can be done by any known suitable technique in the art without any limitations.

The process may further include displaying the results in a system device (such as, a phone) and/or a wearable device (such as, a smartwatch). In some examples, the output of the process as described above may be checked by comparing with video recordings. This may be done to test if the results are correct for a particular distance and body measurements. It may be appreciated that this step may be automated using image processing techniques as known in the art. As discussed, all the necessary calculation for the process as described above may take place at the wearable device itself or the system device, or in a cloud server without departing from the spirit and the scope of the present disclosure.

Referring to FIG. 9A, there is shown a system device 900 depicting a GUI as provided by the app for displaying performance history. Such GUI may provide a coach with performance analysis for various swimmers getting coached under him. The system device 900 gives an overview of the activity performed by the swimmer. As seen from FIG. 9A, for an exemplary day "6th June 2020," the swimmer travelled a distance of about 300 meter in 5m 47s (five minutes and forty-seven seconds). The average time consumed for covering each 100 meter is depicted as approximately lm 55s (one minute and fifty-five seconds). Similarly, the performance for another day "19th February 2020" is also shown.

Referring to FIG. 96, there is shown the system device 900 depicting a GUI as provided by the app for displaying derived insights. Herein, the app provides information of the individual swim performed by the swimmer. Further, the app provides information about the length of the pool, the distance traversed by the swimmer, the time duration spent, the calories burnt, the average heart rate, the average strokes per minute and the average pace per hundred meters (as depicted). Further, a "GET STROKE INSIGHTS" button 902 may be provided by the app that may be clicked for getting more detailed information about the swim. Furthermore, the app provides a "Session Executive" section that displays the basic insights of the swim. Herein, a score (in this case, 39 out of 100) is provided to provide an overall evaluation of how efficient or how poor that particular swim session was.

Referring to FIG. 9C, there is shown the system device 900 depicting a GUI as provided by the app for displaying stroke insights to the swimmer. 5 The movement of each arm for a stroke is rated out of 5 that depicts how poor or how good the particular arm movement was in the stroke. Further, comments are also provided. In the present example, the left arm is rated at 0.2 and the right arm is rated at 1.2. A "Guru's Tip" section 904 is provided which provides the improvement techniques that be 10 applied in the successive swimming session. For example, herein, the improvement needed is deep pull through with both the arms. Further, the app allows to display more detailed information by clicking the GUI element 906.

Referring to FIG. 9D, there is shown the system device 900 depicting a GUI as provided by the app for displaying the detailed view on how to improve the swimmer's technique. This is obtained by clicking the GUI 906 of FIG. 9C. Herein, a "Videos" section 908 is included which provides videos that may be viewed by the swimmer or by the coach in order to get a better understanding of the mistakes that the swimmer was making while swimming.

Referring to FIG. 9E, there is shown the system device 900 depicting a GUI as provided by the app for displaying an overview of progress of the swimmer. Herein, a "Current Training Balance" section 910 gives information regarding fitness, training stress and recovery. Further, a "Summary" section 912 is provided which gives a summary of total distance swum and the number of swimming sessions for a particular time. Herein, the summary for 4 weeks, 12 weeks and all time may be obtained by the clicking the respective icon. Further, a "Fitness Prediction" section 914 provides an overview of how fit the swimmer is.

The present disclosure provides a system and a method for improving swimming techniques, which result in health benefits, prevents injuries and saves time for swimming coaches. The wearable device worn by a swimmer, preferably a wrist-worn device (such as, an Apple® Watch) in connection with an app connected to the wearable device and optionally to a system device is utilized for calculation movement deviation and recommendation of movement correction for swim related activities. The system and the method of the present disclosure is advantageous, as there is no need for direct measurements of force or position. Rather positioning is accurately determined using the data obtained from the gyroscope sensor. This eliminates the need of measuring positions by video or other special positions. Hence, the problem of privacy and privacy regulations which in most cases forbid filming and film analysis, and video sharing of swimmers between swimmers and coaches is solved.

The invention also solves the particular problem of complexity, regulates time and resources needed to comply with regulation of video acquisition particularly for under-aged swimmers, which in most countries mean that videoing is not allowed for swimmers aged below 16 to 18 years old.

It may be understood that, the present disclosure relates to the movement detection, transmission, calculation and correction of swimming needs. Herein, the movements are the ones that are typical and necessary for a swimming stroke to make a body float and move in water. It may be understood other movements may not be related to the present disclosure. For instance, the present disclosure cannot be applied in other application models because only for swimming the start and end of the stroke is at same position and is replicated (generally, every 2 seconds). In other sports, such as, tennis golf, each stroke starts and end on different positions. Similarly, in running, or any other movement sports, the steps or repeated movements follow other parameters, notably different actions or even terrain variations.

Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe and claim the present disclosure are intended to be construed in a nonexclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural.

