Worn sensor based task training system

Abstract

NOT VISIBLE DUE TO STATUS OF PATENT

Description

ATTORNEY DOCKET NO.0215.0006-PCT WORN SENSOR BASED TASK TRAINING SYSTEM _________________________________________________ Related Application

[0001] This application claims priority to, and the benefit of, U.S. Provisional Patent Application No. 63 / 619,205, filed January 9th, 2024, and entitled, “Athlete Training System With Biometric Sensory Input”, which is hereby incorporated by reference herein in its entirety. Field

[0002] This disclosure relates generally to training systems and, more particularly, to sensor-based intelligence systems used to train a user to increase proficiency in completing one or more tasks. Background

[0003] As sensing and computing technology have improved, the size and speed of information analysis, processing, and delivery have advanced. Sensor technology has similarly improved over time to provide increasingly accurate readings and measurements faster than before. Such advancements have allowed greater numbers, and types, of activities to be automated, such as manufacturing, assembly, and operation. However, the real-time use of sensors to improve activity and analysis of user instructions has not been contemplated. Hence, assorted embodiments employ sensors worn by a user in conjunction with modern computing capabilities to interpret user activities and provide intelligent instructions to the user directed to optimize the completion of one or more user tasks. Summary

[0004] Embodiments of the present disclosure are generally directed to training systems, such as but not limited to, task-specific training based on sensed operator activity.

[0005] A smart headband, in some embodiments, may be designed to utilize integrated sensors and AI-driven analytics to monitor biomechanics, physiological data, and mental focus during activities, behaviors, and tasks. A smart headband may utilize biomechanical sensors, physiological sensors, machine learning algorithms, a real-timefeedback system, focus tracking, and state tracking. Biomechanical sensing may measure physical metrics, such as finger spacing, body alignment, movement patterns, and other movement-related data while physiological sensors monitor heart rate, breath, fatigue, and other bodily signals. Machine learning algorithms may analyze the data, identify patterns, and provide real-time feedback on performance while a real-Time Feedback System delivers immediate, personalized feedback that helps improve operational technique, mental clarity, and physical readiness of a user. The tracking of a user’s focus and neurological state may use data inputs, like sleep quality, time of day, and other factors, to generate focus or clarity scores that provide insights into how mental and physical states affect performance.

[0006] In some embodiments, a computing device may be connected to an array of sensors positioned proximal to a user. The computing device may determine a task to be completed by the user in response to information detected by the array of sensors. The computing device may generate a set of instructions directing the user to complete the task and a communication strategy in response to the information detected by the array of sensors with the communication strategy prescribing different communication modes associated with different user actions during the completion of the task. The array of sensors may sense real-time user conditions during the completion of the task along with real-time environmental conditions during the completion of the task. The computing device may determine that the user will fail to complete a predetermined milestone of the task based on the real-time user conditions and the real-time environmental conditions and proceed to adapt at least one instruction of the set of instructions in response to the failure of the user to complete the predetermined milestone, the adaptation of the at least one instruction changing how the user is taught to perform a sub-step of the task.

[0007] These and other features which characterize various embodiments of the present disclosure can be understood in view of the following detailed discussion and the accompanying drawings. Brief Description of the Drawings

[0008] FIG. 1 is a functional block of a training environment in which various embodiments of the present disclosure can be practiced.

[0009] FIG. 2 is a block representation of a training system that may be utilized in the environment of FIG. 1 in accordance with some embodiments.

[0010] FIG. 3 is a block representation of a training module that may be employed as part of an athlete training system in accordance with assorted embodiments.

[0011] FIG. 4 is a block representation of a training system configured and operated in accordance with various embodiments.

[0012] FIG. 5 is a perspective view line representation of portions of a training system arranged in accordance with assorted embodiments.

[0013] FIG. 6 is a flowchart depicting example operation of a training system performed in accordance with various embodiments.

[0014] FIG. 7 is a flowchart depicting example operation of a training system conducted in accordance with some embodiments.

[0015] FIG. 8 is a flowchart depicting example operations of a training system executed in accordance with assorted embodiments. Detailed Description

[0016] Various embodiments relate to wearable technology, specifically a smart headband designed for monitoring and optimizing athletic performance, particularly in sports. The device integrates machine learning, real-time data analytics, and biomechanical tracking to provide personalized feedback for improving performance, mental clarity, and physical effort during sports activities.

