Smart strength training device

Abstract

An exercise device, strike pad, and method are disclosed. The exercise device includes a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion. The exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements and a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements. A strike pad adapted for attachment to an exercise device can include one or more additional sensors configured to capture sensor measurements of actions performed with the strike pad.

Description

Attorney Docket No.: 36610-0120W01SMART STRENGTH TRAINING DEVICECROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Provisional Application No.63 / 748,167, filed on January 22, 2025, and to U.S. Provisional Application No. 63 / 913,241, filed on November 7, 2025, the contents of which are all hereby incorporated by reference.BACKGROUND

[0002] This disclosure relates generally to an item of exercise equipment.

[0003] An athlete can perform a number of different exercises using an item of exercise equipment. An athlete can perform an exercise using exercise equipment by aligning different parts of the body relative to each other and / or the exercise equipment. Some sports demand repetitions of exercises to improve performance of complex actions for the sport.SUMMARY

[0004] This specification describes an apparatus and method for a sensor-based strength training device, also referred to as a smart exercise device configured to apply sensor-based performance tracking for athletes. The disclosed technology captures sensor measurements of athletes performing exercises using an item of exercise equipment, such as a sled, to evaluate the athlete’s using metrics determined from the sensor measurements. The sensor-based strength training device can include a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along the length of the frame structural member and each having a respective top portion and a respective bottom portion. The sensor-based strength training device can include a plurality of sensors and a computing device, each sensor configured to capture sensor measurements and the computing device configured to process sensor measurements to generate an output.

[0005] The smart strength training device obtains sensor data of a person performing an exercise action with a strength training device and determines performance metrics for the exercise action. The smart strength training device can include an integrated computing device with associatedAttorney Docket No.: 36610-0120W01memory, data storage, and processors that allow for on-demand processing of sensor measurements. The computing device can include a sensor engine that utilizes a model to generate output data representing performance metrics of the athlete performing the exercise. The model can be configured to apply machine learning techniques, statistical techniques, a physical-based simulation, or some combination thereof, to generate an output representing performance metrics based on the sensor measurements. The metrics can include an evaluation of the exercise performed by the person and generates an output for the athlete’s performance of the exercise, e.g., a numerical output, an output label, indicating a degree of performance for the exercise. For example, some exercises can demand explosive or dynamic movements, e.g., from a relatively stable pose, and then gradually adjust the amount of power to drive the equipment to successfully execute a complex exercise that meets an acceptable standard for competitive sports.

[0006] The smart strength training device can be, for example, an American football training sled. An American football training sled (also referred to as a “football sled” or a “sled”) is an item of exercise equipment that an athlete can use for American football training by performing full-body exercises using the sled. Examples of exercises with the sled can include pushing the sled, dragging the sled, lifting the sled, flipping the sled, etc. However, the sled need not be limited to American football training, as the exercises can also be performed for strength training in other applications, such as recreational fitness, general fitness, first responder physical training, etc. These applications can demand training with a sled-type strength training device (among other types of strength training devices) to improve a user’s strength (e.g., an ability to carry a load with a particular weight) and endurance (e g., an ability to carry a load for an extended period of duration). For example, a backpacker can perform exercises using a strength training device for endurance and strength training, so that they can carry significant weight in backpacks over long periods of time. As another example, first responders such as search and rescue personnel, firefighters, police officers, etc., can perform exercises using a strength training device for first responder training, due to training demands to carry heavy loads such as equipment, people, debris, etc.

[0007] The smart strength training device can include multiple sensors located at different locations on the strength training device to capture sensor measurements, providing a holistic evaluation of exercise performance from multiple angles and parts of the device. The device can provide improved accuracy and precision by leveraging multiple instances and types of sensors and allow for sensor fusion to generate performance metrics that evaluate the athlete’sAttorney Docket No.: 36610-0120W01performance. The device also includes a computing device to process the measurements and generate metric representations of an exercise action, e.g., pulling a strength training device, flipping a strength training device, pushing a strength training device, or some combination thereof. Exercise actions can include combinations of movements such as pushing then flipping the strength training device. The device can evaluate the exercise over multiple performed exercise actions, e.g., a sequence of exercises, repetitions of the same exercise, to determine performance consistency or deviation of the athlete with respect to a target performance metric.

[0008] The disclosed training device can include an example frame structural member with an adjustable length and width. The frame structural member can be a telescoping structure where an inner portion of the frame structural member is contained in an outer portion, and the inner portion is slidable, e.g., along an axis. The disclosed training device can include a weight adjustment mechanism that can adjust the amount of weight for the training device, such as by a slidable mechanism corresponding to different weights based on the position of the slider in the slidable mechanism. The weight adjustment mechanism can be a weight-stack pin adjustment system, where plates of weight with selector pins that can be inserted in holes disposed into a weight plate and the pin can be inserted in an opening of at a target weight plate to configure the weight of the training device. The disclosed training device can include bumpers to improve safety by absorbing shock (and other types of forces) when the training device lands on the ground. The disclosed technology also includes friction adjustment mechanisms, such as to mimic friction across different types of surfaces with different coefficients of friction. For example, an exercise can be performed with the exercise device on a first surface with a first coefficient of friction. The exercise device can determine a weight adjustment to mimic performing the same exercise with the exercise device on a surface with a different coefficient of friction than the first surface. The disclosed exercise device can allow for efficient and effective athletic training, such as to adapt the exercise device with an adjusted weight for any surface.

[0009] The disclosed training device can include damping systems to reduce the forces between the training device and a surface for performing the exercise action, e.g., to allow the training device to land more softly. Examples of damping mechanisms include spring-systems, hydraulic systems, pneumatic systems, gel-based damping, magnetic damping, and friction damping, among others. The disclosed training device can also include damping materials such as rubber, elastomers, foams, hydraulic fluids, springs, viscoelastic materials, and components. The disclosedAttorney Docket No.: 36610-0120W01training device can include attachments that allow an athlete to perform complex exercise actions, such as by simulating expected blocking impact that occurs during American football training and games. Examples of the attachments can include medicine balls, blocking dummies, and other types of materials of varying shapes and sizes. The disclosed training device can be a propulsionbased exercise training device (also referred to as an exercise device or training device) configured to provide propulsion that allow the training device to mimic complex movements of players, e.g., offensive moments.

[0010] In general, one innovative aspect of the subject matter described in this specification can be embodied in an exercise device. The exercise device includes a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion. The exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements. The exercise device includes a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0011] The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination. In particular, one embodiment includes all the following features in combination.

[0012] In some implementations, the exercise device includes a first strike pad attachment configured to be attached to the first end of the frame structural member. The first strike pad attachment includes at least a first additional sensor configured to capture sensor measurements.

[0013] In some implementations, the exercise device includes a second strike pad attachment configured to be attached to the second end of the frame structural member. The second strike pad attachment includes at least a second additional sensor configured to capture sensor measurements.

[0014] In some implementations, each of the respective top portion and the respective bottom portion for one or both of the first end or the second end includes a respective bumper configured to provide support for the exercise device.

[0015] In some implementations, the computing device includes one or more processors coupled to one or more computer-readable storage media having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations.Attorney Docket No.: 36610-0120W01The operations include obtaining, from a first sensor from the plurality of sensors, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device. The operations include determining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0016] In some implementations, the operations include generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

[0017] In some implementations, the operations further include obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device. The operations can include determining, from the second sensor data, a second metric representing a characteristic for the exercise action.

[0018] In some implementations, the operations include obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device. The operations include determining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.

[0019] In some implementations, the operations further include generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

[0020] In some implementations, the exercise device is a football training sled.

[0021] In some implementations, sensor measurements captured by a sensor from the plurality of sensors is captured prior to an exercise being performed using the exercise device.

[0022] In some implementations, sensor measurements captured by a sensor from the plurality of sensors is captured while an exercise is being performed using the exercise device.

[0023] In some implementations, sensor measurements captured by a sensor from the plurality of sensors is captured after an exercise is performed using the exercise device.

[0024] In some implementations, the plurality of sensors at least one of (i) a radar sensor, (ii) a LIDAR sensor, (iii) a global position system sensor, (iv) an ultrasonic sensor, (v) an infrared sensor, (vi) a camera sensor, (vii) an image sensor, (viii) an optical sensor, (ix) a light sensor, (x) a proximity sensor, (xi) a narrow-beam ultrasonic sensor, (xii) a power sensor, (xiii) a force sensor and (xiv) a load cell.Attorney Docket No.: 36610-0120W01

[0025] In some implementations, the computing device includes a model configured to apply one or more of (i) machine learning techniques, (ii) statistical techniques, or (iii) a physical-based simulation, to generate the output based on the sensor measurements.

[0026] In some implementations, the computing device is configured to: generate, based on the output, an evaluation of an exercise performed; and provide data indicative of the evaluation.

[0027] In some implementations, to provide the data indicative of evaluation includes providing the data for display on a user interface of a device communicatively coupled to the computing device. I data causes the device to configure the user interface to display one or more user interface elements to display the evaluation.

[0028] In some implementations, at least one sensor from the plurality of sensors is mounted onto the exercise device.

[0029] In some implementations, at least one sensor from the plurality of sensors is co-located with the computing device.

[0030] In some implementations, at least one sensor from the plurality of sensors is configured to capture sensor data in a non-visible electromagnetic spectrum frequency band.

[0031] In some implementations, at least one sensor from the plurality of sensors is configured to capture sensor data in one of radio frequency bands, microwave frequency bands, acoustic frequency bands, infrared frequency bands, and ultrasonic frequency band.

[0032] In some implementations, sensor measurements from at least one sensor from the plurality of sensors includes point cloud data.

[0033] In some implementations, the computing device is configured to: determine, based on the sensor measurements, a first value representing friction between a surface and the exercise device; and based on a comparison of the first value representing friction and a second value representing a target metric, generating an indicator to adjust the exercise device.

[0034] In some implementations, the computing device is configured to determine a type of the surface based on the first value representing friction.

[0035] In some implementations, the respective bottom portion of one or both of the first end and the second end includes at least one of (i) brakes, (ii) wheels, (iii) adjustable braking mechanisms, (iv) a ski, (v) a pair of skis, (vi) a disk, (vii) a bowl, or (viii) a saucer.Attorney Docket No.: 36610-0120W01

[0036] In some implementations, the at least one of (i) the brakes, (ii) the wheels, (iii) the adjustable braking mechanisms, (iv) the ski, (v) the pair of skis, (vi) the disk, (vii) the bowl, or (viii) the saucer is rotatable about an axis relative to the exercise device.

[0037] In some implementations, the respective top portion of at least one of the first end or the second end further includes one or more handles.

[0038] In some implementations, at least one handle from the one or more handles is a horizontally oriented handle connecting a first portion of the respective top portion to a second portion of the respective top portion different from the first portion.

[0039] In some implementations, at least one handle from the one or more handles is a horizontally oriented handle connecting a first portion of the respective bottom portion to a second portion of the respective bottom portion different from the first portion.

[0040] In some implementations, at least one handle from the one or more handles is a vertically oriented handle protruding upward from one or both of the respective top portion and the respective bottom portion.

[0041] In some implementations, a handle from the one or more handles is rotatable about an axis.

[0042] In some implementations, a handle from the one or more handles includes a sensor configured to capture measurements.

[0043] In some implementations, the respective bottom portion of the first end includes one or more wheels and the respective bottom portion of the second end includes an adjustable braking mechanism.

[0044] In some implementations, the respective bottom portion of the first end includes at least one ski or at least one wheel that connects the respective bottom portion of the first end to the respective bottom portion of the second end.

[0045] In some implementations, the plurality of sensors is at least one of a camera sensor and an image sensor. The at least one of the camera sensor and the image sensor is configured for facial recognition by the computing device.

[0046] In some implementations, the exercise device includes one or more damping mechanisms, materials, and absorbers.

[0047] In some implementations, the exercise device includes one or more flywheel devices.Attorney Docket No.: 36610-0120W01

[0048] In some implementations, the exercise device includes a third strike pad attachment configured to be attached to the first end of the frame structural member and adjacent to the first strike pad attachment.

[0049] In some implementations, the third strike pad attachment includes at least a third additional sensor configured to capture sensor measurements.

[0050] In general, one innovative aspect of the subject matter described in this specification can be embodied in a strike pad adapted for attachment to an exercise device. The exercise device includes a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion. The exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements. The exercise device includes a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0051] The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination. In particular, one embodiment includes all the following features in combination.

[0052] In some implementations, the strike pad can include one or more additional sensors configured to capture sensor measurements of actions performed with the strike pad.

[0053] In some implementations, the strike pad can include an adapter configured to interface with the exercise device.

[0054] In some implementations, the adapter includes at least one of (i) an electrical connector, and (ii) a mechanical connector, to attach the strike pad to the exercise device

[0055] In some implementations, the electrical connector includes at least one of (i) a wire, (ii) a plug connector, (iii) a coaxial connector, (iv) a magnetic connector, or (v) a spring loaded connector.

[0056] In some implementations, the mechanical connector includes at least one of (i) a snap-fit connector, (ii) a fastener, (iii) a quick-release mechanism, (iv) a clamp, or (v) a hinge.

[0057] In some implementations, the strike pad attachment includes one or both of (i) a processor, or (ii) data storage.Attorney Docket No.: 36610-0120W01

[0058] In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of determining, by a strike pad, that the strike pad has been removably connected to a strength training exercise device; transmitting features of the strike pad to the strength training exercise device; and receiving data from the strength training exercise device based on the features of the strike pad.

[0059] In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of obtaining, by a first sensor of a strike pad, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the strike pad The strike pad is adapted to be removably connected to a strength training exercise device; and determining, from the first sensor data, a first metric representing a characteristic for the exercise action

[0060] In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of obtaining, by at least one sensor of a strike pad adapted for attachment to an exercise device, first sensor data comprising first sensor measurements captured by the at least one sensor for an exercise action with the strike pad; and determining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0061] In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of obtaining, by a first sensor from a plurality of sensors of an exercise device, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device; and determining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0062] The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination. In particular, one embodiment includes all the following features in combination.

[0063] In some implementations, the method can include generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

[0064] In some implementations, the method can include obtaining, by a second sensor from the plurality of sensors, second sensor data comprising second sensor measurements captured by the second sensor for the exercise action with the exercise device. The second sensor is a differentAttorney Docket No.: 36610-0120W01from the first sensor. The method can include determining, from the second sensor data, a second metric representing a characteristic for the exercise action.

[0065] In some implementations, the method can include obtaining, by a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor for the exercise action with the exercise device. In some implementations, the method can include determining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.

[0066] In some implementations, the method can include generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

[0067] In some implementations, the first sensor is a first type of sensor, and the second sensor is a second type of sensor, the second type being different from the first type.

[0068] In some implementations, the method can include determining, by a computing device, a first value representing friction between a surface and the exercise device. The determination is based on the first sensor measurements and / or the second sensor measurements; and based on a comparison of the first value representing friction and a second value representing a target metric, generating an indicator to adjust the exercise device.

[0069] In some implementations, the method can include generating calibration data for the exercise device based on the first sensor measurements.

[0070] In some implementations, the method can include generating feedback data based on the first metric.

[0071] In general, one innovative aspect of the subject matter described in this specification can be embodied in an exercise device. The exercise device includes an adjustable frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the adjustable frame structural member. Each of the first end and the second end of the adjustable frame structural member further include a respective top portion and a respective bottom portion. The adjustable frame structural member includes an inner portion and an outer portion. The inner portion is slidably connected to the outer portion by a mechanism. The mechanism allows the inner portion to be extended or retracted.Attorney Docket No.: 36610-0120W01

[0072] In some implementations, the adjustable frame structural member further includes one or more additional adjustable frame structural members, each of one or more additional adjustable frame structural members being substantially parallel to the adjustable frame structural member.

[0073] In some implementations, each of the one or more additional adjustable frame structural members is connected to the adjustable frame structural member by one or more horizontally oriented structural members.

[0074] In some implementations, one of both of (i) an additional adjustable frame structural members from the one or more additional adjustable frame structural members, and (ii) the adjustable frame structural member, includes one or more compartments.

[0075] In some implementations, the exercise device includes an adjustable weight mechanism. In some implementations, the adjustable weight mechanism is configured to adjust an amount of weight of the exercise device using a slidable pin stack system.

[0076] In some implementations, the slidable pin stack system includes: one or more openings, each opening corresponding to a weight from a plurality of weights; and at least one pin configured to be inserted in an opening from the one or more openings to select the weight corresponding to the opening.

[0077] In some implementations, the exercise device includes one or more damping mechanisms, materials, and absorbers.

[0078] In some implementations, the exercise device includes one or more flywheel devices.

[0079] In general, one innovative aspect of the subject matter described in this specification can be embodied in a rounded exercise pad adapted for attachment to an exercise device. The exercise device includes a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion. The exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements. The exercise device includes a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0080] In some implementations, the rounded exercise pad includes one or more additional sensors configured to capture sensor measurements of actions performed with the rounded exercise pad.Attorney Docket No.: 36610-0120W01

[0081] In some implementations, the rounded exercise pad includes an adapter configured to interface with the exercise device. In some implementations, the adapter includes at least one of (i) an electrical connector, and (ii) a mechanical connector, to attach the rounded exercise pad to the exercise device. In some implementations, the electrical connector includes at least one of (i) a wire, (ii) a plug connector, (iii) a coaxial connector, (iv) a magnetic connector, or (v) a spring loaded connector. In some implementations, the mechanical connector includes at least one of (i) a snap-fit connector, (ii) a fastener, (iii) a quick-release mechanism, (iv) a clamp, or (v) a hinge.

[0082] In some implementations, the rounded exercise pad includes one or both of (i) a processor, or (ii) data storage.

[0083] In some implementations, the rounded exercise pad is connected to the exercise device by a rotating mechanism.

[0084] In some implementations, the rounded exercise pad is one of (i) a medicine ball, (ii) a wall ball, or (iii) a weighted ball.

[0085] In general, one innovative aspect of the subject matter described in this specification can be embodied in a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion; and a propulsion system configured to displace the exercise device along a surface.

[0086] In some implementations, the exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; and a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0087] In some implementations, the exercise device includes a drivetrain system coupled to the propulsion system; one or more driving elements coupled to the drivetrain system; and one or more braking elements coupled to at least one of the one or more driving elements.

[0088] In some implementations, the exercise device includes a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; and a computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.Attorney Docket No.: 36610-0120W01

[0089] In some implementations, the respective bottom portion of one or both of the first end and the second end includes at least one of (i) brakes, (ii) wheels, (iii) adjustable braking mechanisms, (iv) a ski, or (v) a pair of skis, (vi) disk, (vii) bowl, or (viii) saucer.

[0090] In some implementations, the propulsion system includes: a power source configured to provide power; and one or more propulsion elements coupled to the power source and configured to generate force based on power received from the power source.

[0091] In some implementations, the exercise device is configured to receive sensor measurements from a sensor remote from the exercise device. The sensor is a force plate located below the exercise device.

[0092] In some implementations, the force plate is part of a surface below the exercise device.

[0093] In general, one innovative aspect of the subject matter described in this specification can be embodied in a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion; and at least one pulling mechanism coupled to at least one weight. The pulling mechanisms is configured to adjust a position of the at least one weight along the longitudinal axis of the frame structural member.

[0094] In some implementations, the at least one pulling mechanism is configured to adjust the position of the at least one weight in response to a force that is (i) applied to and / or (ii) being applied to, the at least one pulling mechanism.

[0095] In some implementations, an adjustment of the position of the at least one weight along the longitudinal axis is based on, at least in part, the force experienced by the at least one pulling mechanism.

[0096] In some implementations, the force experienced by the at least one pulling mechanism is a force that is (i) applied to and / or (ii) being applied to, the exercise device.

[0097] In general, one innovative aspect of the subject matter described in this specification can be embodied in a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member. Each of the first end and the second end of the frame structural member further include a respective top portion and a respective bottom portion. The respective bottom portion for one or both of the first end and the second end includes a rounded shape.Attorney Docket No.: 36610-0120W01

[0098] The technology described in this specification can be implemented so as to realize one or more of the following advantages.