Claims

CLAIMS1. A system (100) for coaching a swimmer (200), the system comprising: - a sensing unit (102) positioned on a wrist (202) of the swimmer, 5 the sensing unit configured to sense and measure movement of the wrist during a swimming session; - a memory unit (104) comprising stroke related parameters (410) stored therein; - a processor (106) in communication with the sensing unit and the 10 memory unit, the processor configured to: - analyse the sensed and measured movement of the wrist to determine stroke related data (414) for the swimmer; - compare the determined stroke related data with the stroke related parameters to derive at least one insight (406); -retrieve a set of instructions (408) from the memory unit based on the derived at least one insight; and -a display unit (108) in communication with the processor, the display unit arranged to display a graphical representation of the stroke related data and the retrieved set of instructions, wherein the retrieved set of 20 instructions are configured to coach the swimmer.
2. The system (100) according to claim 1, wherein, for analysing the sensed and measured movement of the wrist to determine stroke related data (414) for the swimmer (200), the processor (106) is further 25 configured to: - log data related to movement of the wrist during the swimming session from the sensing unit (102); - identify start and end of stroke from the logged data using pitch rate criteria; -determine orientations from the logged data and the identified start and end of stroke; - calculate direction of average acceleration over stroke cycle using orientations; - determine positions in space from the logged data, based on the identified start and end of stroke and re-determine orientations based on 5 the calculated direction of average acceleration; and - adjust initial velocity of stroke such that the start and end of stroke is in same position.
3. The system (100) according to any of claims 1 or 2, wherein the sensing unit (102) comprises at least one of: a tri-axial accelerometer, a tri-axial gyroscope, and a magnetic angular rate gravity (MARG) sensor, to sense and measure movement of the wrist during a swimming session.
4. The system (100) according to claim 3, wherein the sensing unit is comprises a tri-axial accelerometer and a tri-axial gyroscope.
5. The system (100) according to any of claims 3 or 4, wherein the processor is configured to determine orientations from the logged data obtained from the tri-axial gyroscope.
6. The system (100) according to any of claims 3 to 5, wherein the processor is configured to determine positions in space from the logged data obtained from the tri-axial accelerometer.
7. The system (100) according to any of claims 3 to 6, wherein the processor is configured to determine positions and orientations in space throughout the stroke.
8. The system (100) according to any of the preceding claims, wherein 30 the display unit (108) is configured to display: - a spatial distribution (414) of the determined stroke related data using one of: a heat map (404) or a tabulation; and - the retrieved set of instructions (408) using at least one of: text, images and video.
9. The system (100) according to any of the preceding claims, wherein the determined stroke related data comprises a calculated angle and position of at least one arm (204) of the swimmer (200).
10. The system (100) according to any of the preceding claims, wherein the derived at least one insight comprises two or more insights therein, and wherein the processor (106) is configured to determine a rank of each insight from the two or more insights using an insight hierarchy determination algorithm and retrieve the set of instructions (408) to provide the retrieved set of instructions in an order of development to the swimmer (200) based on the ranking of each insight from the two or more insights provided by the insight hierarchy determination algorithm.
11. The system (100) of any of the preceding claims, wherein the zo memory unit (104) is a remotely located server.
12. The system (100) according to any of the preceding claims, wherein the memory unit (104) is a random-access memory unit located together with at least one of: the sensing unit (102), the processor (106), or the 25 display unit (108).
13. A method for coaching a swimmer (200), the method comprising: - positioning a sensing unit (102) on a wrist (202) of the swimmer; - sensing and measuring, using the sensing unit, movement of the wrist 30 during a swimming session; - analysing, using a processor (106), the sensed and measured movement of the wrist to determine stroke related data (414) for the swimmer; - comparing, using the processor, the determined stroke related data with 5 stroke related parameters to derive at least one insight (406); - retrieving, by the processor, a set of instructions (408) from a memory unit (104) based on the derived at least one insight; and - displaying, using a display unit (108), a graphical representation of the stroke related data and the retrieved set of instructions to coach the 10 swimmer.
14. The method according to claim 13, wherein, for analysing the sensed and measured movement of the wrist to determine stroke related data (414) for the swimmer (200), the method further comprises: -logging data related to movement of the wrist during the swimming session from the sensing unit (102); - identifying start and end of stroke from the logged data using pitch rate criteria; - determining orientations from the logged data and the identified zo start and end of stroke; - calculating direction of average acceleration over stroke cycle using orientations; - determining positions in space from the logged data, based on the identified start and end of stroke and re-determining the orientations 25 based on the calculated direction of average acceleration; and - adjusting initial velocity of stroke such that the start and end of stroke is in same position.
15. The method according to any of claims 13 or 14, wherein the 30 sensing unit (102) comprises at least one of: a tri-axial accelerometer, a tri-axial gyroscope, and a magnetic angular rate gravity (MARG) sensor, to sense and measure movement of the wrist during a swimming session.
16. The method according to claim 15, wherein the sensing unit 5 comprises a tri-axial accelerometer and a tri-axial gyroscope.
17. The method according to any of claims 15 or 16, wherein the processor determines orientations from the logged data obtained from the tri-axial gyroscope.
18. The method according to any of claims 15 to 17, wherein the processor determines positions in space from the logged data obtained from the tri-axial accelerometer.
19. The method according to any of claims 15 to 18, wherein the processor determines positions and orientations in space throughout the stroke
20. The method according to any of claims 13 to 19 further comprising 20 displaying: - a spatial distribution of the determined stroke related data using one of: a heat map (404) or a tabulation; and - the retrieved set of instructions (408) using at least one of: text, images and video.
21. The method according to any of claims 13 to 20, wherein the determined stroke related data comprises of a calculated angle and position of at least one arm (204) of the swimmer (200).
22. The method according to any of claims 13 to 21, wherein the derived at least one insight (406) comprises two or more insights therein, and wherein the processor (106) is configured to determine a rank of each insight from the two or more insights using an insight hierarchy determination algorithm and retrieve the set of instructions (408) to provide the retrieved set of instructions in an order of development to the swimmer (200) based on the ranking of each insight from the two or more insights provided by the insight hierarchy determination algorithm.
23. The method according to any of claims 13 to 22, wherein the method is implemented in a wrist-worn device (300).