[0017] Such a smart headband may have applications in sports training, physical rehabilitation, and mental performance optimization. More specifically, a smart headband may optimize basketball shooting. However, a smart headband may also be applicable to other sports, such as soccer, tennis, and golf, by adapting the system to specific metrics. A smart headband may be useful for rehabilitation programs focusing on improving motor control, muscle memory, and biomechanics. Additionally, a smart headband may monitor and improve focus, and mental clarity, for athletes under competitive pressure.

[0018] Generally, embodiments are at an intelligent training system that utilizes sensors worn by a user to improve the efficiency and / or accuracy of user activities. By employing sensed conditions and user tasks in real-time with sensors, a training system may intelligently generate instructions and communication strategies that allow the user to conduct activities faster, more confidently, and with less errors than if the user conducted the activity without aid. The ability to employ computing devices to analyze user actions and interact with the user in selected communication techniques allows for increased efficiency of instruction obeyance, particularly compared to instruction given from a human coach.

[0019] It is contemplated that current training methods often lack personalized, real- time feedback that can help users optimize their performance and improve skill development. Existing wearable technologies may track basic metrics, such as heart rate, but few provide a comprehensive analysis of athletic movement or integrate with artificial intelligence (AI) and biomechanics for a complete performance evaluation. Accordingly, embodiments of a wearable headband may use AI and machine learning algorithms to measure and analyze various biomechanical metrics such as body movement, finger spacing, dexterity, and physical states, such as breath, heart rate, and fatigue. The use of a smart headband, in accordance with various embodiments, may provide users with continuous feedback, which may help users improve their techniques based on real-time insights derived from the collected data.

[0020] In FIG. 1, a block representation of a training environment 100 is illustrated. Assorted embodiments of a training system may be conducted in the training environment 100 to provide intelligent, automated response to user movements and activity to instruct, and perfect, tasks, such as work operations, medical treatment, or crisis management. It is contemplated that any number of users 110 may participate in poses, movements, and actions that are evaluated by a network of sensors 120 to collect data that may be evaluated, analyzed, and utilized by a connected computing device 130 to produce intelligent feedback, instructions, and guidance to complete one or more tasks and / or activities.

[0021] It is noted that the number, type, and position of sensor 120 utilized to analyze aspects of one or more users 110 is not limited to a particular configuration. Accordingly,some embodiments employ stationary 122, such as cameras, LiDAR, and acoustic detectors, along with dynamic sensors 124 located on an athlete 110, such as accelerometers, force, and ultrasonic detectors. The ability to employ a variety of different sensors 120 concurrently, or sequentially, allows for the collection of a diverse range of data, which can involve environmental data, such as temperature, humidity, and wind, as well as biometric data about the user 110, such as heart rate, body temperature, and brain activity.

[0022] Some embodiments may further collect data, with one or more sensors 120, about the activity of a user 110, such as body position, force applied, and plane of movement. It is noted that selected sensors 120 may be configured to collect data for a selected amount of time, such as one second, one minute, or thirty seconds, while other sensors 120 collect data in response to a trigger, such as a sensed data from a separate sensor 122 / 124 reaching a predetermined threshold. With the detection of redundant, collective, and individual measurements and data from the assorted sensors 120, a variety of different statistical and comparative analysis can be conducted to interpret how the user 110 is acting and performing at selected instances as well as over time. That is, the use of the various sensors 120 provide data that may provide mathematical understanding of how a user 110 is performing compared to a standard, which may be utilized to identify tendencies and cyclic behaviors of the user 110.

[0023] However, the interpretation of the detected information from one or more sensors 120 may be challenging for some computing devices 130. That is, some computing devices 130 may have difficulty in processing multiple different streams of input data from various stationary 122 and / or dynamic 124 sensors to provide feedback and / or instructions that are effective at producing tangible actions from a user 110 to improve the performance of one or more tasks. It is contemplated that a human coach, or advisor, may interpret information from various sensors 120 to provide task instructions in real-time, but greater volumes of sensor 120 data may prove challenging for effective generation and communication of task instructions. Hence, the robust collection of data from the variety of sensors 120 may be difficult and / or inefficient to interpret by the user 110 or other human.

[0024] It is noted that the assorted 120 may be connected to one or more computing devices 130 in a variety of manners, such as wired or wireless signal pathways as well as local transfer from mobile memory. It is further noted that computing devices 130 may be local, such as physically present in proximity to the sensors 120, or remote, such as separated from the sensors 120 by a tangible distance. Despite the incorporation of computer processing and memory with one or more computing devices 130, the translation of collected information from sensors 120 into identified tendencies, behaviors, and deviations from ideal performance may be inefficient.