[0099] The disclosed technology allows for performance tracking of exercise actions such as pushing, pulling, and flipping the exercise device. The disclosed smart exercise devices can include sensors to provide an evaluation of an exercise from multiple angles, thereby providing improved accuracy and precision tracking. The disclosed technology can evaluate sequences of exercise, repetitions of multiple exercises, etc. to determine performance consistency or deviation of the athlete with respect to a target performance metric. In this way, the disclosed technology allows for more accurate, dynamic, and proactive sensor-based strength training, such as by capturing sensor measurements of the athlete’ s stance prior to interacting with the exercise device, the athlete’s motions while they apply force to lift the exercise device, the athlete’s posture while performing motions, and other related motions while the athlete pushes the exercise device. The disclosed technology provides feedback and performance training for the athlete during different phases of an exercise action. The disclosed technology provides a real-time or near real-time analysis of the athlete’s performance to provide feedback for improving exercising training, such as to reduce the risk of injury, improve exercise performance, etc.

[0100] The disclosed technology advantageously allows for more accurate, dynamic, and proactive sensor-based strength training device training, such as football sled training, other types of sports training, general fitness, first responder physical training (e.g., for police officers, firefighters) than otherwise possible with other strength training device training systems that lack these features. A strength training device that does not have this type of sensor-based performance tracking and analysis would be unable to characterize an athlete’s form as they perform an exercise action with the strength training device. The disclosed technology provides improvements by capturing sensor measurements of the athlete’s stance prior to interacting with the sled, the athlete’s motions while they apply force to lift the sled up, the athlete’s posture while performing motions, and other related motions while the athlete pushes the sled. The disclosed sensor-based strength training device can process the sensor measurements to provide useful feedback and performance training for the athlete during different phases of an exercise action.

[0101] Furthermore, the sensor-based strength training device can provide a real-time or near real-time analysis of the athlete’s performance and identify areas for improvement to achieve a target metric or target performance for a given athlete and exercise. The determined metrics canAttorney Docket No.: 36610-0120W01be transmitted to a user device for the athlete. For example, the sensor-based strength training device can generate feedback to indicate which part of an exercise an athlete can put additional effort to achieve the target metric. As another example, the sensor-based strength training device can identify deviations, e.g., insufficient force, excess force, in an athlete’s performance to identify improvements for an athlete when performing an exercise, e.g., pushing and / or pulling motions, flipping, cleans, and other types of combined moves. In some cases, the smart device can detect a deviation in performance and generate an evaluation indicating a likelihood of fatigue, injury, lack of nutrition and / or hydration, unbalanced training, among other sources of degraded performance.

[0102] The disclosed sensor-based strength training device can provide quantitative evaluation of an athlete’s performance for goal setting, e.g., performing an exercise and / or practicing a sport, and provide feedback to assist the athlete in achieving a target goal. For example, the target metric can be a number of flips of the sensor-based strength training device, an amount of force in a movement with the sensor-based strength training device, a distance for pushing, pulling, or driving the sensor-based strength training device, an amount of weight lifted, a number of reps and sets, a power output, etc. In some cases, the sensor-based strength training device can determine and provide a metric representing a personal record of the athlete’s performance of an exercise with the sensor-based strength training device, e.g., a personal best. The sensor-based strength training device can assist the athlete with goal-setting to improve their personal record of an exercise, e.g., beating the athlete’s personal best for an exercise, by providing real-time tracking of the athlete’s performance and feedback to indicate additional effort is demanded to exceed a previous personal record.

[0103] The technology described in this specification also relates to a training device (include a smart, sensor-based training device) with a reduced contract area on at least one end of the training device, such as by including a bottom portion with a disk, single ski, etc. The training device introduces an instability factor for an athlete to train their stability when performing an exercise action. By introducing an instability factor, the disclosed training device allows the athlete to train and improve coordination, muscle activation, functional strength, and other athletic performance metrics. As another example, the instability factor from the training device can allow the athlete to reduce joint stress and / or muscle imbalance, as well as increase flexibility, by performing exercise actions with the training device. The disclosed training device allows an athlete to train with exercises that simulate complex moves, such as those in American football. For example, theAttorney Docket No.: 36610-0120W01athlete can practice movements that simulate driving back a defender and applying torque to laterally displace the defender, i.e., throwing the defender off balance to fall on the ground or throwing a defender off balance and then throwing the defender onto the ground. By utilizing a training device with a reduced contact area with the ground, the training device allows for a wide range of blocking techniques and complex exercise actions to be performed.

[0104] This specification also describes an example frame structural member for a strength training device, such as a frame structure with an adjustable length and width. The frame structural member can be a telescoping structure where an inner portion of the frame structural member is contained in an outer portion, and the inner portion is slidable, e.g., along an axis. An adjustable length for the frame structural member allows for an increase or decrease of the amount of force demanded without adding or removing weights, e.g., plates. For example, increasing the length of the frame structural member can increase the amount of force without adding additional weight, by increasing the length where a force for an exercise action is applied. Increasing the length of the frame structural member can also allow for additional weight to be added, by adding weight pins to hold weight plates. Shortening the length of the frame structural member can decrease the amount of force without removing weight, by decreasing the length where a force for an exercise action is applied. The disclosed adjustable configuration of the frame structural member allows for the training device to be more compact and reduce the amount of space, e.g., for easier storage.

[0105] The disclosed training device can include a weight adjustment mechanism that can adjust the amount of weight for the training device, such as by a slidable mechanism corresponding to different weights based on the position of the slider in the slidable mechanism. The weight adjustment mechanism can be a weight-stack pin adjustment system, where plates of weight with selector pins that can be inserted in holes disposed into a weight plate and the pin can be inserted in an opening of at a target weight plate to configure the weight of the training device. The disclosed training device with adjustable weight mechanisms expands locations on the training device for adding weight, such as closer to or further away from a pivot point, e.g., to make the training device lighter or heavier, respectively. The disclosed frame structural member can also include compartments to allow for additional customization of weight displaced throughout the training device.

[0106] The training devices disclosed in this specification can also include bumpers to improve safety by absorbing shock (and other types of forces) when the training device lands on the ground.Attorney Docket No.: 36610-0120W01For example, one or both ends of the training device can include bumpers such that when performing exercise action, the bumpers land on the surface and absorb resulting forces such as shock or impact. The disclosed training device can also include damping systems to reduce the forces between the training device and a surface for performing the exercise action, e.g., to allow the training device to land more softly. The damping systems reduce noise, reduces wear and tear of the surface of the training device, and reduces wear and tear for the training device (including any of its components). Examples of damping mechanisms include spring-systems, hydraulic systems, pneumatic systems, gel-based damping, magnetic damping, and friction damping, among others. The disclosed training device can also include damping materials such as rubber, elastomers, foams, hydraulic fluids, springs, viscoelastic materials, and components.

[0107] This specification also describes a smart training device with attachments that allow an athlete to perform complex exercise actions, such as by simulating expected blocking impact that occurs during American football training and games. Examples of the attachments can include medicine balls, blocking dummies, and other types of materials of varying shapes and sizes. A medicine ball attachment provides a combination of an ergonomic advantage of the round shape associated a medicine ball with dynamic rotation capabilities provided a rotating mechanism connecting the medicine ball to the training device. The medicine ball attachment allows for the athlete to train and improve hand placement, by practicing hand-placement when performing the exercise action, e.g., for low-blocking techniques and applying leverage. The disclosed attachment for the training device can allow techniques that are in demand for some types of athletes, such as linemen in American football.

[0108] The technology described in this specification also includes a propulsion-based exercise training device (also referred to as an exercise device or training device) configured to provide propulsion that allow the training device to mimic complex movements of players, e.g., offensive moments. The training device can be configured to generate a target amount of propulsion to safely translate the device along a surface to mimic a lineman, running back, etc. in American football. The athlete can interact with the propelled training device to practice blocking techniques. The training device can include sensors to detect changes in motion and instability to trigger automatic braking, e.g., to prevent collisions. The training device can also include proximity sensors, load sensors, etc. to detect obstacles or unsafe weight distribution and effectuate controls to prevent injury. The propulsion-based exercise training device can reset exercise positions to allow anAttorney Docket No.: 36610-0120W01athlete to perform the same exercise in less time, e.g., reducing the need to move the exercise device into position. The propulsion-based exercise training set can also remove the need for another athlete, allowing an athlete to perform an exercise that would otherwise demand two or more athletes.

[0109] The details of one or more embodiments of the subject matter of this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0110] FIG. 1 A is a diagram of an example sensor-based strength training device.

[0111] FIG. IB is a diagram of example exercise actions using a sensor-based strength training device.

[0112] FIG. 1C is a perspective view of an example frame structural members for a sensor-based strength training device.

[0113] FIG. ID is atop view of the example frame structural members for a sensor-based strength training device.

[0114] FIG. IE is a diagram of an example attachment for a strength training device.

[0115] FIG. IF is a diagram of an example damping system for a strength training device.

[0116] FIG. 2 is a block diagram of an example computing environment for a sensor-based strength training device.

[0117] FIG. 3A is a flow diagram of an example process for processing sensor measurements from the sensor-based strength training device.

[0118] FIG. 3B is a flow diagram of an example process for processing sensor measurements from a strike pad for the sensor-based strength training device.

[0119] FIG. 3C is a flow diagram of an example process for processing sensor measurements from a strike pad adapted to be removably connected to a strength training device.

[0120] FIG. 3D is a flow diagram of an example process for processing sensor measurements from a sensor of a strike pad for a strength training device.

[0121] FIG. 4A is a diagram of an example strength training device.Attorney Docket No.: 36610-0120W01

[0122] FIG. 4B is a diagram of another example strength training device.

[0123] FIG. 5A is a diagram of another example strength training device.

[0124] FIG. 5B and 5C are diagrams of an example attachment for the strength training device of FIG. 5A.

[0125] FIG. 6 is a diagram of another example strength training device.

[0126] FIG. 7 is an example computer device for the sensor-based strength training device form training system.

[0127] Like reference numbers and designations in the various drawings indicate like elements.DETAILED DESCRIPTION

[0128] Due to the competitive nature of sports, there is a continual demand to optimize exercise performance. The disclosed sensor-based strength training device can capture the dynamic nature of explosive movements in complex exercises, analyze joint angles and movements patterns for capturing motions in the exercises, and provide feedback to improve the athlete's overall performance. The device can provide performance feedback tailored to a particular athlete's performance in real-time, thereby identifying opportunities for the athlete to improve and focus on a particular phase of the exercise. By identifying and evaluating a phase of the exercise to improve the athlete's performance, the device allows for the athlete to train more efficiently compared to without the smart device.

[0129] A sled is an example of a strength training device that an athlete can use to perform different types of exercises, by performing an action with the sled, e.g., relative to a surface of an athlete’s training environment. For example, some complex exercises for American football training can include performing a “clean, drive, flip” (CDF) with the sled. A CDF is where the athlete performs a clean with the sled, e.g., by lifting one end of the sled using an overhand grip to lift the sled from the ground at an angle, then performs a drive with the sled, e.g., by pushing the sled along a ground surface while holding the sled at an angle, and then flips the sled, e.g., by pushing the end of the sled that is being held so that the sled falls over towards the direction the sled was being pushed towards. Other examples of complex exercises using a strength training device (such as a sled) can include dragging and pulling the strength training device, performing a clean and a press using the strength training device, as well as deadlifting the strength trainingAttorney Docket No.: 36610-0120W01device, among other examples described in this specification. These types of exercises (also referred to as a sequence of power moves) can include multiple phases, which can include a number of pulls, pushes, positioning and / or re-positioning (e.g., stance, grip, posture), recovery, etc. to perform the exercise.

[0130] By collecting and processing sensor measurements of the athlete and the exercise, e.g., images, videos, audio, and other types of sensor measurements, the smart device can provide a data-driven feedback mechanism for improving athlete performance. For example, these movements can include static or dynamic lifting, pushing, flipping actions, etc., and the sensor measurements can quantify these actions in exercises, thereby allowing for the identification of performance deviations. An athlete’s performance can degrade due to fatigue, lack of practice, injury, imbalanced training load, poor hydration and / or nutrition, among other reasons for declining performance. The smart device can identify deviations from metrics based on the sensor measurements representing an athlete’s performance and compare the metrics to reference metrics for the exercise. Examples of reference metrics can include historical metrics for the same athlete, ideal metrics for the exercise, e.g., for a competition level performance. The smart device can also allow for comparison of performance metrics between different athletes. This can allow for realtime feedback and corrective action for the athlete, summary and statistical information about the exercise session, and accurate performance tracking for athletic training.

[0131] Inadequate (e.g., improper, incorrect) performance by an athlete can result in the athlete being unable to successfully perform a simple (e.g., a single movement) or complex motion (e.g., a combination of movements) for an exercise in a particular support. For example, an athlete performing an exercise (e.g., an exercise action) with inadequate performance, e.g., effort, form, may be unable to push or pull an item of training equipment such as a sled and may also be unable to lift the sled and / or push the sled after lifting to flip the sled. In some cases, inadequate performance can include an insufficient amount of force or excessive force, resulting in poor athletic performance, e.g., the athlete may fail to achieve a target performance metric demanded for a particular sport. For general fitness, a person with inadequate performance may be unable to lift a strength training device with the desired weight but may also risk inadequate training for personal fitness, delaying progress in their fitness journey. As another example, a first responder (e.g., police, firefighter) may perform physical training that include complex actions using strength training devices, and inadequate performance in these scenarios can also result in poorAttorney Docket No.: 36610-0120W01performance, e.g., for rescue operations, team exercises, including scenarios where lives may be at risk.

[0132] In some cases, poor performance metrics can result in the athlete performing the exercise action without engaging a target muscle or muscle group that the exercise action is designed to train, resulting in an exercise action that does not meet the target performance metrics. As another example, poor performance of an exercise includes performing the exercise without performing the full range of motion associated with the exercise. While the exercise may be fully performed by the athlete without completing the full range of motion, the athlete may fail to meet performance metrics for the exercise because the benefits associated with performing the full range of motion are not fully realized. By analyzing and generating evaluations of athletic performance for the exercise, the disclosed technology reduces injury risk and damage to athletic equipment, improving safety for the athlete and extending the lifespan of exercise equipment, respectfully.

[0133] The disclosed technology allows for personalized, sensor-based metrics for evaluating an athlete’s performance of an exercise. While two athletes may both be able to successfully lift and push the strength training device, e.g., to flip the strength training device, one athlete may have a lower performance metric, e.g., lower output power, compared to the second athlete. With this information, an athlete with worse performance can improve, and the performance of two athletes can be quantitatively compared using objective, sensor-based measures.

[0134] FIG. 1A shows an environment 100 including an example sensor-based strength training device 104 with sensors. The environment 100 shows an athlete 102 and a strength training device 104, which can also be referred to as smart training device 104 or training device 104. The athlete 102 can perform a number of different exercises using the training device 104. The training device 104 can be, for example, an American football training sled. The training device 104 can optionally include additional components for athletic training, such as weighted plates, bumpers, handles, frame accessories, strike pads, etc.

[0135] The environment 100 also shows the training device 104 on a surface 103, e.g., the ground, and the athlete 102 can perform exercises such as dragging, pushing, etc., the training device 104 against the surface 103. In some cases, exercises can include lifting one end of the training device 104 upward and away from the surface 103. Exercises can also include flipping the training device 104, as described in reference to FIG. IB below.Attorney Docket No.: 36610-0120W01

[0136] The training device 104 includes a frame structural member 108 having a first end 107a and a second end 107b opposite of the first end 107a. The first end 107a and the second end 107b define a longitudinal axis 109 along a length of the frame structural member 108. Each of the first end 107a and the second end 107b can optionally include a respective top portion and a respective bottom portion. Although FIG. 1A depicts the training device 104 having a top portion Illa protruding upward from the first end 107a and a bottom portion 110a protruding downward from the first end 107a, each of the top portion Illa and the bottom portion 110a can be optionally included. Similarly, each of the top portion 111b protruding upward from the second end 107b and the bottom portion 110b protruding downward from the second end 107b can be optionally included.

[0137] As an example, a training device 104 configured to operate in a single mode as sled may include the top portions Illa and 111b, e.g., as handled to allow for pushing, pulling, dragging, etc., of the training device. The top portions Illa and 111b may be weight-loading pins that allow for additional weight to be added to the training device. The training device 104 also includes one or more weight loading pins 113 (also referred to as “weight pins”) that are part of (e.g., integrally formed with) or attached to the frame structural member 108. The one or more weight pins allow for additional weight (e.g., weight plates) to be added to the training device 104.

[0138] In some implementations, the training device 104 can be configured to operate in a single mode for flipping and include the bottom portions 110a and 110b but absent top portions 11 la and 111b. In a flipping configuration, the training device 104 can be absent of the top portion Illa, top portion 11 lb, bottom portion 110a, and bottom portion 110b.

[0139] As illustrated in FIG. 1A and referring to the training device 104, the first end 107a includes a top portion Illa and a bottom portion 110a while the second end 107b includes a top portion 111b and a bottom portion 110b. As depicted in FIG. 1A, the top portions Illa and 111b can be a set of handles (e.g., a pair of handles) protruding from the first end 107a and the second end 107b, respectively. For example, the top portion Illa can include a respective set of handles 115a, such as a horizontally oriented handle connecting one part of the top portion 11 la to another part of the top portion 11 la or the training device 104, e.g., parallel to the surface 103. As another example, the handles 115a for the top portion Illa can be a pair of handles, e.g., protruding outward from the top portion Illa. Similarly, the top portion 111b can also include a respectiveAttorney Docket No.: 36610-0120W01set of handles 115b, e.g., one or more handles for performing an exercise with the training device 104.

[0140] The bottom portions 110a and 110b are depicted in FIG. 1A as skis to reduce friction, improve traction and stability, as well as improve maneuverability of the training device 104, e.g., while an athlete pushes or pulls the training device 104 when performing an exercise. For example, each of the bottom portions 110a and 110b can include a pair of skis, e.g., a first left ski and a first right ski for the bottom portion 110a and a second left ski and a second right ski for the bottom portion 110b. In some implementations, each of the bottom portions 110a and 110b include a single ski.

[0141] In some implementations, each of the bottom portions 110a and 110b can be wheels that optionally include brakes and / or other types of dragging or braking mechanisms. This configuration allows for the training device 104 to be adapted to different types of surfaces, e.g., a surface 103. One or both of bottom portions 110a and 110b can include one or more wheels. For example, one or both of the bottom portions 110a and bottom portions 110b can include a single wheel, or multiple wheels, e.g., a pair of wheels. In some implementations, a bottom portion of the strength training device can include one or more wheels, e.g., a pair of wheels, while a different bottom portion of the strength training device includes a braking mechanism. For example, the bottom portion 110a can include a pair of wheels and the bottom portion 110b can include a braking mechanism. The wheels can facilitate easier movement and / or storage of the training device 104, by lifting the training device 104 at one end and rolling the device along the surface 103 on an end with wheels.

[0142] In some implementations, the strength training device can include one or more wheels that span a length or width of the strength training device, e.g., along the longitudinal axis 109 or perpendicular to the longitudinal axis 109. The one or more wheels can be made up of a material to mitigate forces experienced by the training device, such as rubber. For example, the training device can include a single wheel that spans a length or width of the strength training device. The training device include a pair of wheels that each span a length or width of the strength training device, e.g., each of the wheels being substantially parallel to each other. In some implementation, the strength training device includes multiple pairs of wheels, each pair being substantially parallel to another. In some implementations, the training device includes a single pair of wheels. In someAttorney Docket No.: 36610-0120W01implementations, a wheel can be shaped like a ski, e.g., relative long, and can be made of any material, e.g., rubber, polymer.

[0143] In some implementations, a first end of the training device can include wheels with a second end (e.g., opposite the first end) having a braking mechanism, skis, etc. In such a configuration, the load of the training device relies on the resistance on the second end, e.g., a skied resistance for the second end having skies, to bear a majority of the load of the training device. This may be preferably because, for instance, performing a single sled push can be difficult. Configuring the training device to have wheels on one end can allow for additional types of exercise actions to be performed, e.g., mimicking a pushing or pulling motion of a wheelbarrow, as well as allow for heavy loads to be lifted by the athlete.

[0144] In some implementations, accessories and subcomponents of the training device can be rotatable. For example, handles, skis, striking pads, and other components can be rotatable by a fixed range of motion, e.g., 90 degrees outward and away from the strength training device, 90 degrees inward and toward the strength training device. As an example, an axis 117 is illustrated in FIG. 1 A and a disk, ski, etc., from the bottom portion 110a can be rotated. Similarly, the handles at a top portion can also be rotatable. In some implementations, the accessories and subcomponents such as the handles, skis, etc., can be configured to rotate 360 degrees.

[0145] In some implementations, the skis at the bottom portion of the strength training device 104 can be caster-type skis, e.g., a ski coupled to the training device 104 by a caster-type assembly that includes a mount and allows the ski to be fixed to move along a straight line path or pivot to automatically align to a direction of travel. In some implementations, the strength training device 104 can include a pair of caster-type skis, multiple pairs of caster-type skis, etc.

[0146] In some implementations, the bottom portions 110a and 110b can include adjustable braking mechanisms that allow the training device 104 to mimic a target amount of friction between the training device 104 and the surface 103, with the same amount of weight (or lack thereof) configured on the training device 104.

[0147] For example, one or both of the bottom portions 110a and 110b can include adjustable braking mechanism that allow for the training device 104 to mimic the amount of friction associated with a particular weight and / or first type of surface 103, while the athlete 102 performs exercises using the training device 104 another a second type of surface 103 different than the first type of surface 103. Thus, the disclosed training device 104 can allow for performance of exerciseAttorney Docket No.: 36610-0120W01actions and fitness training for different types of surfaces, such as concrete, grass, sand, pavement, indoor floors, wood, laminate, rubber, etc. An athlete can perform exercises using the training device 104, which can use sensor measurements to determine a first type of the surface 103 and provide adjustments to the adjustable braking mechanisms in the bottom portions 110a and 110b to mimic the amount of friction that would be associated with another second type of surface 103, e.g., compared to the first type of the surface 103.

[0148] In some implementations, each of the bottom portions 110a and 110b can be a disk, with a flat facing portion of the disk oriented parallel to the surface 103. In this way, the training device 104 can be turned more readily, e.g., compared to a pair of skis. This allows for the athlete 102 to perform exercises that include twisting motions, also add twisting motions to exercises that may not have previously included a twisting motion. For example, an athlete 102 can perform a clean, drive, flip with the training device 104 to practice driving or blocking exercises in American football. Each part of the CDF can be performed along a same axis or direction, such as by performing a clean to lift one end of the training device 104 and driving the training device 104 along an axis in a substantially straight direction then flipping the training device 104. By adding disks to the bottom portions 110a and 110b of the training device 104, the training device 104 can be twisted and / or flipped in a direction different from the direction of the drive for the CDF. This provides the advantage of more accurately simulating moves in American football such as finishing a block, e.g., compared to the CDF that does not include twisting or turning motions. In some cases, the training device 104 can include additional accessories that allow for a degree or rotation (e.g., 90 degrees, 180 degrees, 360 degrees), e.g., handles on left and right sides of the first end 107a and / or the second end 107b to allow for side-to-side motions, twisting motions, and turning motions with a wide turn radius.

[0149] The training device 104 can include sensors 120-1 through 120-N (collectively “sensors 120”), e.g., any number of sensors. The sensors 120 can be located at one or more different portions of the training device 104. For example, sensor 120-1 is located at the first end 107a of the training device 104 and sensor 120-N is located at the second end 107b of the training device 104. In some implementations, a sensor from the sensors 120 can be mounted onto a surface of the training device 104. In some implementations, a sensor from the sensors 120 can be embedded in a surface of the training device 104. A third sensor 120-3 is depicted relatively closer to the bottom portion 110a ofthe first end 107a, e.g., in contrast to the first sensor 120-1 is depicted in FIG. 1 A relativelyAttorney Docket No.: 36610-0120W01closer to the top portion 11 la of the first end 107a. The training device 104 can also include a fourth sensor 120-4, depicted on a left-hand side of a central portion of the frame structural member 108 connecting the first end 107a to the second end 107b. Similarly, the training device 104 can also include a fifth sensor 120-5 depicted on a right-hand side of a central portion of the frame structural member 108 connecting the first end 107a to the second end 107b. In some cases, the training device 104 can include a sixth sensor 120-6 depicted at an interface or surface of training device 104 that contacts the ground, e.g., an area below the training device 104.

[0150] In some implementations, the handles of the training device can include sensors to capture sensor measurements for estimating force, e.g., measuring forces applied to the handles of the training device 104 to estimate the force applied to the training device 104. For example, FIG. 1A depicts a handle from the set of handles 115a having a sensor 120-7. The sensor measurements can also be processed by a computing device to determine a metric for the exercise action, e.g., speed of the exercise action, applied force, weight, etc. In this way, the training device 104 (e.g., using the computing device to process sensor measurements) can determine friction between the training device 104 and a surface 103 for using the training device 104 for exercise action(s).

[0151] Sensors for measuring force (such as sensors in the handles) can include load cells configured to convert a force (e.g., weight, pressure) into an electrical signal for processing by the computing device. For example, load cells can include strain gauges configured to adjust electrical resistance when subjected to forces such as stress, strain, etc. In some implementations, the load cells can include piezoelectric materials to adjust the electrical resistance when subjected to forces. The change in electrical resistance can be amplified and calibrated to provide a measurement of the applied force. Load-cell based sensors can be incorporated in components of the training device 104 where an athlete can grip the training device 104 to perform an exercise action. In this way, the disclosed technology can provide a measure of force applied by the athlete when performing an exercise action with the training device 104, e.g., pushing, pulling, lifting and / or flipping. The sensors can provide a consistent measurement of force independent of the coefficient of friction between the training device 104 and the surface 103.

[0152] Load cells can be integrated into the training device 104 to provide real-time power output, thereby providing real-time performance tracking and training. Data collected by load cells can be used to determine and power metrics, allowing for explosive power and force to be measured and improved upon through training with the training device 104. For example, loadAttorney Docket No.: 36610-0120W01cells can be used to measure force, velocity, etc., to provide consistent performance metrics independent of the surface type. For example, the load cells can capture force measurements of an athlete performing an exercise action with the training device 104 on a surface with a lower coefficient of friction, e.g., wet grass, compared to a surface with a higher coefficient, e.g., dry turf, and a computing device 124 for the training device 104 determines the amount of force that incorporates differences in the coefficient of friction, e.g., to provide a surface-independent comparison of exercise performance. By providing real-time force information about an exercise action, the training device can provide opportunities for the athlete to perform real-time or near-real time adjustments of the exercise action and improve their overall athletic training. The measurements provided by load cells and other types of sensors can provide performance benchmarks, e.g., watts, data-based progression, remote-based coaching, and the ability to integrate with other strength and / or conditioning platforms.

[0153] For instance, load cells can be integrated into handles of the training device 104 to measure horizontal force applied during a sled push exercise action performed using the training device 104. In some cases, the horizontal forces measured by the load cells can be combined with velocity, accelerator, and other types of measurements captured by an accelerometer and / or an inertial measurement unit (IMU). As another example, load cells can be integrated into another set of handles for the training device to capture vertical and horizontal forces during the lift phase of a flip exercise action performed using the training device 104. Movement speed, angles, and other measurements can be combined with force data, e.g., from the vertical and horizontal forces, to compute an output wattage indicative of explosive lift power and / or finish power when performing a flip exercise action. In some implementations, the training device 104 includes a display, e.g., a screen, to output measurements and other types of values, performance metrics, etc. In some instances, the training device 104 also include a wireless module for integration with cloud platforms for exercise training. In some cases, the module can be a Bluetooth module configured to transmit and receive data between the training device and a client device, e.g., a computing device or mobile device for an athlete.

[0154] In some implementations, the sensors 120 can include integrated sensors that combine multiple types of sensors. For example, a load cell can be combined with accelerometers to improve performance tracking and training with the training device 104, such as by capturing the real-time power, e.g., in Watts, generated while performing exercise actions with the trainingAttorney Docket No.: 36610-0120W01device, e.g., tire flips, sled pushes. By combining a first type of sensor, e.g., configured to measure force or weight, and a second type of sensor, e.g., configured to measure accelerator, vibration, etc., to provide synchronized force and motion measurements. An advantage of integrating multiple sensor types can include providing improved real-time sensor measurements, as both sensors are synchronized to each other. In this way, sources of sensor inaccuracy such as timing errors can be reduced in the sensor measurements. In some implementations, a sensor can be an inertial measurement unit based on a combination of sensors to provide multi-axis force measurements.

[0155] The sensors 120 are configured to capture sensor data, such as sensor measurements, detections, etc. of the athlete 102 prior to, during, and / or after completing, an exercise with the training device 104. The training device 104 includes a computing device 124 configured to process the sensor measurements and determine performance metrics of an exercise. The output of the computing device 124 of the training device 104 is shown as exercise performance output 122 (also referred to as “output 122”), which can be transmitted to one or more user devices, e.g., smart watch, smartphone, laptop computer, to provide a metric describing, e.g., quantifying, the performance of an exercise by the athlete 102 using the training device 104.

[0156] The sensors 120 can be configured to measure forces, speeds, accelerations, etc. for different types of forces, including linear forces, rotational forces, and angular forces experienced by the strength training device. For example, the sensors 120 can be configured to capture sensor measurements of linear forces experienced by the training device 104, e.g., translation motions of the training device 104 along a direction, such as applied forces, tension, friction, normal forces, and inertia-induced forces. As another example, the sensors 120 can be configured to capture sensor measurements of rotational forces experienced by the training device 104, e.g., rotation about an axis of the training device 104, such as different types of torque, e.g., inertia-induced, applied torque, tension-induced torque, frictional torque. As another example, the sensors 120 can be configured to capture sensor measurements of angular forces experienced by the training device 104, e.g., change in rotational motion of the training device 104, such as angular acceleration and angular momentum.

[0157] One or more of the sensors 120 may be located on different portions of the training device 104, e.g., surfaces of the training device 104, as well as at the top portion 1 Ila, top portion 11 lb, bottom portion 110a, bottom portion 110b, or some combination thereof. The one or more sensorsAttorney Docket No.: 36610-0120W01can be configured to detect impact forces of the training device 104 as the training device 104 interacts with the surface 103. For example, the training device 104 can be lifted, flipped, etc., and the sensors 120 can capture sensor measurements of any resulting impact forces to determine how “hard” the sled is being flipped.

[0158] A sensor from the sensors 120 can include one or more hardware devices capable of receiving input from the environment and converting that input into data accessible by computing elements. For example, the one or more sensors 120 can capture image data, point cloud data, or another format of data capturing two-dimensional (2D) or three-dimensional (3D) information of the training environment, e.g., including an athlete. The sensor 120 can capture temporal data of the training environment, e.g., to capture exercise actions performed by an athlete over time.

[0159] Although FIG. 1A depicts an example of different locations for the sensors 120 on the training device 104, the sensors 120 can be located anywhere on the surface of the device, embedding in the device, etc. Referring to the computing device 124, the computing device 124 can include graphical processing units (GPUs), application-specific integrated circuits (ASICs), digital signal processors (DSPs), and other types of electronic circuitry. The computing device 124 can include processors, other types of hardware, software, or some combination thereof.

[0160] The training device 104 can also include a number of additional components that can also include sensors. For example, the training device 104 can include an attachment 112 and a sensor 120-2 mounted onto or embedded in the attachment 112. Examples of the attachment 112 can include strike pads, blocking pads, tackling dummies, resistance harnesses, agility arms, shield attachments, etc. For example, the attachment 112 can be a strike pad attachment 112 that includes one or more sensors, e.g., sensor 120-2 to capture force measurements, among other types of measurements. The computing device 124 can be configured to determine an output 122 indicating metrics such as reaction time, sustained force, etc., for striking the strike pad attachment. The strike pad attachment 112 can be any shape or size. For example, the strike pad attachment 112 can be in the shape of a human body, e.g., a defender in American football, and can include a number of sensors to track movements and forces from different positions on the strike pad attachment 112.

[0161] The sensor measurements captured by sensors of the strike pad attachment 112 can be processed by the computing device 124 to determine if an athlete 102 is striking one or more particular locations on the strike pad attachment. In the example of American football, striking a particular location on the strike pad attachment may be desirable to successfully achieve the targetAttorney Docket No.: 36610-0120W01exercise, e.g., blocking or interrupting opponent’s actions in American football. As another example, the training device 104 can include exercises to determine a distribution of the force exerted onto the training device 104 by the athlete 102 based on the sensors being at different locations of the attachment, e.g., a left-hand side of the attachment 112 compared to a right-hand side of the attachment 112 or a top portion of the attachment 112 compared to a bottom portion side of the attachment 112.

[0162] Although FIG. 1A illustrates a single computing device 124, the computing device 124 can be a distributed computing system, e.g., the training device 104 can have multiple computing devices 124 with a single computing device 124 configured as a master computing device 124 and the remaining computing devices 124 configured as non-master computing devices 124. For example, a non-master computing device 124 can be co-located with a sensor from the sensors 120 to provide edge computing, e g., applying signal processing techniques to raw sensor measurements captured by the sensor. The processed sensor measurements can be transmitted to the master computing device 124 for further processing, e.g., to determine metrics associated with an exercise and / or the athlete.

[0163] The computing device 124 determines metrics associated with an exercise performed by the athlete 102 using the training device 104. For example, metrics can include a value representing output power, e.g., wattage power, based on a force and velocity of an exercise movement using the training device 104. Power can also be based on an amount of work performed over a period of time. Examples of exercises performed using the training device 104 are described in reference to FIG. IB below. In some implementations, the computing device 124 can compute a graphical visualization of metrics associated with an exercise and transmit data related to the visualization to a user device. In some implementations, the computing device 124 includes a screen or any type of graphical display to show the visualization. In some cases, the computing device 124 can determine numerical values for output.

[0164] As another example, the training device 104 can be used as a sled and the computing device 124 can determine, based on sensor measurements from sensors 120, an output 122. Examples of the output 122 for a sled exercise can include wattage output (e.g., power), speed (e.g., how fast the sled was pushed), distance (e.g., how far the sled was pushed), time (e.g., how long the sled was pushed), acceleration, etc. As another example, the training device 104 can be used for tire flipping exercises and the computing device 124 can determine an output 122 such asAttorney Docket No.: 36610-0120W01wattage output (e.g., power), counts (e.g., counting a number of flips), distance, duration, speed, etc. In some cases, the output 122 for exercises that include rotation, twisting, etc., can include metrics such as torque, moment of force, angular momentum, shear stress, tension, etc.

[0165] In some cases, wattage output can indicate how quickly an athlete can move a certain amount of weight for the training device 104, e.g., the training device 104 by itself or the training device 104 with additional weights. In some implementations, the computing device 124 can determine an amount of power for a particular phase of an exercise. For example, the athlete 102 can perform a clean, drive, flip (CDF) with the sled and the computing device 124 can determine an output 122 indicating a first amount of power associated with the drive phase and a second amount of power associated with the finishing phase. In some cases, the computing device 124 can determine an output 122 that includes rolling statistics of metrics. In some cases, the output 122 can indicate a maximum power output, e.g., indicating how hard an athlete can throw or move weight associated with the training device 104.

[0166] In some implementations, the training device 104 can include audio and / or graphical devices that can communicate commands, feedback, etc., to the athlete 102. For example, the training device 104 can include an audio device that emits sounds to instruct the athlete 102 when performing the exercise. For example, the training device 104 can transmit a first sound to indicate that the athlete can proceed with performing the exercise and a second sound to indicate that the athlete can conclude performing the exercise. In some cases, the second sound can indicate the conclusion of a phase of an exercise and that the athlete 102 can proceed to the next phase of the exercise. In this way, the training device 104 can provide timing training to the athlete 102 for performing the exercise, e.g., to improve metrics related to the duration or timing of an exercise. In some implementations, the commands and / or feedback can be provided to the athlete 102 through a graphical display. The graphical display can indicate, using colors, texts, symbols, images, or some combination thereof, messages to the athlete 102.

[0167] The training device 104 including audio devices and / or graphical displays can provide a guide for the athlete 102 to perform the exercise, which is particularly suited for American football training. Some exercises in American football can demand performing a clean using the training device 104 within a first time period, driving the training device 104 for a number of yards within a second time period, and flipping the training device 104 within a third time period. As an example, the training device 104 can provide instructions to the athlete 102 to perform the cleanAttorney Docket No.: 36610-0120W01using the training device 104 within one and a half seconds, drive the training device 104 for ten yards, and flip the training device 104 within one second. The training device 104 allows the athlete 102 to perform the exercise regardless of the environment 100, which may not have timing mechanisms (e.g., stopwatches, clocks, timers) or visual indicators (e.g., markings on the surface 103) to guide the athlete 102 through the exercise.

[0168] In some implementations, the training device 104 can be communicatively coupled to a network, e.g., the Internet. In this way, the training device 104 can provide internet-based workout programs that can be continually updated to meet a particular performance output, e.g., output 122. In some implementations, the training device 104 can be configured to provide output 122 for a training profile associated with the athlete 102. Profiles of metrics for two different athletes can be compared based on their profiles. In some cases, the training device 104 can receive a training template indicating a number of different exercises. The exercises can be communicated to the athlete 102 by training device 104 and the athlete 102 can perform the exercises as part of the training template. Examples of internet-based workouts and templates can be tailored to a particular type of training such as high-interval intensity training, Tabata workouts, power training, calorie burning, strength, agility, conditioning, among others.

[0169] The computing device 124 can determine metrics for exercises in the training example based on sensor measurements collected during the exercises and the metrics can be provided for output to one or more user devices. The computing device 124 can determine metrics for different profiles, each associated with a different athlete, and determine a ranking of the athletes based on the respective performance metrics for each athlete. The athlete rankings can be use to compare performance metrics between different athletes, e.g., for collegiate recruiting.

[0170] In some implementations, training templates and related workouts can be based on metrics that include maximum calories burned, duration of time, distance, number of flips, heart rate, and / or other physiological measurements captured by the sensors 120. The training device 104 can provide a feedback output to indicate to the athlete 102 one or more recommendations that can be made to achieve a target performance output, e.g., output 122. Examples of recommendations can include changing the amount of weight, such as by adding or removing weights, as well as adjusting an amount of time for performing a phase of an exercise, e.g., by shortening or increasing the amount of time spent in a phase of the exercise. In some implementations, the training deviceAttorney Docket No.: 36610-0120W01104 can indicate to the athlete 102 to perform as many reps (or rounds) as possible for in strength training or high-intensity interval training.[001711 In some implementations, the computing device 124 for the strength training device 104 can be configured to generate workout performance data that allows for competition between multiple users, and / or for a user to compete against with their previous performance. The performance data can include any exercise action, such as those described in reference to FIG. IB below. Examples of competition can include a first user performing the same or similar workout as a second user and the computing device 124 evaluating a performance of the first user in relation to the second user. In some cases, the second user is the same person as the first user — e.g., performing a workout at a different time or different location.

[0172] In general, during a workout, the sensor-based strength training device 104 can track movement of the athlete 102 while they are exercising, e.g., using sensors, to generate metrics evaluating the performance of the athlete for the workout. Evaluation can include using previously tracked data as a baseline to evaluate a current performance — e.g., allowing players to compete against themselves performing the same or similar workouts through previously captured sensor data, leaderboard ranking — e.g., global or local ranking corresponding to workout, drill within workout, etc. During the workout, the system can track a user while they are exercising, e.g., using sensors 120, to identify specific areas for improvement, evaluate and provide evaluation results, or make adjustments in a workout. In some cases, the disclosed technology allows players to compete against prior recorded workouts of themselves performing the same or similar workouts. Allowing such competition can increase a rate of improvement for players.

[0173] The training device 104 can provide responsive, dynamic adjustment of exercise training to the athlete 102 by generating workout signatures that allow for performance tracking, e.g., between different instances of performing the same exercise. For example, the computing device 124 of the training device 104 can provide results of evaluating the generated signatures. Results can include one or more of adjustments to subsequent workouts, drills, etc. that are determined by the system using obtained sensor data, as well as recommendations for technique adjustments, specific drills, progress indication, etc.