[0025] FIG. 2 conveys a block representation of portions of a training assembly 200 arranged and operated in accordance with various embodiments to provide feedback, instructions, and / or guidance to a user 110 in the environment 100 of FIG. 1. The training assembly 200 may include any number of computing devices 130 that employ one or more processors 210, such as a microcontroller, application specific integrated circuit (ASIC), or other programmable circuitry, to translate raw data from one or more sensors, such as sensors 122 / 124, into verbal, textual, and / or graphical information that conveys practical accomplishment of portions of a task.

[0026] In the non-limiting examples shown in FIG. 2, a computing device 130 may compile sensor data into representations of one or more movements, actions, or behaviors of a user 110. The processor 210 may convey such compiled sensor data to a user 110 via a graphical plotting of activity (solid line 222) compared to one or more plots (segmented line 224) corresponding with ideal activity for maximum performance. That is, information derived from sensed athlete activity can be presented to the user 110 in graphical form, once compiled by the processor 210, compared to ideal athletic activity. Such graphical comparison of actual behavior to ideal behavior may efficiently convey a relatively large amount of information. However, a user 110 may not efficiently translate the graphical comparison into practical changes, drills, or positions that will improve performance over time.

[0027] Other embodiments may utilize the computing processor 210 to provide audible and / or visual checklist of ideal behavior element list 232 compared to the sensed behavior and activity of a user 110. In other words, the processor 210 may identify goals,elements, and other aspects of an ideal such as moving a ball, performing a jump shot, lifting a weight, racing a vehicle, or hitting a ball. The identified aspects of an ideal behavior may be displayed, or spoken, to a user 110 as tasks that have been met (check mark), been missed (x mark), or yet to be achieved (question mark), as shown in list 232. Through the identification of ideal behavior aspects to be conducted by a user 110, overall behaviors, movements, and actions can be efficiently conveyed. However, a user 110 may not efficiently, or accurately, understand how to achieve certain identified ideal elements, which can correspond with inefficient growth, learning, or improvement through practice.

[0028] Embodiments of the computing device 130 utilize a processor 210 to capture aspects of a user’s behavior and identify deviations from a predetermined standard stored in memory 212 accessible by the processor 210. Such identification of how behavior is different than a standard position, motion, or application of force can be characterized as visual feedback 242 that augments a picture, video, or series of pictures with text, arrows, circles, or other identifying marks to indicate where the athlete is not performing according to a standard.

[0029] Although the computer processor 210 may efficiently analyze a user’s behavior and incorporate one or more identifiers to visual direct a user 110 to what aspects are worst, or best, as illustrated by solid arrows, such visual feedback 242 may be inefficient at conveying how to change to correct the identified issue. For instance, visual feedback 242 may identify the presence and / or intended use of one or more tools and equipment to complete portions of a task and overall activity that may comprise multiple different tasks. While the visual feedback 242 may convey where a user’s actions are deficient, such as posture, arm movement, or timing of the application of force, such feedback 242 may be inefficient at indicating what a user 110 can do to improve performance. In other words, visual feedback 242 may identify what is wrong without indicating how the user 110 can change to reach the ideal performance. Indeed, some users 110 may understand visual feedback 242 well, but others may misunderstand, or not comprehend, what is needed to conduct ideal behavior for maximum performance.

[0030] In some embodiments, the computing device 130 may generate multiple different types of feedback for a user 110. As an example, the computing processor 210may generate visual feedback 242 and analytics 222 / 224 that may be selected by a user 110 or shown concurrently to the user 110. However, the diversity of feedback may be insufficient and / or inefficient at improving the performance of a user 110. It is contemplated that conventional feedback, such as graphical analytics, visual feedback 242, and checklist elements 232, may lack progressive actions or activities to gradually correct a user’s behavior.

[0031] Instead, feedback may simply show the differences in a user’s behavior compared to an ideal standard, which may be challenging to correctly implement into some behaviors, particularly complex behaviors commonly utilized in relatively complex activities, such as heavy equipment operation, medical procedures, crisis management, and behavioral manipulation. As an example of a complex activity, the computing device 130 may provide feedback to guide a user through identifying, handling, correcting, or preventing condition in others, such as autism, heart attack, stroke, choking, physical therapy, occupational therapy, or speech therapy.

[0032] The use of a single standard for ideal behavior for a user 110 may add challenges to accurate implementation, despite efficient visual representation of a user’s deviation from ideal behavior. That is, ideal performance may be different for different users 110 and may correspond with different postures, positions, motions, and movements to reach maximum possible performance of a user’s capabilities. For instance, a single ideal standard for behavior may not factor a user’s capabilities due to body type, previous injury, current skill level, or environment. As such, the computing processor 210 can lack analysis and / or instruction to efficiently improve a user’s performance and / or reaching a user’s performance potential.