[0174] In some cases, data recorded during a workout can be optimized or filtered, e.g., sensors of a system can be selectively activated during workouts. Selective activation can help to reduce storage of data that might be unused. In general, selective activation can reduce storageAttorney Docket No.: 36610-0120W01requirements and required bandwidth, e.g., for providing data from a sensor to the computing device 124 or between multiple connected devices. In some cases, workout pace can be adjusted — e.g., based on sensed input from a user working out. By adjusting a pace of a workout, energy can be saved from faster operation of a sensor and / or the computing device 124, or using more power than demanded for a particular user at a particular time.

[0175] In some implementations, the athlete 102 competes against themselves. For example, the computing device 124 can obtain a signature of the athlete 102 performing a same drill or same workout using the training device 104. The signature can be generated by the computing device 124, e.g., during a previous workout performed by the athlete 102. The computing device 124 can then evaluate the athlete 102 against the previously generated signature. Evaluation can include indicating whether a particular performance metric of an exercise action was met or missed. Evaluation can include providing data to one or more machine learning models to determine whether a current workout is an improvement or decline compared to the previously generated signature, e.g., from sensor-based performance metrics.

[0176] For example, one or more models described in reference to FIG. 2 below can be trained using one or more signatures labeled with one or more values indicating an improvement or decline compared with one or more other signatures. In some cases, each signature can be represented using a vector or single value. A machine learning model, e.g., model 206 described in FIG. 2 below, can be trained to generate a vector or single value that represents a signature — e.g., provided a signature as input and output a vector or single value that is compared with labels for corresponding signatures. A distance between a vector or single value of a previously generated signature and a current workout of the athlete 102 can then indicate an evaluation of the athlete 102 — e.g., where a positive difference can indicate improvement and negative difference can indicate a decline.

[0177] The computing device 124 can adjust indications of the performance based on a time of the workout. For example, the value of personal record (PR) can be updated to show the value previously recorded for the PR, or baseline, for that point in the previously recorded workout. In some cases, PR can show any other suitable values representing previously recorded workout data. Similarly, the value of the current workout can be adjusted based on sensor data captured of the athlete 102. In this way, the athlete 102 can compete, in real time, against themselves or another user.Attorney Docket No.: 36610-0120W01

[0178] In some implementations, the training device 104 can include a feedback mechanism with adjustable friction features at bottom portions of the training device 104, e.g., bottom portions 110a and 110b. Because some exercises can be dependent on the friction between the training device 104 and the surface 103, the training device 104 can have adjustable friction mechanisms to achieve a target amount of friction to standardize exercises between different athletes and / or surfaces. The training device 104 can be configured to determine, e.g., estimate, a coefficient of friction of the surface 103 based on sensor measurements captured during an exercise with the training device 104, e.g., dragging, pushing, or pulling the training device 104 against the surface 103. In this way, the training device 104 can provide adjustable friction according to a type of surface for performing the exercise action to mimic performing the exercise action on a different type of surface with a different coefficient of friction. For example, the training device 104 can be configured to generate a value representing an amount of weight to add or remove to the training device 104 based on a particular coefficient of the surface for performing the exercise action, such that adjusting the amount of weight for the training device mimics performing the exercise action on a different surface having a different coefficient of friction. In some implementations, the value may also be based on a performance metric for a previous exercise action, such as an output wattage (or another form of output power).

[0179] The training device 104 can include a weight adjustment system, dynamic skid plates, adjustable resistance pads, drag brakes, or some combination thereof, to adjust the amount of friction between the training device 104 and the surface 103. The training device 104 can provide adjustable friction to provide an equivalent amount of friction based on the type of surface. For example, the surface 103 can have a particular coefficient of friction based on the type of surface 103, e.g., artificial grass, grass (including dry grass, thick grass, etc.), hybrid turf, turf, sand, dirt, asphalt, foam, rubber, wood, among other types of surfaces. The training device 104 can be configured to determine a coefficient of friction of the surface 103 and adjust the friction levels of training device 104 to achieve a target amount of friction. In this way, the training device 104 can standardize and calibrate exercises across different types of surfaces. In some implementations, the training device 104 can generate an output indicating an amount of weight, e.g., to increase or decrease an amount of friction.

[0180] In some implementations, a sensor from the sensor 120 is a camera, e.g., CMOS, CCD, or other device capable of capturing image data and / or video data. The sensor 120 can beAttorney Docket No.: 36610-0120W01configured to capture image data and / or video data in visible wavelengths, infrared wavelengths, or other spectra. In some implementations, a sensor from sensor 120 can be an acoustic sensor capable of capturing acoustic signal data. For example, a sensor 120-1 can be an ultrasonic sensor configured to capture ultrasonic frequencies reflected from an athlete’s body and capture sensor measurements of the athlete using the reflected ultrasonic frequencies. The sensor 120-1 can include an ultrasonic source, e.g., an ultrasonic transducer, to detect objects at distance of detection, e.g., up to about 25 meters away from the source. In some examples, an ultrasonic sensor can be integrated with a source, where the transducer of the source functions as a microphone to receive and send the ultrasonic waves and measures the distance to a target by measuring a time lapse (e.g., time of flight) between sending / receiving the ultrasonic pulse. Ultrasonic waves can have the advantage of being able to detect clear or transparent objects, e.g., liquids, and can be stable over a large array of surface finish, material, and color. Ultrasonic-based sensing subsystems can be insensitive to ambient lighting conditions. In some examples, a narrow beam ultrasonic sensor can be used to provide increased resolution within a distance range useful for the training environment. In another example, a sensor can be configured to capture low-level acoustic signals reflected from and / or transmitted through an athlete’s body and capture sensor measurements of the athlete 102 performing the exercise using the low-level acoustic signals. In some implementations, the training device 104 can be configured to capture sensor data of an exercise action, determine performance metrics, and provide real-time or near real-time feedback to assist the athlete to improve the performance of the exercise action. For example, the training device 104 can capture data to help calibrate an athlete’s performance of an exercise action, also referred to as a “calibration period” for the exercise. For example, the training device 104 can capture data while a drill is being performed, e.g., for an exercise action or a sequence of exercise actions and provide feedback to the athlete while they perform the drill to adjust their technique. Examples of feedback can include indicators (e.g., visual, audio) provided to the athlete, e.g., by a display and / or speakers of the training device, to indicate to the athlete to perform an adjustment while performing an exercise action, such as slowing down, speeding up, angling the device upward or downward, maintaining a particular pace, adjusting weight of the machine, among other adjustments. In some implementations, the strength training device 104 includes distance sensors such as a light detection and ranging (LIDAR) sensor, a time-of-flight sensor, etc. The data, images, and / or video captured by the image sensor may also be processed to reduce the noise and / or vibration in theAttorney Docket No.: 36610-0120W01signal using. In some implementations, a sensor 120 collects pre-data whi ch the computing device 124 can process to extract sensor measurements of the athlete. For example, a sensor 120 can collect point cloud data from which the computing device 124 can pre-process, e.g., filter, and extract biometric data for the athlete. In some implementations, the computing device 124 can pre-process the sensor data to reduce si nal-to-noise, filter, or otherwise clean up the sensor measurements prior to computing the output 122. The system can pre-process the data by flattening 3D sensor data to a 2D space. In some implementations, a sensor 120 from sensors 120 can be a radar sensor, e.g., mmWave sensor, collecting radar data.

[0181] In some implementations, the training device 104 includes two or more sensors 120 of a same or different type relative to each other. In some examples, the training device 104 includes at least one camera and at least one non-visible electromagnetic radiation sensor. In such examples, the training device 104 can collect multi-modal data input, e.g., image data and point cloud data, which can be used collectively by the system to generate performance metrics.

[0182] In some implementations, a sensor 120 (e.g., one or sensors from sensors 120-1 through 120-N) can be an image sensor configured to image data for facial recognition, e.g., by the image sensor and / or the training device 104. The image sensor can be a camera, a depth sensor, e g., an infrared sensor, a time-of-flight sensor to measure light reflection, or some combination thereof, to capture sensor measurements for facial recognition of a user of the training device 104. In some implementations, the image sensor can be a camera sensor configured to capture images of the athlete’s form prior to, during, and / or after performing the exercise action using the strength training device 104. In this way, the strength training device 104, e.g., by the computing device 124, can provide an output indicating the athlete’s form and / or efficiency prior to, during, and / or after performing the exercise action.

[0183] The sensor 120 and / or the computing device 124 can be configured to perform facial recognition of a user of the training device 104 to identify the user to a profile for their workouts, e.g., performing the exercise actions. By performing facial recognition, the training device 104 can execute applications upon recognizing the face of a particular athlete, such as loading a workout profile for the athlete based on the conditions of the environment.

[0184] The computing device 124 of the training device 104 can automatically log, e.g., store, sensor measurements and other types of collected data for the identified user. For example, the athlete can approach the training device and the training device 104 (e.g., by a sensor and / or theAttorney Docket No.: 36610-0120W01computing device) can identify the athlete, load a workout program for the athlete based on a longterm workout plan and / or environmental conditions (among other factors), and automatically store workout data and metrics related to exercise actions performed using the training device.

[0185] In some implementations, a sensor 120 can be a Global Positioning System (GPS)-enabled sensor to capture sensor measurements to determine position, speed, distance, acceleration, and other types of state information. For example, a GPS-enabled sensor 120 can provide real-time or near real-time measurement of state information for the strength training device 104 and / or the athlete 102. In some implementations, the training device 104 can include accelerometers, gyroscopes, GPS sensors, and other types of movement and orientation tracking sensors. For example, the sensors 120 can include accelerometers to measure changes in velocity, orientation, etc., to allow for acceleration and movement tracking of the training device 104 in multiple directions. Sensors 120 and / or the computing device 124 can be configured to determine speed, distance traveled, etc., during an exercise session with the training device 104. In some cases, GPS sensors can also be configured to capture data during an exercise session with the training device 104 and complement the accelerometer data. For example, the GPS enabled sensor can provide real-time or near real-time position information to track an athlete’s movement across an area or portion of the environment 100.

[0186] Examples of sensors 120 can include a non-visible wavelength source, e.g., emitter, and a sensor, e.g., detector, configured to detect reflections of the source from the training environment 100 surrounding the training device 104. In such instances, the wavelengths emitted by the source can be, for example, millimeter wave (mmWave), LIDAR, pulsed coherent radar, microwave radar, or other non-visible wavelength bands of the electromagnetic spectrum. An operating wavelength band of a sensor 120 can be selected in part using various factors, for example, accuracy, precision, sensing speed, resolution, and range of detection. The operating wavelength band can be selected, for example, based on a resolution requirement for classifying the different aspects of the exercise action.

[0187] Non-visible wavelength-based sensing subsystems can offer increased robustness to environmental factors over visible-based sensing, e.g., cameras. For example, non-visible based sensing can be robust to lighting conditions, weather interference (e.g., rain), contaminants such as dust or other particles, and the like. In some implementations, non-visible frequency-based sensing can be incorporated in addition to, or alternatively to, camera-based sensing in the strengthAttorney Docket No.: 36610-0120W01training device form training system. Non-visible sensing data, e.g., point cloud data, can be used instead of or to enrich visible-wavelength camera-based imaging data. In some examples, non-visible wavelength sensing subsystems can increase privacy of users by limiting feature resolution of the collected sensor data, e.g., not including facial information.

[0188] In some examples, a mmWave-based sensing subsystem can provide high-precision, low latency object (e.g., athlete and strength training device) sensing capabilities. A mmWave-based sensor 120 can provide added robustness, e.g., over alternative radar technologies, to environmental interference. In some examples, mmWave-based sensing can additionally include Doppler functionality, e.g., measuring speed / velocity of an athlete before, during, and after the exercise action. For example, a mmWave-based sensing subsystem can offer increased robustness to variations in lighting conditions, e.g., low light, dust / particles, and weather conditions such as rain or fog (if in an outdoor environment). In cases where mmWave-based sensing is used, the collected sensor data, e g., raw point cloud data, can be processed to protect the privacy of the athletes captured in the sensor data.

[0189] While described herein with respect to a sensor-based exercise tracking with a strength training device such as a sled, it should be understood that aspects of sensor-based exercise tracking can be applied to other types of strength training devices. Furthermore, although FIG. 1A depicts an example strength training device 104, the disclosed technology can be applied to other designs and configurations for an American football training sled, e.g., a flippable exercise device. In some implementations, the training device 104 can have a different shape or arrangement of components, e.g., the training device 104 can include components such as handles and skis for a sled configuration, compared to a training device 104 with components such as bumpers and skis for a tire flip configuration.

[0190] FIG. IB is a diagram of example exercise actions using a sensor-based strength training device, e.g., system 104. FIG. IB shows a view 130 with the athlete 102 performing three different exercise actions 106-1, 106-2, and 106-3. The view 130 shows the athlete 102 performing exercise action 106-1, which includes the athlete 102 approaching the training device 104 and then lifting one end of the training device 104 over an opposite end of the training device 104, e.g., to flip the training device 104. The training device 104 can obtain sensor data capturing data related to the performance of the exercise action 106-1, including sensor measurements of the athlete 102. TheAttorney Docket No.: 36610-0120W01training device 104 can capture sensor measurements during one or more time instances prior to, during, and / or after the exercise is performed.

[0191] For example, sensors (e.g., sensors 120) can capture measurements of position, velocity, acceleration (e.g., state information) of the athlete 102, as the athlete 102 approaches the training device 104. The sensors of the training device 104 can also capture measurements relating to the forces experienced by the training device 104 as the athlete 102 performs the exercise, such as the changes in state information, e.g., compared to before the athlete 102 makes contact with the training device 104. The sensors can capture measurements relating to forces applied by the athlete 102 (e.g., pulling, pushing, lifting), as well as other types of forces experienced by the training device 104. Similarly, the training device 104 can obtain sensor data, e.g., by capturing sensor measurements using sensors 120, of the athlete 102 as they perform the exercise action, which can include the forces applied by the athlete 102 while lifting and / or pushing the training device 104 and other forces experienced by the training device 104. The sensor measurements can also include changes in force during the performance of the exercise action. As another example, the sensors 120 can be configured to capture sensor information of the athlete 102 and the training device 104 after performing the exercise action 106-1. In this way, the computing device 124 of the training device 104 can determine metrics for the exercise action 106-1 performed by the athlete 102 to quantify characteristics of the performance of the exercise action 106-1, e.g., to evaluate the athlete’s performance and identify areas of improvement.

[0192] Examples of forces can include frictional forces, e.g., between the training device 104 and a surface under the training device 104 as the training device 104 is moved, normal forces, e.g., a force between the training device 104 and the surface supporting the training device 104, torque, e.g., rotational force applied to flip the training device 104, air resistance, e.g., resistance between the training device 104 and the air in the environment for the device, among other types of forces. A number of different types of metrics can be determined by the computing device 124 of the training device 104 for the exercise action 106-1, including metrics relating to force, energy, work, kinematics, rotational metrics, and mechanical resistance metrics, among other types of metrics. For example, the computing device 124 of the training device 104 can determine metrics for energy and work such as an amount of work done, power output, etc. As another example, the computing device 124 can determine kinematics metrics such as displacement, velocity, displacement, etc., as well as rotational metrics such as torque, angular velocity, moment of inertia, etc. OtherAttorney Docket No.: 36610-0120W01examples of metrics can include output wattage for the exercise action, counts (e.g., number of flips), power-to-weight ratio, impact energy, duration (e.g., amount of time), and other types of statistical measures (e.g., average force, average duration per flip), driving power, finishing power, etc. In some cases, the statistical measures can include rolling statistics (e.g., averages).

[0193] In some implementations, the sensors of the training device 104 can be configured to capture sensor measurements relating to the athlete’s pose, such as the stance or posture of the athlete 102. The computing device 124 can be configured to process the pose information from the sensor measurements of the athlete, in addition to information captured related to the device captured by the sensors, e.g., relating to forces experienced by the training device 104, to determine metrics related to the exercise action 106-1.

[0194] As another example, the view 130 shows athlete 102 performing an exercise action 106-2 using the training device 104. The exercise 106-2 can be a subset of exercise 106-1, e.g., only flipping the training device 104, rather than lifting, driving, and flipping the training device 104 shown in exercise 106-1. As another example, the view 130 depicts the training device 104 configured to capture sensor data of exercise action 106-3, e.g., using the sensors 120, to compute metrics relating to the exercise action. The view 130 shows athlete 102-1 and athlete 102-2 (collectively “athletes 102”) performing an exercise action 106-3 with the training device 104 together. The exercise action 106-3 can be an exercise to push or pull the strength training device 104 across a surface of the training environment, e.g., referring to environment 100. The computing device 124 of the training device 104 can be configured to capture multiple sets of sensor data for each of the athletes 102 performing the exercise action 106-2 and determine metrics for each of the athletes and / or both athletes. For example, the training device 104 can be configured to determine performance metrics for each athlete based on their respective contributions (e.g., placement, detected force) prior to, during, and / or after performing the exercise action 106-3.

[0195] The training device 104 can be configured to capture sensor measurements of the athlete’s poses, movements, stances, grip, and other aspects of the athlete’s performance to perform the exercise action 106. Exercise actions can include dragging and pulling the sled. For example, the athlete can drag and pull the sled, e.g., the athlete lifts the sled at an angle relative to a ground using an underhand grip, and then the athlete pulls the sled along the ground surface at the angle while moving backward. In another example, the athlete can perform a clean and press (e.g., a fullbody exercise that engages multiple muscle groups like the legs, core, and upper body of theAttorney Docket No.: 36610-0120W01athlete) using the sled. The athlete performs the exercise action by first performing a clean from an initial posture and stance, e.g., lifting one end of the sled using an overhand grip to lift the sled from the ground at an angle, performing a press with the sled, e.g., pushing the sled away in a horizontal direction away from the athlete, and then returning to the initial posture and stance. In some cases, rather than performing a clean to lift the sled to a shoulder height, the athlete can perform a standing press to lift the sled and can perform repetitions of a press, e.g., lifting the weight overhead from shoulder level, to perform an exercise that focuses on training upper body muscle groups.

[0196] In some implementations, the training device 104 can be configured for tractor pull exercises. For example, FIG. IB depicts an example exercise action 106-4 referred to as a tractor pull using the training device 104. The training device 104 includes a pulling mechanism and weights, such that as the pulling mechanism is pulled by the athlete 102, the weights 105 move closer to the pulling mechanism and increases the friction between the training device 104 and a surface 103, e.g., the ground. As the training device 104 is pulling along a surface, the weights 105 are pushed ahead of structural members, e.g., a frame structural member and / or any horizontally oriented frame structure members, thereby pushing the front portion of the training device 104 into the surface. In this way, the training device 104 can mimic additional weight for the athlete performing the tractor pull, e.g., increasing the difficulty of the exercise and improving the athlete’s training. The athlete can continue pulling the training device 104 until the athlete is unable to perform the exercise action, e.g., unable to generate the amount of force to overcome the friction between the training device and the ground, or able to continually pull the training device 104, e.g., generating the amount of force to exceed the amount of friction between the training device and the ground.

[0197] In some implementations, such as the training device 104 having an adjustable weight system to increase the difficulty for performing tire flips. For example, an athlete can pick up one end of the training device and drive the training device. A weight of the adjustable weight system can moved, e.g., by one or more mechanisms such as a pulley, towards the end of the training device that is lifted by the athlete. In some implementations, the weight is part of a weight-pin system for the training device. In some implementations, the weight is part of an automatic adjustable weight system, such as a weight system controlled by the computing device 124.Attorney Docket No.: 36610-0120W01

[0198] By moving the weight towards the lifted end of the training device, lifting and driving the device can be made more difficult, e.g., the added weight making the training device harder to hold and drive. This also simulates game-like scenarios in American football, such as varying resistance at different phases of an exercise action, e.g., hitting to make contact, lifting, driving, and finishing. In this way, the resistant load of the training device 104 can be adjusted throughout a sequence of phases for an exercise action, e.g., lighter resistance at the earlier phases of the exercise action and increased resistance at later phases of the exercise action.

[0199] An example of an exercise action 106 can also include a sled push, where the athlete adopts a low, forward leaning posture and stance (e.g., bending the knees slightly and angling the athlete’s chest towards the sled) and uses an overhand grip to hold a top portion of a sled, e.g., a handle. The athlete can then perform an initial drive, by pushing their feet into the ground and then extending their legs to generate an amount of force that moves the sled. The athlete can then continue applying force to keep the sled in momentum until the athlete releases the sled and reverts to standing posture and stance. The athlete can then repeat the exercise, e.g., according to a workout training program.

[0200] The athlete can also perform exercise actions that are tailored towards particular sports or applications. In the context of American football, the athlete can perform a lineman throw using the strength training device by approaching the strength training device from a kneeling position. The athlete can then adopt an overhand grip at a first end of the strength training device, lift the strength training device to shoulder level while maintaining the kneeling position, switch to an underhand grip, and push the strength training device such that the first end rotates around a point located at a second end of the strength training device, the second end being opposite from the first end. In this way, the athlete can mimic explosive pushing and throwing actions in a game of American football by performing exercise actions, with the sensors 120 of the training device 104 continually determining metrics for the athlete’s performance of the exercise action and providing one or both of metrics and evaluations for the exercise action. In this way, the training device 104 can responsively provide accurate, sensor-based evaluations of the athlete’s performance of the exercise action, among other benefits of athlete performance tracking.

[0201] The athlete can use a strength training device (e.g., such as a sled with or without additional weights) to perform exercise across different applications such as general or recreational fitness, e g., to improve one’s day-to-day strength and endurance or for recreationalAttorney Docket No.: 36610-0120W01acti vities / sports, as well as applications in first responder physical training, e.g., for lifting debris, carrying people and / or equipment for rescue, firefighting, etc. The athlete can perform exercises such as a tire flip, barrel rolls, deadlifts, among others. For example, the athlete can perform a tire lift by first adopting a stance where the athlete’s feet are shoulder- width apart, performing a squat by bending their knees, and gripping the strength training device to perform a lift, e.g., by drive their feet into the ground to push one end of the strength training device upward. After lifting the strength training device, e.g., to chest or shoulder height and standing from the squat position, e.g., by extending knees with a straight back, the athlete can push the strength training device forward to flip the strength training device.

[0202] The athlete can perform a barrel roll exercise using the strength training device. For example, the athlete can grip a left portion or a right portion of a first end of the strength training device, e.g., using an underhand grip, with one or both hands. After gripping the strength training device, the athlete can pull the strength training device along a direction perpendicular to a center axis of the strength training device. In this way, the athlete flips the strength training device in along the perpendicular direction. The athlete can repeat the exercise using one or both hands to flip the strength training device in the same perpendicular direction again or can flip the strength training device in a direction opposite to the perpendicular direction, e.g., to return the strength training device to its original position. This is a particularly useful exercise in circumstances when space is limited and angles of approach to the object being lifted are limited, such as debris that can only be approached from a narrow range of angles, e.g., by a first responder.

[0203] As another example, the athlete can perform a deadlift using the strength training device (e.g., a sled), which includes the athlete using an overhand grip on a handle on one end of the strength training device while the athlete is in a squat position, then lifting the strength training device upward from the ground using a pulling motion at the end being held. The athlete can shift from a squat position to a standing position when lifting the strength training device and then lower the strength training device back to the ground while returning from the standing position to the squat position. Although this specification describes different exercises using the strength training device, the training device 104 can be applied to combinations of exercises. For example, the training device 104 can identify and improve the performance of an athlete performing any workout using any combination of exercises, e.g., a workout that mixes, repeats, re-orders different exercises using the strength training device.Attorney Docket No.: 36610-0120W01

[0204] FTG. 1C shows a perspective view 140 of example frame structural members for a sensorbased strength training device. The view 140 shows a frame structural member 142 having an adjustable length for the frame, such as by having a telescoping structure where an inner portion of the frame structural member is contained in an outer portion, and the inner portion is slidable, e.g., along an axis. For example, the view 140 shows an inner portion 144 contained in an outer portion 146. Both the inner portion 144 and the outer portion 146 can have a shape similar to a hollowed rectangular prism, with the inner portion 144 being a smaller size than the outer portion 146 to allow the inner portion 144 to slide in a direction 145, e.g., along an axis that spans the length of the frame structural member 142. By sliding the inner portion 144 into the outer portion 146, the length of the frame structural member 142 can be shortened, whereas sliding the inner portion 144 out of the outer portion 146 can increase the length of the frame structural member 142.

[0205] Increasing the length of the frame structural member 142 can increase the amount of force without adding additional weight, e.g., by increasing the length where a force for an exercise action is applied. Increasing the length of the frame structural member 142 can allow for additional weight to be added, e.g., by adding weight pins to hold weight plates. In contrast, shortening the length of the frame structural member 142 can decrease the amount of force without removing weight, e.g., by decreasing the length where a force for an exercise action is applied. The length of the frame structural member can also be adjusted for storage, e g., decreasing the length of the frame structural member so that the training device 104 takes up less space compared to a fully or partially extended frame structural member.

[0206] In some implementations, the frame structural member 142 can include a weight adjustment mechanism. The weight adjustment mechanism can adjust the amount of weight for the training device, such as by a slidable mechanism corresponding to different weights based on the position of the slider in the slidable mechanism. In some implementations, the weight adjustment mechanism can be a weight-stack pin adjustment system, e.g., plates of weight with selector pins that can be inserted in holes disposed into a weight plate. The pin can be inserted in an opening of at a particular weight plate to configure the weight of the training device.

[0207] As an example, by moving the weight pin and inserting into an opening of the weightstack pin adjustment system that is closer to a pivot point of the training device 104, the training device 104 can be light, e.g., easier to lift. In contrast, by moving the weight pin and inserting intoAttorney Docket No.: 36610-0120W01an opening of the weight-stack pin adjustment system that is further away from a pivot point of the training device 104, the training device 104 can be heavier, e.g., more difficult to lift.

[0208] In some implementations, the weight adjustment mechanism can be a weight pin system with one or more slidably movable pins that adjust the amount of weight by moving weight along an axis, e.g., closer to an end where an athlete makes contact with, e.g., gripping, pulling, lifting, flipping, the training device 104 to perform an exercise. Moving one or more of the weight pins can move an amount of weight closer or farther away from the athlete, in which the adjustment in weight placement modifies an amount of force demanded to move the training device 104, e.g., for lifting or pushing. For example, the farther away the weight is relative to an end where the training device 104 is being lifted, the more difficult the training device 104 can be to lift, flip, push, etc. A weight adjustment mechanism having one or more slidable weight cans can allow for adjustable amounts of weight for the training device and adjustable weight placement, e.g., placing weight on different locations of the training device 104. In some instances, the amount of weight for the training device 104 can be a fixed amount of weight but the placement of the fixed weight can be configured along different locations of the training device 104, e.g., along a weight-pin adjustment system.

[0209] FIG. 1C also shows a perspective view 150 of an example frame structural member 152 for a sensor-based strength training device. The frame structural member 152 includes vertical structural members 151-1, 151-2, and 151-3, coupled together by horizontal structural members 154-1, 154-2, and 154-3. Although FIG. 1C depicts the frame structural member 152 having three vertical structural members and three horizontal structural members, the frame structural member 152 can have a different number of horizontal and vertical structural members. For example, the frame structural member 152 can include two vertical structural members and two vertical structural members. Each of the vertical structural members and each of the horizontal structural members can be an example of a telescoping frame, such as the frame structural member 142 depicted in view 140. In this way, a width of the training device 104 can be adjusted to widen or narrow the width of the training device 104, e.g., by adjusting the length of the horizontal structural members, e.g., indicated by bi-directional arrows to depict a direction 155. In addition to or instead of adjusting the width, a length of the training device 104 can be adjusted to shorten or elongate the length of the training device 104, e.g., by adjusting the length of the vertical structural members.Attorney Docket No.: 36610-0120W01

[0210] In some implementations, a frame structural member, e.g., frame structural member 108 of FIG. 1A, frame structural members 142 and 152 of FIG. 1C can include one or more compartments such as compartment 148. The compartment can allow for storage of sand, water, or another material to increase the weight of the training device 104. In some instances, the compartment can include a slidable weight mechanism. The slidable weight mechanism can be electronically controlled, e.g., by a computing device 124 to adjust the amount of weight.

[0211] FIG. ID is a top view 160 of the example frame structural members. The top view 160 shows the frame structural member 162 having a vertical structural member 166 and two horizontal structural members 164-1 and 164-2. The horizontal structural member 164-1 is coupled to a first end of the vertical structural members 166 and the horizontal structural member 164-2 is coupled to a second end of the vertical structural members 166, the second end being opposite the first end.

[0212] The top view 170 shows the frame structural member 172 having vertical structural members 178-1, 178-2, and 178-3 (collectively “vertical structural members 178”) and three horizontal structural members 174-1, 174-2, and 174-3 (collectively “horizontal structural members 174”). The horizontal structural member 174-1 is coupled to a first end of each of the vertical structural members 178, and the horizontal structural member 174-3 is coupled to a second end of each of the vertical structural member 178, the second end being opposite the first end. The second horizontal structural member connects the first vertical structural member 178-1 to the second vertical structural member 178-2, and the second vertical structural member 178-2 to the third vertical structural member 178-3. Each of the vertical structural members and each of the horizontal structural members can be an example of a telescoping frame, such as the frame structural member 142 and 152 described in reference to FIG. 1C.

[0213] FIG. IE is a diagram 180 of example attachments for a strength training device. The diagram 180 shows an attachment 182-1 and attachment 182-2, e.g., an example of an attachment 112 described in reference to FIG. 1A above and 5A through 5C below. The attachment 182-1 can be a blocking pad, such as to allow the athlete to simulate tackling exercises in American football. The attachments 182-1 and / or attachment 182-2 can be attached to the training device by a first structural member 185 that can be connected to the training device, e.g. training device 104.. In some implementations, the first structural member 185 can be connected to another structural member, e.g., a second structural member 187. The second structural member 187 allows for the attachment 182-2 to be adjusted at an angle 0 relative to axis 189 and axis 181 that is depictedAttorney Docket No.: 36610-0120W01substantially perpendicular to axis 189. Tn this way, the height of the attachment 182-2 can be adjusted relative to a ground surface, e.g., to account for varying heights for different athletes.

[0214] The angle of orientation 9 depicted in FIG. IE for the attachment 182-2 can be adjusted by the structural member 517. Although the attachment 182-2 can be connected to the first structural member 185 and the first structural member 185 can be connected to a first end of training device 104, the first structural member 185 can be connected to the first end of the training device 104 by an optional structural member 187. As an example, the attachment 182-2 can be attached or connected to the first structural member 185, which is attached or connected to the first end of training device 104. As another example, the attachment 182-2 can be attached or connected to the first structural member 185, which can be attached to or connected to the second structural member 187, which can be attached or connected to the first end of training device 104.

[0215] In some implementations, the attachment 182-2 can be attached or connected to the second structural member 187, which can be attached or connected to the first structural member 185, which can be attached or connected to the first end of training device 104. The attachment 182-2, the first structural member 185, the second structural member 187, or some combination thereof, can be attached to the training device 104 by the second end of the training device. Although FIG. IE depicts the attachment 182-2 attached to the first structural member 185 and / or the second structural member 187, the attachment 182-1 can be attached to the first structural member 185 and / or the second structural member 187. The attachment 182-1 can be attached to a first end and / or a second end of the training device 104. The attachment 182-1 can be used to allow the athlete to practice tackling and / or driving (e.g., pushing) the training device. For example, the training device 104 can be configured as a single sled, e.g., for complex exercise actions such as tackling by one athlete, a double-team sled, e.g., for tackling by two athletes. The attachment 182-1 can be any shape, size, material, etc.

[0216] FIG. IF is a diagram 190 of an example damping system for a strength training device 194. The training device 194 can be an example of training device 104 and one or both of the first end 195a and the second end 195b of the sensor-based strength training device 194 can include bumpers to improve safety by absorbing shock (and other types of forces) when the strength training device 194 lands on the ground. For example, the first end 195a and / or the second end 195b of the strength training device can include bumpers such that when performing exercise action, e g., lifted, pushed, and flipped over, one or more bumpers of the strength training deviceAttorney Docket No.: 36610-0120W01194 lands on the ground and the bumpers absorb resulting forces, e.g., shock, impact. The diagram 190 in FIG. IF depicts bumpers 196-1, 196-2, 196-3, and 196-4 (collectively “bumpers 196”). Bumper 196-1 is a bumper on a top portion of the first end 195a of training device 194, bumper 196-2 is a bumper on a bottom portion of the first end 195a of training device 194, bumper 196-3 is a bumper on a top portion of the second end 195b of training device 194, and bumper 196-4 is a bumper on a bottom portion of the second end 195b of training device 194.

[0217] In some implementations, the strength training device can include other types of damping systems (in addition to or instead of the bumpers), to reduce the forces between the strength training device and a surface for performing the exercise action. For example, the diagram 190 shows dampers 192-1, 192-2, 192-3, and 192-4 (collectively “dampers 192”). The damper 192-1 connects a portion of a frame structural member to a top portion of the first end 195a of the training device 194. The damper 192-2 connects a portion of a frame structural member to a bottom portion of the first end 195a of the training device 194. The damper 192-3 connects a portion of a frame structural member to a top portion of the second end 195b of the training device 194. The damper 192-4 connects a portion of a frame structural member to a bottom portion of the second end of the training device 194.

[0218] The damping systems provide a way to perform an exercise action with the strength training device (such as flipping the device) so that it lands more softly, e.g., in contrast to a strength training device without bumpers and / or damping systems. By softening the load and reducing forces between strength training device and a surface of the exercise environment, the bumpers and / or damping systems reduce noise, reduce wear and tear for the surface of the exercise environment, reduce wear and tear for the strength training device. For example, the surface (e.g., a driveway) can be made of concrete, asphalt, or another type of material. The bumpers and / or the damping mechanisms can reduce forces acting on the materials of the surface to prevent wear and tear resulting in defects in the surface, e.g., cracks, potholes, depressions, and other types of defects.

[0219] Different types of damping mechanisms, materials, and absorbers can be included in the exercise device to reduce vibrations and / or oscillations, as well as absorb and dissipate kinetic energy from the exercise action. Examples of damping mechanisms can include spring-systems, hydraulic systems, pneumatic systems, gel-based damping, magnetic damping, and frictionAttorney Docket No.: 36610-0120W01damping, among others. Examples of materials can include rubber, elastomers, foams, hydraulic fluids, springs, viscoelastic materials, and components.

[0220] FIG. 2 is a block diagram 200 of an example computing environment for a sensor-based strength training device. The block diagram 200 shows the sensors 120 of the training device 104, e.g., described in reference to FIGS. 1A and IB above, communicatively coupled to a computing environment 202 by a communication network 218. Each of the sensors from sensors 120 can provide sensor data to the computing environment 202. Streams of sensor data 222-1 through 222-N (collectively “sensor data 222”) can be provided to the computing environment 202, each stream of sensor data 222 corresponding to a respective sensor from sensors 120. For example, a sensor 120-1 can provide a stream of sensor data 222-1 to the computing environment 202 and a sensor 120-N can provide a stream of sensor data 222-N to the computing environment 202. The computing environment 202 receives sensor data from sensors 120 and generates output data 216 based on the sensor measurements for the sensor data 222. The output data 216 can include one or more performance metrics determined by the computing device 124 for an exercise performed by the athlete, based on sensor measurements captured by sensors 120.

[0221] The computing environment 202 provides the output data 216 to one or more user computing devices 220 communicatively coupled to the computing environment 202. User computing devices 220-1 through 220-N (collectively “user computing devices 220”) can be any computer system, device, mobile device, smart watch, smartphone, etc. An athlete can review evaluations and performance metrics for exercises through a user computing device 220.

[0222] The communication network 218 can utilize one or more communication device(s) to communicate with external devices, e.g., sensors 120, via one or more wired or wireless communication networks, or both. Communication device(s) can include any one or more communication devices, such as network interface cards (e.g., Ethernet cards), optical transceivers, radio frequency transceivers, Bluetooth transceivers, 3G or 4G transceivers, and WiFi radio computing devices.

[0223] Referring to the computing environment 202, the computing device 124 includes a sensor engine 204 that includes a model 206. The sensor engine 204 can receive sensor measurements from any of the sensors 120 and use the sensor measurements as an input to model 206. The model 206 can be configured to apply or perform a number of machine learning techniques, statistical techniques, physics-based simulations, or some combination thereof, to generate a model outputAttorney Docket No.: 36610-0120W01indicating a performance metric for the exercise performed by athlete 102 using the training device 104. The model output can be shown as output data 216, which can be an example of exercise performance output 122. In some cases, the model output from model 206 can also include data indicating feedback for the athlete to improve the performance of the exercise 106 being performed. In some cases, the model output can include graphical interface data that configures a display of a device (e.g., a user computing device 220-1) to display the performance adjustment, but the model output can also be an alert provided to the athlete 102.

[0224] The computing device 124 can also include hardware such as a processor 208 for implementing operations based on instructions to process the sensor data and data storage 210 for storing sensor measurements and other inputs, intermediate outputs, and output data 216 generated by the computing device. In some implementations, the computing device 124 is optionally connected to a server 212 and / or computing platform 214 for processing. For example, the computing device 124 can receive the sensor data 222 and transmit at least a portion (or the entirety) of the sensor data for a server 212 and / or the computing platform 214 for processing. In some implementations, data processed by the computing device 124 can be transmitted to the server 212 and / or the computing platform 214 for further transmit, e.g., to computing devices, systems, and platforms that are not connected to the computing device 124, as well as other types of user devices 220 that may not be connected to the computing device 124 through the communication network 218.

[0225] The sensor engine 204 can be configured to train the model 206 using sensor data from sensor 120-1. The sensor engine 204 can also be configured to train the model 206 using sensor data from any of sensors 120-2 through 120-N, e.g., instead of or in addition to sensor data from sensor 120-1. The model 206 can be trained by the sensor engine 204 using a variety of training techniques to improve the accuracy of inference tasks performed by the model. These training techniques can include supervised and unsupervised learning. The model 206 can include any form of boosting techniques such as gradient boosting but can include deep learning techniques to perform an inference task.

[0226] For example, an inference task for the model 206 can include generating scores, labels, or some combination thereof indicating an evaluation of an athlete’s performance of an exercise. A label can indicate a degree of performance for the exercise, e.g., excellent, great, fair, poor, incomplete. A numerical value can indicate a score, e.g., a higher value indicating betterAttorney Docket No.: 36610-0120W01performance than a lower value. A numerical value can also indicate a metric representing a performance of the exercise. An inference task can also include generating, based on the sensor measurements of the athlete’s exercise, one or more adjustments to improve the athlete’s performance of the exercise. In some examples, a model performs hybrid-learning techniques to improve accuracy of model output. Training processes for the model 206 can include any number of iterative processes, each performing a number of iterations to train the model to achieve a target performance value, e.g., an error rate below a threshold value, a generated classification label that matches the ground truth label.

[0227] The training of the model 206 can be performed using obtained ground truth data that includes known labels, associations, classifications, etc., coupled with a corresponding input, e.g., some or all of the sensor data. The sensor engine 204 can adjust one or more weights or parameters of the model 206 to match estimates or predictions from the to the ground truth data, e.g., sensor data captured of an athlete performing an exercise with a target performance metric. In some implementations, the model 206 includes one or more fully or partially connected layers. Each of the layers can include one or more parameter values indicating an output of the layers. The layers of the model 206 can generate outputs for which the model can use for performing one or more inference tasks. The model 206 can be validated and tuned through holdout and test techniques, model comparison, and model selection.

[0228] The model 206 can process input data such as sensor data 222 to generate output data 216 representing an athlete’s power output, fatigue levels, and other indicators of exercise performance. In some implementations, the output data 216 can be a time-series evaluation of the athlete’s performance to describe the athlete’s performance over a period of time, e.g., power output over a time period. The model 206 can generate output data 216 indicating a recommendation for training intensity, training plan, recovery period, etc. that can optimize an athlete’s training session. In this way, the model 206 provides feedback to the athlete 102 to allow the athlete to optimize their workout, e.g., training in an optimal way that achieves peak performance. For example, the output data 216 can indicate an amount of weight, a type of exercise, cardiovascular efficiency, and other types of metrics to provide real-time or near realtime feedback to the athlete. The feedback from the model 206 can allow the athlete 102 to stay within target zones for aerobic or anaerobic training and achieve a target performance for sport, competitions, personal fitness, etc.Attorney Docket No.: 36610-0120W01

[0229] Examples of output data 216 can include performance trends, e.g., trends of performance metrics for phases of exercises and / or multiple exercises, personalized workout plans, predictions of peak performance capacity, among other types of exercise metrics. The model 206 can also identify patterns and generate predictions, such as when an athlete is likely to hit a plateau or risk overtraining and offer feedback indicating adjustments to avoid setbacks in the athlete’s training plan. As described in this specification, the disclosed technology applies sensor-based modeling of the strength training device to help athletes improve training efficiency, make adjustments in training plans to meet a particular performance metric, and provide monitoring of training progress. In some cases, the performance metrics in the output data 216 can include athlete physiological metrics, e.g., respiratory rate, oxygen saturation, heart rate variability, lactate threshold, and environmental metrics, e.g., temperature, humidity, contact pressure, stability. As another example, the performance metrics in the output data 216 can also kinetic and load metrics, e.g., force distribution, load tolerance, as well as temporal metrics, e.g., cadence in exercises, and time under tensions, e.g., duration of muscle engagement during a lift or hold for an exercise.