[0033] Accordingly, embodiments of the computing device 130 utilize intelligence, machine learning, and / or other circuitry to identify at least one type of feedback to communicate efficiently and accurately with a user 110 based on the determination of what tasks are being conducted, the user’s actions in real-time, and available equipment and tools that may be utilized to perform a task. Other embodiments of the training assembly 200 utilize the computing processor 210 to identify the mental and / or physical capabilities of user 110 based on sensed activity current skill, capabilities, and / or body abilities, andgenerate practical actions and / or for an athlete 110 to conduct to improve athletic behavior to reach more of an athlete’s performance potential.

[0034] FIG. 3 is a block representation of portions of a training system 300 configured and operated in accordance with various embodiments. The training system 300 may employ any number and type of computing device 130 to provide intelligent collection, analysis, and feedback to a user 110 participating in a training environment 100. The computing device 130 may employ one or more processors 210 to translate assorted input information into a variety of determinations, strategies, and milestones that can be used to efficiently communicate with a user 110 and convey training activities optimized to improve the user’s execution and / or performance of one or more tasks that can make up an activity.

[0035] While not limiting or required, various embodiments input at least data from sensors 120, biographical data about the subject user 110, environmental data, and model data to compute the mental and / or physical potential of the user 110. The assorted input data and data generated by the local, or remote 302, processors 210 may be temporarily, or permanently, stored in memory 212, which may make computations, analysis, and intelligent generation of feedback more efficient than if a memory 212 was not utilized for data storage. The ability to utilize multiple processors 210, either local or remotely located, provides robust computing capabilities that feed various operational modules. It is noted, however, that some embodiments utilize separate processors 210 for the respective operational modules of the computing device 130.

[0036] The assorted operational modules 310 / 320 / 330 / 340 shown in FIG. 3 are respectively illustrated in block form for simplicity, but may include one or more circuits physically present in a local, or remote, computing device 130, such as a tablet, smartphone, laptop, or desktop computer, that operate to translate input signals into at least one decision, determination, or conclusion. Hence, each module 310 / 320 / 330 / 340 may include numerous separate signal pathways, integrated circuits, chips, system on chips (SoC), or other programmable circuitry that provides automated selection of an answer or determination with regard to at least what type of feedback is optimal for an athlete, whatthe athletic potential of an athlete is, what can be conducted by an athlete to improve towards their potential, and if a training milestone has been achieved.

[0037] Although the processor(s) 210 / 302 of the computing device 130 may operate alone to translate input data into assorted training determinations, such as physical potential and communication means, various embodiments structurally configure the computing device 130 with separate circuitry directed to carrying out specific tasks alone, or with the aid of the available processors 210 / 302. Such task specific circuitry can be characterized as a module, but in no way limits the possible circuit configurations and operational components of the computing device 130 that may intelligently translate input data into user training aspects. In other words, circuitry of the computing device 130 can operate alone, or in combination with other circuitry, to carry out data analysis, computations, determinations, and content generation.

[0038] The computing device 130 may have a task module 310 directed at determining what a user 110 is preparing to conduct, actually conducting, or should be conducting to perform an activity comprising multiple separate tasks. Task determination by the task module 310 may involve any number, type, and sequence of input information. For instance, the task module 310 may determine a future, current, or suggested task from information ascertained from one or more sensors, known biographical information about a user, and / or model data associated with a task and / or activity. The task module 310 may produce any amount, and type, of information that may be utilized by the instruction module 320 to provide practical guidance directed to aid a user’s efficiency, accuracy, safety, and / or precision of task execution.

[0039] The instruction module 320 may input any amount, and type, of information from one or more sources to generate user actions that are within the user’s capabilities and instructions directed to execution of those actions. In various embodiments, input information, such as biographical data and sensed user behavior, is processed by the learning module 330 to learn the user’s reactions, tendencies, capabilities. Such learning may involve execution of tasks, or actions, that are generated by the learning module 330 solely to discern how a user tends to behave or physically react as well as the mental and / or physical capabilities of the user. For instance, the learning module 330 and instructionmodule 320 may create instructions for behavior, such as movement, action, or problem solving, that are unrelated to an identified task to be completed by the user in an effort to ascertain an accurate understanding of the physical and mental aspects of the user, which may allow for more effective generation of instructions, milestones, and selection of communication modes.