[0230] In some implementations, components of the strength training device can be compartments for storing weight. For example, the frame structural member can include one or more compartments, e.g., openings with attached covers to cover the opening, for adding weight to the strength training device 104, e.g., in addition to or instead of adding weights to a weight pin 113 of the training device 104. While the training device can include one or more weight pins 113 to hold weight plates, the training device can be configured to store weight in compartments of the frame structural member. In some implementations, the compartments can include sand, water, or another material to add weight and adjust the type of resistance (e.g., dynamic, static). For example, storing sand or water in the compartments can provide dynamic resistance when performing exercise actions with the training device, e.g., in contrast to the static resistance provided by weight plates.

[0231] In some implementations, the strength training device can be a stationary exercise device. For example, the strength training device can include one or more flywheel devices to provide fanned resistance for performing exercise actions such as tire flips, sled pushes, sled pulls. The strength training device can be a fanned sled strength training device. The resistance for the fanned sled strength training device can automatically adjust the resistance, e.g., providing variable resistance for the exercise action. The flywheel devices can generate resistance by pushing air asAttorney Docket No.: 36610-0120W01the exercise action, e.g., flipping a portion of the strength training device, and increasing the resistance as the amount of force for flipping the portion of the strength training device increases, e.g., thereby adjusting the resistance. As another example, an athlete can perform a sled push using the fanned sled strength training device and the fanned resistance can be configured to increase, e.g., by the flywheel devices, as the sled is pushed faster by the athlete. Other forms of resistance training can be provided by the strength training device based on accessories and other components of the strength training device.

[0232] For example, the strength training device 104 can include a strike pad attachment 112 or blocking pad 112 to configure the strength training device 104 as a double-team throwable sled, e.g., allowing two players to perform an exercise action in American football such as a two-player throw. In some implementations, the training device 104 (e.g., by the frame structural member 108) can include a motor and the bottom portions 110a and 110b can include wheels or skis, thereby providing a motorized blocking sled configuration for the training device 104 that moves towards the athlete. In this way, the athlete 102 can practice blocking exercises to practice American football using the training device 104. In some implementations, the training device 104 need not include or use the motor, but instead have another athlete 102 pick-up and drive the training device 104 towards the player practicing the blocking exercise action.

[0233] In some implementations and as depicted in FIG. 1A, the environment 100 can further include a force plate 126. The force plate 126 can include sensors to measure forces generated by the athlete 102 and / or the strength training device 104 on top of the force plate 126 or moving across the force plate 126. In this way, the force plate 126 can generate sensor measurements and the computing device 124 can generate an output indicating or describing the athlete’s balance, gait, efficiency, stand, and other biomechanical parameters for exercise actions.

[0234] FIG. 3A is a flow diagram of an example process 300 for processing sensor measurements from the sensor-based strength training device (also referred to as an exercise device). For convenience, the process 300 will be described as being performed by a system of one or more computers, located in one or more locations, and programmed appropriately in accordance with this specification. For example, a training device, e.g., the sensor-based strength training device 104 of FIG. 1A can perform the process 300. As another example, the process 300 can be performed by, for example, the training device 194 of FIG. IF, the training device 404 of FIG. 4A,Attorney Docket No.: 36610-0120W01the training device 454 of FIG. 4B, the training device 504 of FIG. 5A, and the training device 604 of FIG. 6.

[0235] Briefly, the process 300 includes obtaining, from a first sensor of a plurality of sensors of an exercise device, first sensor data including first sensor measurements captured by the first sensor for an exercise action with the exercise device (302), determining, from the first sensor data, a first metric representing a characteristic for the exercise action (304), and generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric (306). Measurements for the sensor-based exercise device can be stored in data storage of the computing device 124. In some implementations, the measurements can be stored, transmitted to, etc., to another device coupled to the sensor-based exercise device, e.g., by wireless communication, wired communication.

[0236] In some implementations, the process 300 includes obtaining, by a second sensor from the plurality of sensors, second sensor data including second sensor measurements captured by the second sensor for the exercise action with the exercise device. The second sensor can be a different sensor than the first sensor. The process 300 can include determining, from the second sensor data, a second metric representing a characteristic for the exercise action.

[0237] In some implementations, the process 300 includes obtaining, by a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor for the exercise action with the exercise device. The process 300 includes determining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.

[0238] In some implementations, the process 300 includes generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

[0239] In some implementations, the first sensor is a first type of sensor, and the second sensor is a second type of sensor, the second type being different from the first type.

[0240] In some implementations, the process 300 includes determining, by a computing device, a first value representing friction between a surface and the exercise device, wherein the determination is based on the first sensor measurements and / or the second sensor measurements. The process 300 includes, based on a comparison of the first value representing friction and a second value representing a target metric, generating an indicator to adjust the exercise device.Attorney Docket No.: 36610-0120W01

[0241] In some implementations, the process 300 includes generating calibration data for the exercise device based on the first sensor measurements.

[0242] In some implementations, the process 300 includes generating feedback data based on the first metric.

[0243] FIG. 3B is a flow diagram of an example process 320 for processing sensor measurements from a strike pad for the sensor-based strength training device. Briefly, the process 320 includes determining, by a strike pad, that the strike pad has been removably connected to a strength training exercise device (322), transmitting features of the strike pad to the strength training exercise device (324), and receiving data from the strength training exercise device based on the features of the strike pad (326). For convenience, the process 320 will be described as being performed by a system of one or more computers, located in one or more locations, and programmed appropriately in accordance with this specification. For example, a training device, e.g., the sensor-based strength training device 104 of FIG. 1A can perform the process 320. A strike pad attachment, e.g., strike pad attachment 112 of FIG. 1A, attachments 182-1 and 182-2 of FIG. IE, attachments 512, 562, 564 of FIG. 5 A through 5C, can perform the process 320. As another example, the process 320 can be performed by, for example, the training device 194 of FIG. IF, the training device 404 of FIG. 4A, the training device 454 of FIG. 4B, the training device 504 of FIG. 5A, and the training device 604 of FIG. 6.

[0244] FIG. 3C is a flow diagram of an example process 340 for processing sensor measurements from a strike pad adapted to be removably connected to a strength training device. Briefly, the process 340 includes obtaining, by a first sensor of a strike pad, first sensor data including first sensor measurements captured by the first sensor for an exercise action with the strike pad (342). The strike pad is adapted to be removably connected to a strength training exercise device. The process 340 includes determining, from the first sensor data, a first metric representing a characteristic for the exercise action (344).

[0245] For convenience, the process 340 will be described as being performed by a system of one or more computers, located in one or more locations, and programmed appropriately in accordance with this specification. For example, a training device, e.g., the sensor-based strength training device 104 of FIG. 1A can perform the process 340. A strike pad attachment, e.g., strike pad attachment 112 of FIG. 1A, attachments 182-1 and 182-2 of FIG. IE, attachments 512, 562, 564 of FIG. 5A through 5C, can perform the process 340. As another example, the process 340 can beAttorney Docket No.: 36610-0120W01performed by, for example, the training device 194 of FIG. IF, the training device 404 of FIG. 4A, the training device 454 of FIG. 4B, the training device 504 of FIG. 5A, and the training device 604 of FIG. 6.

[0246] FIG. 3D is a flow diagram of an example process 360 for processing sensor measurements from a sensor of a strike pad for a strength training device. Briefly, the process 360 includes obtaining, by at least one sensor of a strike pad adapted for attachment to an exercise device, first sensor data including first sensor measurements captured by the at least one sensor for an exercise action with the strike pad (362). The process 360 includes determining, from the first sensor data, a first metric representing a characteristic for the exercise action (364).

[0247] For convenience, the process 360 will be described as being performed by a system of one or more computers, located in one or more locations, and programmed appropriately in accordance with this specification. For example, a training device, e.g., the sensor-based strength training device 104 of FIG. 1A can perform the process 360. A strike pad attachment, e.g., strike pad attachment 112 of FIG. 1A, attachments 182-1 and 182-2 ofFIG. IE, attachments 512, 562, 564 of FIG. 5A through 5C, can perform the process 360. As another example, the process 360 can be performed by, for example, the training device 194 ofFIG. IF, the training device 404 ofFIG. 4A, the training device 454 ofFIG. 4B, the training device 504 ofFIG. 5A, and the training device 604 ofFIG. 6.

[0248] FIG. 4A is a diagram 400 of an example strength training device 404 (“training device 404”). The training device 404 can be an example of a training device 104 described in reference to FIG. 1A above. The training device 404 includes a frame structural member 408, e.g., similar to frame structural member 108 of training device 104 in FIG. 1 A, and includes one or more weight loading pins 413 (also referred to as “weight pins”) that are part of (e.g., integrally formed with) or attached to the frame structural member 408. The one or more weight pins allow for additional weight (e.g., weight plates) to be added to the training device 404. The frame structural member 408 connects a first end 412a of the training device to a second end 412b.

[0249] The training device 404 includes handles 407-1 through 407-4 (collectively “handles 407”) at the first end 412a. The training device 404 also includes a bottom portion 410 that includes a pair of skis to make contact with a ground surface, e.g., ground surface 103 described in reference to FIG. 1A above. An athlete can lift the training device 404 by holding any of the handles 407,Attorney Docket No.: 36610-0120W01including pairs of handles. For example, an athlete can lift the training device 104 by lifting a pair of handles, e.g., handles 407-2 and 407-4, handles 407-1 and 407-3.

[0250] The athlete can lift the training device 404 by any of the handles 407 in a vertical direction 409 and flip the training device 404, e.g., similar to exercise action 106-1, in a direction 405. For example, the athlete lifts a first end 412a of the training device 404, e.g., the first end 412a of the training device 404 with the handles 407. The athlete can push the first end 412a of the training device while the first end 412a is lifted such that the first end 412a is pushed over, e.g., about a pivot point, a second end 412b of the training device 404. The second end 412b of the training device 404 is opposite the first end 412a.

[0251] The bottom portion 410 of the training device 404 includes a pair of skis that can provide two points of contact between the training device 404 and a ground surface, e.g., ground surface 103. The training device 404 can provide stability that allow the athlete to push the training device 404 in a forward direction 411 and / or provide an upward force in a vertical direction 409 to lift the training device 404, e.g., to flip the training device.

[0252] FIG. 4B is a diagram 450 of another example strength training device 454 (“training device 454”). The training device 454 can be referred to as a “single touchpoint blocking sled” and can be thrown, flipped, rotated, etc. in a horizontal direction 459. The training device 454 can be an example of a training device 104 described in reference to FIG. 1A above.

[0253] The training device 454 includes a frame structural member 458, e.g., similar to frame structural member 108 of training device 104 in FIG. 1A, and includes one or more weight loading pins 413 (also referred to as “weight pins”) that are part of (e.g., integrally formed with) or attached to the frame structural member 458. The one or more weight pins allow for additional weight (e.g., weight plates) to be added to the training device 454. The frame structural member 458 connects a first end 452a of the training device to a second end 452b.

[0254] The training device 454 includes handles 457-1 through 457-4 (collectively “handles 457”). The training device 454 includes a bottom portion 460 that includes a disk to make contact with a ground surface, e.g., ground surface 103 described in reference to FIG. 1A above. The bottom portion 460 can be a bowl, a saucer, a ski, etc. The bottom portion 460 of the training device 454 can have a substantially smaller surface area than the bottom portion 410 of the training device 404.Attorney Docket No.: 36610-0120W01

[0255] The training device 454 can introduce an instability factor that demands additional control by the athlete, thereby allowing the athlete to train stability and control when performing an exercise action. By introducing an instability factor, the training device 454 can allow the athlete to train and improve coordination, muscle activation, functional strength, and other athletic performance metrics. As another example, the instability factor from the training device 454 can allow the athlete to reduce joint stress and / or muscle imbalance, as well as increase flexibility, by performing exercise actions with the training device 454.

[0256] Similar to the training device 404 in FIG. 4A, an athlete can lift the training device 454 by holding any of the handles 457, including pairs of handles. The bottom portion 460 of the training device 454 includes provides a single point of contact between a first end 452a of the training device 454 and a ground surface, e.g., ground surface 103. As depicted in FIG. 4B, the second end 452b of the training device 454 can have multiple points of contact. In some implementations, the second end 452b of the training device 454 can be similar to the first end 452a, e.g., including a bottom portion 460 with a substantially round shape.

[0257] The training device 454 can allow for an athlete to simulate complex exercise actions such as driving back a defender in American football and applying a rotational force, e.g., indicated by direction 455, to simulate a lateral displacement of the defender, e.g., in a direction 459. The training device 454 allows the athlete to perform exercise actions that simulate sport strategy to provide a competitive advantage, such as by throwing a defender off balance to fall on the ground or throwing a defender off balance and then throwing the off-balance defender onto the ground. By utilizing a training device with a bottom portion 460 having a smaller contact area with the ground, e.g., compared to bottom portion 410 of FIG. 4A, the training device 454 allows for a wide range of blocking techniques and complex exercise actions to be performed. For example, the training device 454 can allow for a wide variety of blocking and finishing moves that demand movement with torque, e.g., indicated by direction 455.

[0258] In some implementations and as described in reference to FIGS. 5A and 5B below, the training devices (e g., training device 104, 404, 454) can include handholds, blocking pads, and / or other types of attachments.

[0259] FIG. 5 A is a diagram of another example strength training device. Similar to FIG. 1A, FIG. 5A shows an environment 500 including an example sensor-based strength training device 504 with sensors 520-1 - 520-N (collectively “sensors 520”). The strength training device 504,Attorney Docket No.: 36610-0120W01which can also be referred to as smart training device 504 or training device 504, can be an American football training sled that optionally includes additional components for athletic training, such as weighted plates, bumpers, handles, frame accessories, strike pads, etc. Similar to the training device 104, the strength training device 504 includes a frame structural member 508 having a first end 507a and a second end 507b opposite of the first end 507a, and both ends define a longitudinal axis 509 along a length of the frame structural member 508. Although FIG. 5A depicts the training device 504 having a top portion 511a protruding upward from the first end 507a and a bottom portion 510a protruding downward from the first end 507a, each of the top portion 511a and the bottom portion 510a can be optionally included. Similarly, each of the top portion 511b protruding upward from the second end 507b and the bottom portion 510b protruding downward from the second end 507b can be optionally included.

[0260] The frame structural member 508 includes one or more weight loading pins 513 (also referred to as “weight pins”) that are part of (e.g., integrally formed with) or attached to the frame structural member 508. The frame structural member 508 connects the first end 507a of the training device 504 to a second end 507b. The training device 504 includes a computing device 530 configured to process the sensor measurements and determine performance metrics of an exercise, e.g., similar to computing device 124 to generate an exercise performance output.

[0261] The first end 507a of the training device 504 includes a top portion 511a and a bottom portion 510a while the second end 507b includes a top portion 511b and a bottom portion 510b. The bottom portions 510a and 510b are depicted in FIG. 5 A as skis to reduce friction, improve traction and stability, as well as improve maneuverability of the training device 504, e.g., while an athlete pushes or pulls the training device 504 when performing an exercise. Similar to the bottom portions 110a and 110b described in reference to FIG. 1A, the bottom portions 510a and 510b can be a disk, braking mechanism, caster ski, etc. Similar to the training device 104 of FIG. 1A, the training device 504 of FIG. 5A also includes sensors 520-1 through 520-N (collectively “sensors 520”).

[0262] Similar to the training device 104 depicted in FIG. 1A, the training device 504 depicted in FIG. 5A includes an attachment 512 and a sensor 520-2 mounted onto or embedded in the attachment 512. The attachment 512 for the training device 504 is referred to as a “medicine ball attachment” that allows the player to simulate expected blocking impact that occurs during American football training and games. The medicine ball attachment 512 can be attached to theAttorney Docket No.: 36610-0120W01training device by a rotating mechanism 515 that allows the medicine ball attachment 512 to spin or rotate about an axis, as described in reference to FIG. 5B below. In some implementations, the medicine ball attachment 512 rotates about the longitudinal axis 509 of the training device 504 using the rotating mechanism 515.

[0263] The medicine ball attachment 512 simulates the blocking impact by combining the ergonomic advantage of the round shape associated a medicine ball with dynamic rotation capabilities provided by the rotating mechanism 515. Although FIG. 5B depicts the rotating mechanism 515 extending outward from a surface of the training device 504, e.g., depicted in the left-hand side of the page, the rotating mechanism can be at least partially disposed in an opening of the training device 504. For example, the opening can be disposed in a surface of the first end 507a. In some implementations, the rotating mechanism 515 is embedded into the first end 507a but with an amount of space that permits the rotating mechanism 515 to rotate about an axis.

[0264] The medicine ball attachment 512 may also be referred to a multi-directional rotational attachment for the training device 504. In some implementations, the medicine ball attachment 512 can be attached to the training device 504 by a structural member 517 that connects the medicine ball attachment 512 to the training device 504. As described in reference to FIG. 5B, the structural member 517 can be adjusted to configure the medicine ball attachment 512 at an angle 0 relative to axis 509 and axis 551 that is depicted substantially perpendicular to axis 509. In this way, the height of the medicine ball attachment 512 can be adjusted relative to a ground surface, e g., to account for varying heights for different athletes.

[0265] The angle of orientation 0 depicted in FIG. 5B for the medicine ball attachment 512 can be adjusted by the structural member 517. Although FIGS. 5 A and 5B depict the medicine ball attachment 512 connected to the rotating mechanism 515 and the rotating mechanism 515 connected to the first end 507a of training device 504, the rotating mechanism 515 can be connected to the first end 507a of the training device 504 by an optional structural member 517. As an example, the medicine ball attachment 512 can be attached or connected to the rotating mechanism 515, which is attached or connected to the first end 507a of training device 504. As another example, the medicine ball attachment 512 can be attached or connected to the rotating mechanism 515, which can be attached to or connected to the structural member 517, which can be attached or connected to the first end 507a of training device 504. In some implementations, the medicine ball attachment 512 can be attached or connected to the structural member 517, whichAttorney Docket No.: 36610-0120W01can be attached or connected to the rotating mechanism 515, which can be attached or connected to the first end 507a of training device 504.

[0266] FIG. 5B is a diagram of an example attachment for the strength training device of FIG.5A. The diagram 550 depicts the medicine ball attachment 512 having a rounded surface 554 and a flat surface 552. The rounded surface 554 of the medicine ball attachment 512 allows for the athlete 102 to train and improve hand placement, e.g., by practicing hand-placement when performing the exercise action. In particular, this allows the athlete to practice low-blocking techniques and applying leverage, e.g., techniques that are in demand for certain types of athletes such as linemen in American football.

[0267] The medicine ball attachment 512 can have any size, weight, and other physical characteristics associated with medicine balls, wall balls, or another type of rounded, cushioned surface for exercise. The medicine ball attachment 512 includes the rounded surface 554 to different striking angles for blocking exercise actions, which allows for the athlete to improve their ability to practice different blocking exercising actions for sport training. The medicine ball attachment 512 can be in a fixed and rotatable position relative to the training device 504 when the training device is stationary. The medicine ball attachment 512 can be rotatable about an axis, such as the rotational motion 556 represented by the arrows in diagram 550, e.g., motion 556 about an axis 509. The medicine ball attachment 512 can be any type or shape for a medicine ball, e.g., dual-grip medicine ball, slam ball, wall ball, elongated medicine ball. In some implementations, the medicine ball attachment 512 can include handles, e.g., on the surface of the medicine ball and / or integrally formed in the medicine ball.

[0268] FIG. 5C is a diagram 560 of another example attachment for the strength training device of FIG. 5 A. The diagram 560 depicts an example attachment 562 that has a substantially ellipsoid shape, e.g., prolate, oblate, triaxial, and an example attachment 564 that has a substantially spherical shape. Although diagram 560 shows attachments 562 and 564 with substantially round shapes, an attachment for the strength training device can be any shape, e.g., elongated, substantially rectangular, etc. Similar to the medicine ball attachment 512, the attachment 562 and / or attachment 564 can have any size, weight, shape, etc., and can be attached to a rotating mechanism 515 of the strength training device 504. In some instances, the attachments 562 and / or 564 can be attached to structural member 517, e.g., in addition to or instead of the rotating mechanism 515.Attorney Docket No.: 36610-0120W01

[0269] FTG. 6 is a diagram 600 of another example strength training device 604. The strength training device 604 is an example of a propulsion-based strength training device, such as an American football sled that can be propelled toward or away from an athlete, e.g., such as direction 652 indicated by bi-directional arrows. The direction for propelling the training device 604 can be substantially parallel to the ground 603. Similar to FIG. 1A, FIG. 6 shows an example sensorbased strength training device 604 with sensors 620-1 - 620-N (collectively “sensors 620”).

[0270] The strength training device 604, which can also be referred to as smart training device 604 or training device 604 , can be an American football training sled that optionally includes additional components for athletic training, such as weighted plates, bumpers, handles, frame accessories, strike pads, etc. Similar to the training device 104, the training device 604 includes a frame structural member 608 having a first end 607a and a second end 607b opposite of the first end 607a, and both ends define a longitudinal axis 609 along a length of the frame structural member 608. The frame structural member 608 includes one or more weight loading pins 613 (also referred to as “weight pins”) that are part of (e.g., integrally formed with) or attached to the frame structural member 608. The frame structural member 608 connects the first end 607a of the training device 604 to a second end 607b. The training device 604 includes a computing device 630 configured to process the sensor measurements and determine performance metrics of an exercise, e.g., similar to computing device 124 to generate an exercise performance output.

[0271] Although FIG. 6 depicts the training device 604 having a top portion 611a protruding upward from the first end 607a and a bottom portion 610a protruding downward from the first end 607a, each of the top portion 611a and the bottom portion 610a can be optionally included. Similarly, each of the top portion 611b protruding upward from the second end 607b and the bottom portion 610b protruding downward from the second end 607b can be optionally included.

[0272] The first end 607a of the training device 604 includes a top portion 611a and a bottom portion 610a while the second end 607b includes a top portion 611b and a bottom portion 610b. The bottom portions 610a and 610b are depicted in FIG. 6 as skis to reduce friction, improve traction and stability, as well as improve maneuverability of the training device 604, e.g., while an athlete pushes or pulls the training device 604 when performing an exercise.

[0273] Similar to the bottom portions 110a and 110b described in reference to FIG. 1A, the bottom portions 610a and 610b can be a disk, braking mechanism, caster ski, etc. Similar to the training device 104 of FIG. 1A, the training device 604 of FIG. 6 also includes sensors 620-1Attorney Docket No.: 36610-0120W01through 620-N (collectively “sensors 620”). Similar to the training device 104 depicted in FIG.1A, the training device 604 depicted in FIG. 6 includes an attachment 612 and a sensor 620-2 mounted onto or embedded in the attachment 612.

[0274] The training device 604 includes a propulsion system 640 that further includes a power source 642 and propulsion elements 644. The propulsion system 640 can include computing devices, hardware processors, software, or some combination thereof. In some implementations, the propulsion system 640 can be controlled by the computing device 630.

[0275] The propulsion system 640 is configured to generate force and / or power for moving the training device 604, such as to drive the training device 604 toward an athlete or away from an athlete. The propulsion system 640 connects to a drivetrain system 646 that connects to the first end 607a and / or the second end 607b to transfer force and / or power to the driving elements 648. The driving elements 648 can be part of, e.g., included, in the bottom portion 610a for the first end 611a and / or the bottom portion 610b for the second end 607b, of the training device 604. The driving elements 648 are coupled to the drivetrain system 646 and allow the transferred force and / or power from the drivetrain system 646 to interact with a ground surface of the training device, e g., turf, asphalt, wood, etc., thereby causing the training device to move forward or backward, e.g., direction 652 The driving elements 648 translate the rotational input from the drivetrain system 646 into linear displacement and can be configured for different types of terrain, e.g., different types of surfaces that form a ground 103. The training device 604 includes braking system 650 configured to decelerate or halt motion of the training device 104, e g., by applying resistive force to the driving elements 648 or associated drivetrain components in the drivetrain system 646.

[0276] The drivetrain system 646 can include components such as gearboxes, transmissions, drive shafts, chain drives, belt drives, differentials, axles, couplings, bearings, etc. Examples of driving elements 648 can include wheels, tracks, treads, skis, rollers, and other types of components for translating the rotational input from the drivetrain system 646 into linear displacement along the surface 103, e.g., the ground.

[0277] The braking system 650 can include components such as disc brakes, drum brakes, friction pads, mechanical latches, locks, hydraulic brakes, electronic braking components and systems, reverse torque motors. The braking system can include any combination of mechanical, hydraulic, pneumatic, or regenerative braking mechanisms, and can be actuated manually, electronically, orAttorney Docket No.: 36610-0120W01both. The braking system 650 can provide additional safety features by decelerating the training device 604.

[0278] Examples of the propulsion system 640 can include an electric motor, an internal combustion engine, a hydraulic motor, a pneumatic actuator, a spring release mechanism, a flywheel drive system, or some combination thereof. Examples of the power source 642 can include batteries, fuel cells, fuel systems, air pressure tanks, pneumatic pistons, hydraulic oil, hydraulic accumulators, solar panels, etc. Although FIG. 6 depicts the propulsion system 640 adjacent and below the frame structural member 608, the propulsion system 640 can be part of the frame structural member 608, above the frame structural member 608, adjacent to and on a lefthand or right-hand side of the frame structural member 608, etc.

[0279] For tension-based implementations of the propulsion system 640, the propulsion system 640 can include springs as the propulsion elements 644 and can further include a compression mechanism and a locking mechanism. The springs can be mechanically deformed, e.g., by compressing the compression mechanism and a locking mechanism can used to hold the spring in place. Examples of springs can include coil springs, torsion springs, leaf springs, gas springs, etc.

[0280] For flywheel-based implementations of the propulsion system 640, the flywheel-drive can receive input from the power source 642 and any generated power from the power source can be used to spin the flywheel in the propulsion elements 644. The flywheel can store energy as rotational kinetic energy and provide the rotational kinetic energy to a drivetrain system 646. Alternatively or additionally, energy for the flywheel can be supplied by forces provided by the athlete, e.g., an athlete flipping and / or pushing the training device 604. The flywheel-based implementation of the propulsion system 640 can include the flywheel drive providing the stored energy, e.g., to transfer rotational energy from the flywheel to the driving elements 648 (e.g., wheels, skis) through the drivetrain system 646.

[0281] The sensor-based training device 604 includes sensors 620 to measure and / or determine speed, acceleration, etc. to adjust propulsion. For example, the computing device 630 can be configured to provide control for the propulsion system 640, power source 642, propulsion elements 644, drivetrain system 646, driving elements 648, and braking system 650. As an example, sensors such as accelerometers and gyroscopes can detect sudden changes in motion or instability and triggering automatic braking to prevent collisions. In some implementations, the sensors 620 can include proximity sensors, load sensors, etc. to detect obstacles or unsafe weightAttorney Docket No.: 36610-0120W01distribution and the computing system 630 can effectuate controls to prevent movement and therefore prevent injury. The training device 604 can be configured to adjust its motion, e.g., by providing a constant motion or variable motion for driving the training device 604. An athlete can perform blocking exercises using the training device 604 configured to increase motion until a threshold value. The threshold value can be associated with an amount of force that athlete provides, e.g., by pushing against the attachment 612.

[0282] Although multiple different implementations of the training device are described in this specification, some implementations can include combinations of the implementations described herein. As an example, the sensor-based smart technology fortraining device 104 can be combined with any of the disclosed training devices, e.g., the training device 104 of FIG. 1 A through ID, the training device 404 of FIG. 4A, the training device 454 of FIG. 4B, the training device 504 of FIG.5A, and / or the training device 604 of FIG. 6. As another example, the weight adjustment mechanisms, damping systems and / or mechanisms, and / or bumpers, can also be implemented with the training devices described in reference to FIGS. 1A through ID, 4A, 4B, 5A, and 6. Similarly, the attachments described in reference to FIGS. 5A through 5C can also be combined with the training devices described in FIG. 1A through ID, 4A, 4B, and FIG. 6. As another example, the frame structural member described in FIGS. 1C and ID can be combined with any of the disclosed training devices, e.g., the training device 104 of FIG. 1A through ID, the training device 404 of FIG. 4A, the training device 454 of FIG. 4B, the training device 504 of FIG. 5A, and / or the training device 604 of FIG. 6. The motorized training device 604 described in FIG. 6 can be combined with the features of the training devices depicted in FIGS. 1 A through ID, FIG. 4A, FIG. 4B, and FIG.5A.

[0283] FIG. 7 is a block diagram of computing devices 700, 750 that may be used to implement the systems and methods described in this document, as either a client or as a server or multiple servers. Computing devices 700 and 750 represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The components shown here, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations described and / or claimed in this document.

[0284] Computing device 700 includes a processor 702, memory 704, a storage device 706, a high-speed interface 708 connecting to memory 704 and high-speed expansion ports 710, and aAttorney Docket No.: 36610-0120W01low speed interface 712 connecting to low speed bus 714 and storage device 706. Each of the components 702, 704, 706, 708, 710, and 712, are interconnected using various busses, and may be mounted on a common motherboard or in other manners as appropriate. The processor 702 can process instructions for execution within the computing device 700, including instructions stored in the memory 704 or on the storage device 706 to display graphical information for a GUI on an external input / output device, such as display 716 coupled to high speed interface 708. In other implementations, multiple processors and / or multiple buses may be used, as appropriate, along with multiple memories and types of memory. Also, multiple computing devices 700 may be connected, with each device providing portions of the necessary operations, e.g., as a server bank, a group of blade servers, or a multi-processor system.

[0285] The memory 704 stores information within the computing device 700. In one implementation, the memory 704 is a computer-readable medium. In one implementation, the memory 704 is a volatile memory unit or units. In another implementation, the memory 704 is a non-volatile memory unit or units.

[0286] The storage device 706 is capable of providing mass storage for the computing device 700. In one implementation, the storage device 706 is a computer-readable medium. In various different implementations, the storage device 706 may be a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid-state memory device, or an array of devices, including devices in a storage area network or other configurations. In one implementation, a computer program product is tangibly embodied in an information carrier. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer- or machine-readable medium, such as the memory 704, the storage device 706, or memory on processor 702.

[0287] The high-speed controller 708 manages bandwidth-intensive operations for the computing device 700, while the low speed controller 712 manages lower bandwidth-intensive operations. Such allocation of duties is exemplary only. In one implementation, the high-speed controller 708 is coupled to memory 704, display 716, e.g., through a graphics processor or accelerator, and to high-speed expansion ports 710, which may accept various expansion cards (not shown). In the implementation, low-speed controller 712 is coupled to storage device 706 and low-speed expansion port 714. The low-speed expansion port, which may include various communication ports, e.g., USB, Bluetooth, Ethernet, wireless Ethernet, may be coupled to one or moreAttorney Docket No.: 36610-0120W01input / output devices, such as a keyboard, a pointing device, a scanner, or a networking device such as a switch or router, e.g., through a network adapter.

[0288] The computing device 700 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a standard server 720, or multiple times in a group of such servers. It may also be implemented as part of a rack server system 724. In addition, it may be implemented in a personal computer such as a laptop computer 722. Alternatively, components from computing device 700 may be combined with other components in a mobile device (not shown), such as device 750. Each of such devices may contain one or more of computing device 700, 750, and an entire system may be made up of multiple computing devices 700, 750 communicating with each other.

[0289] Computing device 750 includes a processor 752, memory 764, an input / output device such as a display 754, a communication interface 766, and a transceiver 768, among other components. The device 750 may also be provided with a storage device, such as a microdrive or other device, to provide additional storage. Each of the components 750, 752, 764, 754, 766, and 768, are interconnected using various buses, and several of the components may be mounted on a common motherboard or in other manners as appropriate.

[0290] The processor 752 can process instructions for execution within the computing device 750, including instructions stored in the memory 764. The processor may also include separate analog and digital processors. The processor may provide, for example, for coordination of the other components of the device 750, such as control of user interfaces, applications run by device 750, and wireless communication by device 750.

[0291] Processor 752 may communicate with a user through control interface 758 and display interface 756 coupled to a display 754. The display 754 may be, for example, a TFT LCD display or an OLED display, or other appropriate display technology. The display interface 756 may include appropriate circuitry for driving the display 754 to present graphical and other information to a user. The control interface 758 may receive commands from a user and convert them for submission to the processor 752. In addition, an external interface 762 may be provided in communication with processor 752, so as to enable near area communication of device 750 with other devices. External interface 762 may provide, for example, for wired communication, e.g., via a docking procedure, or for wireless communication, e.g., via Bluetooth or other such technologies.Attorney Docket No.: 36610-0120W01

[0292] The memory 764 stores information within the computing device 750. Tn one implementation, the memory 764 is a computer-readable medium. In one implementation, the memory 764 is a volatile memory unit or units. In another implementation, the memory 764 is a non-volatile memory unit or units. Expansion memory 774 may also be provided and connected to device 750 through expansion interface 772, which may include, for example, a SIMM card interface. Such expansion memory 774 may provide extra storage space for device 750 or may also store applications or other information for device 750. Specifically, expansion memory 774 may include instructions to carry out or supplement the processes described above and may include secure information also. Thus, for example, expansion memory 774 may be provided as a security module for device 750 and may be programmed with instructions that permit secure use of device 750. In addition, secure applications may be provided via the SIMM cards, along with additional information, such as placing identifying information on the SIMM card in a non-hackable manner.

[0293] The memory may include, for example, flash memory and / or MRAM memory, as discussed below. In one implementation, a computer program product is tangibly embodied in an information carrier. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer- or machine-readable medium, such as the memory 764, expansion memory 774, or memory on processor 752.

[0294] Device 750 may communicate wirelessly through communication interface 766, which may include digital signal processing circuitry where necessary. Communication interface 766 may provide for communications under various modes or protocols, such as GSM voice calls, SMS, EMS, or MMS messaging, CDMA, TDMA, PDC, WCDMA, CDMA2000, or GPRS, among others. Such communication may occur, for example, through radio-frequency transceiver 768. In addition, short-range communication may occur, such as using a Bluetooth, WiFi, or other such transceiver (not shown). In addition, GPS receiver module 770 may provide additional wireless data to device 750, which may be used as appropriate by applications running on device 750.

[0295] Device 750 may also communicate audibly using audio codec 760, which may receive spoken information from a user and convert it to usable digital information. Audio codec 760 may likewise generate audible sound for a user, such as through a speaker, e.g., in a handset of device 750. Such sound may include sound from voice telephone calls, may include recorded sound, e.g.,Attorney Docket No.: 36610-0120W01voice messages, music files, etc., and may also include sound generated by applications operating on device 750.

[0296] The computing device 750 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a cellular telephone 780. It may also be implemented as part of a smartphone 782, personal digital assistant, or other similar mobile device.

[0297] In this specification, the term “engine” or “software engine” refers to a software implemented input / output system that provides an output that is different from the input. An engine can be an encoded block of functionality, such as a library, a platform, a software development kit (“SDK”), or an object. Each engine can be implemented on any appropriate type of computing device, e.g., servers, mobile phones, tablet computers, notebook computers, music players, e-book readers, laptop or desktop computers, PDAs, smart phones, or other stationary or portable devices, that includes one or more processors and computer readable media. Additionally, two or more of the engines may be implemented on the same computing device, or on different computing devices.

[0298] The term “data processing apparatus” encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. Data processing apparatus can include specialpurpose logic circuitry, e.g., an FPGA (field programmable gate array), an ASIC (applicationspecific integrated circuit), or a GPU (graphics processing unit). The apparatus can also include, in addition to hardware, code that creates an execution environment for computer programs, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

[0299] Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs, computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0300] These computer programs, also known as programs, software, software applications or code, include machine instructions for a programmable processor, and can be implemented in aAttorney Docket No.: 36610-0120W01high-level procedural and / or object-oriented programming language, and / or in assembly / machine language. As used herein, the terms “machine-readable medium” “computer-readable medium” refers to any computer program product, apparatus and / or device, e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0301] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0302] This specification uses the term “configured to” in connection with systems, apparatus, and computer program components. A system of one or more computers is configured to perform particular operations or actions means that the system has installed on it software, firmware, hardware, or a combination of them that in operation cause the system to perform the operations or actions. One or more computer programs is configured to perform particular operations or actions means that the one or more programs include instructions that, when executed by data processing apparatus, cause the apparatus to perform the operations or actions. Special-purpose logic circuitry is configured to perform particular operations or actions means that the circuitry has electronic logic that performs the operations or actions.

[0303] The systems and techniques described here can be implemented in a computing system that includes a back-end component, e.g., as a data server, or that includes a middleware component such as an application server, or that includes a front-end component such as a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here, or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication such as, a communicationAttorney Docket No.: 36610-0120W01network. Examples of communication networks include a local area network (“LAN”), a wide area network (“WAN”), and the Internet.

[0304] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0305] In addition, certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed. For example, in some embodiments, a user’s identity may be treated so that no personally identifiable information can be determined for the user, or a user’s geographic location may be generalized where location information is obtained (such as to a city, ZIP code, or state level), so that a particular location of a user cannot be determined. Thus, the user may have control over what information is collected about the user, how that information is used, and what information is provided to the user.

[0306] In addition to the embodiments described above, the following embodiments are also innovative:

[0307] Embodiment 1 is an exercise device comprising:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion;a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0308] Embodiment 2 is the exercise device of embodiment 1, further comprising:a first strike pad attachment configured to be attached to the first end of the frame structural member, wherein the first strike pad attachment comprises at least a first additional sensor configured to capture sensor measurements.

[0309] Embodiment 3 is the exercise device of embodiment 1, further comprising:

[0310] a second strike pad attachment configured to be attached to the second end of the frame structural member, wherein the second strike pad attachment comprises at least a second additionalAttorney Docket No.: 36610-0120W01sensor configured to capture sensor measurements. Embodiment 4 is the exercise device of embodiment 1, wherein each of the respective top portion and the respective bottom portion for one or both of the first end or the second end comprises a respective bumper configured to provide support for the exercise device.

[0311] Embodiment 5 is the exercise device of embodiments 1 to 4, wherein the computing device comprises one or more processors coupled to one or more computer-readable storage media having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:obtaining, from a first sensor from the plurality of sensors, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device; anddetermining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0312] Embodiment 6 is the exercise device of embodiment 5, wherein the operations further comprise:generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

[0313] Embodiment 7 is the exercise device of embodiment 5, wherein the operations further comprise:obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device; anddetermining, from the second sensor data, a second metric representing a characteristic for the exercise action.

[0314] Embodiment 8 is the exercise device of embodiment 5, wherein the operations further comprise:obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device; anddetermining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.Attorney Docket No.: 36610-0120W01

[0315] Embodiment 9 is the exercise device of embodiment 7, wherein the operations further comprise:generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

[0316] Embodiment 10 is the exercise device of embodiment 1, wherein the exercise device is a football training sled.

[0317] Embodiment 11 is the exercise device of embodiment 1, wherein sensor measurements captured by a sensor from the plurality of sensors is captured prior to an exercise being performed using the exercise device.

[0318] Embodiment 12 is the exercise device of embodiment 1, wherein sensor measurements captured by a sensor from the plurality of sensors is captured while an exercise is being performed using the exercise device.

[0319] Embodiment 13 is the exercise device of embodiment 1, wherein sensor measurements captured by a sensor from the plurality of sensors is captured after an exercise is performed using the exercise device.

[0320] Embodiment 14 is the exercise device of embodiment 1, wherein the plurality of sensors at least one of (i) a radar sensor, (ii) a LIDAR sensor, (iii) a global position system sensor, (iv) an ultrasonic sensor, (v) an infrared sensor, (vi) a camera sensor, (vii) an image sensor, (viii) an optical sensor, (ix) a light sensor, (x) a proximity sensor, (xi) a narrow-beam ultrasonic sensor, (xii) a power sensor, (xiii) a force sensor and (xiv) a load cell.

[0321] Embodiment 15 is the exercise device of embodiment 1 wherein the computing device comprises a model configured to apply one or more of (i) machine learning techniques, (ii) statistical techniques, or (iii) a physical-based simulation, to generate the output based on the sensor measurements.

[0322] Embodiment 16 is the exercise device of embodiment 1, wherein the computing device is configured to:generate, based on the output, an evaluation of an exercise performed; andprovide data indicative of the evaluation.

[0323] Embodiment 17 is the exercise device of embodiment 1, wherein to provide the data indicative of evaluation comprises providing the data for display on a user interface of a deviceAttorney Docket No.: 36610-0120W01communicatively coupled to the computing device, wherein the data causes the device to configure the user interface to display one or more user interface elements to display the evaluation.[003241 Embodiment 18 is the exercise device of embodiment 1, wherein at least one sensor from the plurality of sensors is mounted onto the exercise device.

[0325] Embodiment 19 is the exercise device of embodiment 1, wherein at least one sensor from the plurality of sensors is co-located with the computing device.

[0326] Embodiment 20 is the exercise device of embodiment 1, wherein at least one sensor from the plurality of sensors is configured to capture sensor data in a non-visible electromagnetic spectrum frequency band.

[0327] Embodiment 21 is the exercise device of embodiment 1, wherein at least one sensor from the plurality of sensors is configured to capture sensor data in one of radio frequency bands, microwave frequency bands, acoustic frequency bands, infrared frequency bands, and ultrasonic frequency band.

[0328] Embodiment 22 is the exercise device of embodiment 1, wherein sensor measurements from at least one sensor from the plurality of sensors comprises point cloud data.

[0329] Embodiment 23 is the exercise device of embodiment 1, , wherein the computing device is configured to:determine, based on the sensor measurements, a first value representing friction between a surface and the exercise device; andbased on a comparison of the first value representing friction and a second value representing a target metric, generating an indicator to adjust the exercise device /

[0330] Embodiment 24 is the exercise device of embodiment 23, wherein the computing device is configured to determine a type of the surface based on the first value representing friction.

[0331] Embodiment 25 is the exercise device of embodiment 1, wherein the respective bottom portion of one or both of the first end and the second end comprises at least one of (i) brakes, (ii) wheels, (iii) adjustable braking mechanisms, (iv) a ski, (v) a pair of skis, (vi) a disk, (vii) a bowl, or (viii) a saucer.

[0332] Embodiment 26 is the exercise device of embodiment 25, wherein the at least one of (i) the brakes, (ii) the wheels, (iii) the adjustable braking mechanisms, (iv) the ski, (v) the pair of skis, (vi) the disk, (vii) the bowl, or (viii) the saucer is rotatable about an axis relative to the exercise device.Attorney Docket No.: 36610-0120W01

[0333] Embodiment 27 is the exercise device of embodiment 1, wherein the respective top portion of at least one of the first end or the second end further comprises one or more handles.

[0334] Embodiment 28 is the exercise device of embodiment 27, wherein at least one handle from the one or more handles is a horizontally oriented handle connecting a first portion of the respective top portion to a second portion of the respective top portion different from the first portion.

[0335] Embodiment 29 is the exercise device of embodiment 27, wherein at least one handle from the one or more handles is a horizontally oriented handle connecting a first portion of the respective bottom portion to a second portion of the respective bottom portion different from the first portion.

[0336] Embodiment 30 is the exercise device of embodiment 27, wherein at least one handle from the one or more handles is a vertically oriented handle protruding upward from one or both of the respective top portion and the respective bottom portion.

[0337] Embodiment 1 is the exercise device of embodiment 27, wherein a handle from the one or more handles is rotatable about an axis.

[0338] Embodiment 32 is the exercise device of embodiment 27, , wherein a handle from the one or more handles comprises a sensor configured to capture measurements.

[0339] Embodiment 33 is the exercise device of embodiment 1, wherein the respective bottom portion of the first end comprises one or more wheels and the respective bottom portion of the second end comprises an adjustable braking mechanism.

[0340] Embodiment 34 is the exercise device of embodiment 1, wherein the respective bottom portion of the first end comprises at least one ski or at least one wheel that connects the respective bottom portion of the first end to the respective bottom portion of the second end.

[0341] Embodiment 35 is the exercise device of embodiment 1, the plurality of sensors is at least one of a camera sensor and an image sensor, wherein the at least one of the camera sensor and the image sensor is configured for facial recognition by the computing device.

[0342] Embodiment 36 is the exercise device of embodiment 1, further comprising one or more damping mechanisms, materials, and absorbers.

[0343] Embodiment 37 is the exercise device of embodiment 1, further comprising one or more flywheel devices.

[0344] Embodiment 38 is the exercise device of embodiment 2, further comprising a third strike pad attachment configured to be attached to the first end of the frame structural member and adjacent to the first strike pad attachment.Attorney Docket No.: 36610-0120W01

[0345] Embodiment 39 is the exercise device of embodiment 38, wherein the third strike pad attachment comprises at least a third additional sensor configured to capture sensor measurements.

[0346] Embodiment 40 is a strike pad adapted for attachment to an exercise device of embodiment 1. The exercise device comprises:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion;a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0347] Embodiment 41 is the strike pad of embodiment 40, comprising one or more additional sensors configured to capture sensor measurements of actions performed with the strike pad.

[0348] Embodiment 42 is the strike pad of embodiment 40, comprising an adapter configured to interface with the exercise device.

[0349] Embodiment 43 is the strike pad of embodiment 42, wherein the adapter comprises at least one of (i) an electrical connector, and (ii) a mechanical connector, to attach the strike pad to the exercise device.

[0350] Embodiment 44 is the strike pad of embodiment 43, wherein the electrical connector comprises at least one of (i) a wire, (ii) a plug connector, (iii) a coaxial connector, (iv) a magnetic connector, or (v) a spring loaded connector.

[0351] Embodiment 45 is the strike pad of embodiment 43, wherein the mechanical connector comprises at least one of (i) a snap-fit connector, (ii) a fastener, (iii) a quick-release mechanism, (iv) a clamp, or (v) a hinge.

[0352] Embodiment 46 is the strike pad of embodiment 40, comprising one or both of (i) a processor, or (ii) data storage.

[0353] Embodiment 47 is a method comprising:determining, by a strike pad, that the strike pad has been removably connected to a strength training exercise device;transmitting features of the strike pad to the strength training exercise device; andAttorney Docket No.: 36610-0120W01receiving data from the strength training exercise device based on the features of the strike pad.[003541 Embodiment 48 is a method comprising:obtaining, by a first sensor of a strike pad, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the strike pad, wherein the strike pad is adapted to be removably connected to a strength training exercise device; and determining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0355] Embodiment 49 is a method comprising:obtaining, by at least one sensor of a strike pad adapted for attachment to an exercise device, first sensor data comprising first sensor measurements captured by the at least one sensor for an exercise action with the strike pad; anddetermining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0356] Embodiment 50 is a method comprising:obtaining, by a first sensor from a plurality of sensors of an exercise device, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device; anddetermining, from the first sensor data, a first metric representing a characteristic for the exercise action.

[0357] Embodiment 51 is the method of embodiment 50, further comprising:generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

[0358] Embodiment 52 is the method of embodiment 50,obtaining, by a second sensor from the plurality of sensors, second sensor data comprising second sensor measurements captured by the second sensor for the exercise action with the exercise device, wherein the second sensor is a different from the first sensor; and determining, from the second sensor data, a second metric representing a characteristic for the exercise action.

[0359] Embodiment 53 is the method of embodiment 50, further comprising:Attorney Docket No.: 36610-0120W01obtaining, by a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor for the exercise action with the exercise device; anddetermining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.

[0360] Embodiment 54 is the method of embodiment 53, further comprising:generating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

[0361] Embodiment 55 is the method of embodiment 53, wherein the first sensor is a first type of sensor, and the second sensor is a second type of sensor, the second type being different from the first type.

[0362] Embodiment 56 is the method of embodiment 53, further comprising:determining, by a computing device, a first value representing friction between a surface and the exercise device, wherein the determination is based on the first sensor measurements and / or the second sensor measurements; andbased on a comparison of the first value representing friction and a second value representing a target metric, generating an indicator to adjust the exercise device.

[0363] Embodiment 57 is the method of embodiment 50, further comprising generating calibration data for the exercise device based on the first sensor measurements.

[0364] Embodiment 58 is the method of embodiment 50, further comprising generating feedback data based on the first metric.

[0365] Embodiment 59 is an exercise device comprising:an adjustable frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the adjustable frame structural member, wherein each of the first end and the second end of the adjustable frame structural member further comprise a respective top portion and a respective bottom portion;wherein the adjustable frame structural member comprises an inner portion and an outer portion, wherein the inner portion is slidably connected to the outer portion by a mechanism, wherein the mechanism allows the inner portion to be extended or retracted.Attorney Docket No.: 36610-0120W01

[0366] Embodiment 60 is the exercise device of embodiment 59, wherein the adjustable frame structural member further comprises one or more additional adjustable frame structural members, each of one or more additional adjustable frame structural members being substantially parallel to the adjustable frame structural member.

[0367] Embodiment 61 is the exercise device of embodiment 60, wherein each of the one or more additional adjustable frame structural members is connected to the adjustable frame structural member by one or more horizontally oriented structural members.

[0368] Embodiment 62 is the exercise device of embodiment 59, wherein one of both of (i) an additional adjustable frame structural members from the one or more additional adjustable frame structural members, and (ii) the adjustable frame structural member, comprises one or more compartments.

[0369] Embodiment 63 is the exercise device of embodiment 59, further comprising an adjustable weight mechanism.

[0370] Embodiment 64 is the exercise device of embodiment 63, wherein the adjustable weight mechanism is configured to adjust an amount of weight of the exercise device using a slidable pin stack system.

[0371] Embodiment 65 is the exercise device of embodiment 64, wherein the slidable pin stack system comprises:one or more openings, each opening corresponding to a weight from a plurality of weights; andat least one pin configured to be inserted in an opening from the one or more openings to select the weight corresponding to the opening.

[0372] Embodiment 66 is the exercise device of embodiment 59, further comprising one or more damping mechanisms, materials, and absorbers.

[0373] Embodiment 67 is the exercise device of embodiment 59, further comprising one or more flywheel devices.

[0374] Embodiment 68 is a rounded exercise pad adapted for attachment to an exercise device of embodiment 1. The exercise device comprises:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structuralAttorney Docket No.: 36610-0120W01member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion;a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0375] Embodiment 69 is the rounded exercise pad of embodiment 68, comprising one or more additional sensors configured to capture sensor measurements of actions performed with the rounded exercise pad.

[0376] Embodiment 70 is the rounded exercise pad of embodiment 68, comprising an adapter configured to interface with the exercise device.

[0377] Embodiment 71 is the rounded exercise pad of embodiment 70, wherein the adapter comprises at least one of (i) an electrical connector, and (ii) a mechanical connector, to attach the rounded exercise pad to the exercise device.

[0378] Embodiment 72 is the rounded exercise pad of embodiment 71, wherein the electrical connector comprises at least one of (i) a wire, (ii) a plug connector, (iii) a coaxial connector, (iv) a magnetic connector, or (v) a spring loaded connector.

[0379] Embodiment 73 is the rounded exercise pad of embodiment 71, wherein the mechanical connector comprises at least one of (i) a snap-fit connector, (ii) a fastener, (iii) a quick-release mechanism, (iv) a clamp, or (v) a hinge.

[0380] Embodiment 74 is the rounded exercise pad of embodiment 68, comprising one or both of (i) a processor, or (ii) data storage.

[0381] Embodiment 75 is the rounded exercise pad of embodiment 68, wherein the rounded exercise pad is connected to the exercise device by a rotating mechanism.

[0382] Embodiment 76 is the rounded exercise pad of embodiment 68, wherein the rounded exercise pad is one of (i) a medicine ball, (ii) a wall ball, or (iii) a weighted ball.

[0383] Embodiment 77 is an exercise device comprising:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion; andAttorney Docket No.: 36610-0120W01a propulsion system configured to displace the exercise device along a surface.

[0384] Embodiment 78 is the exercise device of embodiment 77, further comprising:a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0385] Embodiment 79 is the exercise device of embodiment 78, further comprising:a drivetrain system coupled to the propulsion system;one or more driving elements coupled to the drivetrain system; andone or more braking elements coupled to at least one of the one or more driving elements.

[0386] Embodiment 80 is the exercise device of embodiment 78, further comprising:a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

[0387] Embodiment 81 is the exercise device of embodiment 78, wherein the respective bottom portion of one or both of the first end and the second end comprises at least one of (i) brakes, (ii) wheels, (iii) adjustable braking mechanisms, (iv) a ski, or (v) a pair of skis, (vi) disk, (vii) bowl, or (viii) saucer.

[0388] Embodiment 82 is the exercise device of embodiment 78, wherein the propulsion system comprises:a power source configured to provide power; andone or more propulsion elements coupled to the power source and configured to generate force based on power received from the power source.

[0389] Embodiment 83 is the exercise device of any preceding embodiment, wherein the exercise device is configured to receive sensor measurements from a sensor remote from the exercise device, and wherein the sensor is a force plate located below the exercise device.

[0390] Embodiment 84 is the exercise device of embodiment 83, wherein the force plate is part of a surface below the exercise device.

[0391] Embodiment 85 is an exercise device comprising:Attorney Docket No.: 36610-0120W01a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion; andat least one pulling mechanism coupled to at least one weight, wherein the pulling mechanisms is configured to adjust a position of the at least one weight along the longitudinal axis of the frame structural member.

[0392] Embodiment 86 is the exercise device of embodiment 85, wherein the at least one pulling mechanism is configured to adjust the position of the at least one weight in response to a force that is (i) applied to and / or (ii) being applied to, the at least one pulling mechanism.

[0393] Embodiment 87 is the exercise device of embodiment 86, wherein an adjustment of the position of the at least one weight along the longitudinal axis is based on, at least in part, the force experienced by the at least one pulling mechanism.

[0394] Embodiment 88 is the exercise device of embodiment 86, wherein the force experienced by the at least one pulling mechanism is a force that is (i) applied to and / or (ii) being applied to, the exercise device.

[0395] Embodiment 89 is an exercise device comprising:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion, wherein the respective bottom portion for one or both of the first end and the second end comprises a rounded shape.

[0396] A number of embodiments have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosed technology. Accordingly, other embodiments are within the scope of the following claims. While this specification contains many specific implementation details, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment.Attorney Docket No.: 36610-0120W01

[0397] Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a sub combination or variation of a sub combination.

[0398] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system modules and components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0399] Particular embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. For example, the actions recited in the claims can be performed in a different order and still achieve desirable results. As one example, some processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results.

[0400] What is claimed is:

Claims

Attorney Docket No.: 36610-0120W01CLAIMS1. An exercise device comprising:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion;a plurality of sensors, each sensor from the plurality of sensors configured to capture sensor measurements; anda computing device configured to process the sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

2. The exercise device of claim 1, further comprising at least one of:(i) a first strike pad attachment configured to be attached to the first end of the frame structural member, wherein the first strike pad attachment comprises at least a first additional sensor configured to capture a first set of sensor measurements; and(ii) a second strike pad attachment configured to be attached to the second end of the frame structural member, wherein the second strike pad attachment comprises at least a second additional sensor configured to capture a second set of sensor measurements.

3. The exercise device of claim 1, wherein the computing device comprises one or more processors coupled to one or more computer-readable storage media having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:obtaining, from a first sensor from the plurality of sensors, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device; anddetermining, from the first sensor data, a first metric representing a characteristic for the exercise action.

4. The exercise device of claim 3, wherein the operations further comprise:Attorney Docket No.: 36610-0120W01generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

5. The exercise device of claim 3, wherein the operations further comprise:obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device; anddetermining, from the second sensor data, a second metric representing a characteristic for the exercise action.

6. The exercise device of claim 3, wherein the operations further comprise:obtaining, from a second sensor from the plurality of sensors different from the first sensor, second sensor data comprising second sensor measurements captured by the second sensor of the exercise action with the exercise device; anddetermining, from the second sensor data and the first sensor data, a second metric representing a characteristic for the exercise action.

7. The exercise device of claim 5, wherein the operations further comprise:generating, by a model of the computing device and for the exercise action, an evaluation of the exercise action based on the first metric and the second metric.

8. The exercise device of claim 1, wherein the exercise device is a football training sled.

9. The exercise device of claim 1, wherein the sensor measurements captured by a sensor from the plurality of sensors is captured: (i) prior to an exercise action being performed using the exercise device, (ii) while an exercise action is being performed using the exercise device, or (iii) after an exercise action is performed using the exercise device.

10. The exercise device of claim 1, wherein the plurality of sensors at least one of (i) a radar sensor, (ii) a LIDAR sensor, (iii) a global position system sensor, (iv) an ultrasonic sensor,Attorney Docket No.: 36610-0120W01(v) an infrared sensor, (vi) a camera sensor, (vii) an image sensor, (viii) an optical sensor, (ix) a light sensor, (x) a proximity sensor, (xi) a narrow-beam ultrasonic sensor, (xii) a power sensor, (xiii) a force sensor and (xiv) a load cell.

11. The exercise device of claim 1, wherein the computing device comprises a model configured to apply one or more of (i) machine learning techniques, (ii) statistical techniques, or (iii) a physical-based simulation, to generate the output based on the sensor measurements.

12. The exercise device of claim 1, wherein the computing device is configured to:generate, based on the output, an evaluation of an exercise performed; and provide data indicative of the evaluation by providing the data for display on a user interface of a device communicatively coupled to the computing device, wherein the data causes the device to configure the user interface to display one or more user interface elements to display the evaluation.

13. The exercise device of claim 1, wherein at least one sensor from the plurality of sensors is (i) mounted onto the exercise device, or (ii) co-located with the computing device.

14. The exercise device of claim 1, wherein at least one sensor from the plurality of sensors is configured to capture sensor data in one of (i) a non-visible electromagnetic spectrum frequency band, (ii) radio frequency bands, (iii) microwave frequency bands, (iv) acoustic frequency bands, (v) infrared frequency bands, and (vi) ultrasonic frequency band.

15. The exercise device of claim 1, wherein the sensor measurements from at least one sensor from the plurality of sensors comprises point cloud data.

16. The exercise device of claim 1, wherein the respective top portion of at least one of the first end or the second end further comprises one or more handles, and wherein a handle from the one or more handles comprises a sensor configured to capture measurements.Attorney Docket No.: 36610-0120W0117. The exercise device of claim 1 , wherein the plurality of sensors is at least one of a camera sensor and an image sensor, wherein the at least one of the camera sensor and the image sensor is configured for facial recognition by the computing device.

18. A strike pad adapted for attachment to an exercise device, the strike pad comprising one or more additional sensors configured to capture sensor measurements of actions performed with the strike pad, wherein the exercise device comprises:a frame structural member having a first end and a second end opposite of the first end, the first end and the second end defining a longitudinal axis along a length of the frame structural member, wherein each of the first end and the second end of the frame structural member further comprise a respective top portion and a respective bottom portion;a plurality of sensors, each sensor from the plurality of sensors configured to capture the sensor measurements; anda computing device configured to process the sensor measurements from the plurality of sensors and generate an output based on the sensor measurements.

19. The strike pad of claim 18, comprising an adapter configured to interface with the exercise device, wherein the adapter comprises at least one of (i) an electrical connector, and (ii) a mechanical connector, to attach the strike pad to the exercise device.

20. A method comprising:obtaining, by a first sensor from a plurality of sensors of an exercise device, first sensor data comprising first sensor measurements captured by the first sensor for an exercise action with the exercise device;determining, from the first sensor data, a first metric representing a characteristic for the exercise action; andgenerating, by a model of a computing device and for the exercise action, an evaluation of the exercise action based on the first metric.

Images