[0040] With an understanding of at least one task to be completed by a user and the capabilities of the user from the learning module 330, the instruction module 320 may generate a series of instructions to be completed by the user to satisfy a task and / or activity. The instruction module 320 may create completely new dialogue in response to the sensed, or computed, understanding of the capabilities and tendencies of the user. That is, the learning module 330 may input assorted information to determine what words efficiently effectuate user action, what instructions are most efficient to produce user action, and what instructions are outside of the user’s capability to physically, or mentally, translate into task-specific actions, which may be used by the instruction module 320 to provide optimized task-specific instructions customized to the user. As a non-limiting example, the learning module 330 may interpret previously logged user behavior, real-time user actions, and model user data to allow the instruction module 320 to generate an initial list of task completion instructions that is subsequently adapted over time by the instruction module 320 in response to the identification that user actions are not optimized for task completion efficiency, accuracy, or safety.

[0041] The intelligent interpretation of the user’s behavior along with the understanding of the tendencies and reactions of the user may additionally be employed by the computing device 130 to generate, and maintain, a communication strategy that prescribes particular modes of communication in response to various aspects of task completion. For instance, a communication strategy may assign different communication modes, such as graphical, textual, audible, or visual, to convey different instructions to a user. The communication strategy, in some embodiments, may assign a hierarchy of communication modes that are sequentially used in response to user actions not producing intended task completion milestones.

[0042] With the communication generated prior to a user beginning a task, and in response to what the learning module 330 has ascertained about how the user behaves in response to different forms of communication, the instruction module 320 may seamlessly transition between different instructions and / or communication types to effectively convey information that is most likely to result in aspects of a task being completed, such as designated milestone progressions designated by the computing device 130. It is noted that the computing device 130 may generate multiple different communication strategies and / or change previously assigned aspects of a communication strategy as sensed conditions change over time, such as environmental conditions, user physical capabilities, or user mental capabilities.

[0043] While not required or limiting, the instruction module 320 may develop task completion instructions by altering a default list of user actions. Such customization may involve changing the pace of instruction delivery, the communication mode, and / or changing a single instruction into a series of sub-steps. As such, the instruction module 320 can provide instructions optimized for the user’s capabilities and real-time operational condition as well as environmental conditions, such as the availability of tools or equipment. In some embodiments, various modules 310 / 320 / 330 / 340 are present on-site with a user, such as in a smart phone or equipment worn by the user, while other embodiments utilize computing hardware off-site, such a server, computer, or mainframe, to conduct one or more data analysis, task determination, and instruction generation aspects that are transmitted to the on-site computing device 130 for delivery to the user.

[0044] In addition to the assorted data collection and intelligent analysis provided by the aspects of the computing device 130, a video module 340 may operate to process various sensor input into at least one compiled video that may be recorded and replayed at any time. The video module 340, in some embodiments, converts sensor input, such as video, audio, environmental conditions, and sensed measurements, into compiled video and audio to allow streaming of user activity to one or more destinations. That is, the video module 340 may alter, remove, encrypt, condense, and aggregate different information into a single, compiled video that may be efficiently broadcast over wireless, or wired, signal pathways by the local processor 210 and / or a remote processor 302. The ability to streamcompiled video efficiently may allow for human coaching in real-time along with subsequent user review, analysis, and storage, which may be utilized by the learning module 330 to improve the instructions and / or instruction delivery to a user during completion of a later task.

[0045] In accordance with some embodiments, the computing device 130 may convert any number, type, and volume of input information into a format conducive to software analysis, organization, and interactions. For instance, the assorted device modules 310 / 320 / 330 / 340 may alter, consolidate, compress, or summarize sensed conditions and activity to allow a software application (app) to present information to a user, coach, and / or spectator in an efficient manner. An application may, in various embodiments, provide task information, user capabilities, user activity history, task milestones, and interactive aspects generated and / or computed by the computing device 130. The interactive aspects of an app may allow for selective manipulation of information, such as switching communication modes, reciting historical data, or viewing ideal task execution criteria.

[0046] Through the intelligent processing of input information by the computing device 130, a user may enjoy real-time instructions that allow for efficient understanding and completion of one or more tasks that may be characterized as an activity. FIG. 4 illustrates a block representation of portions of a training system 400 utilizing a computing device 130 in accordance with various embodiments. The system 400 employs a number of separate sensors 120 that can be stationary 122 or dynamic 124 to provide data and signals to the assorted circuitry of the computing device 130. Through the analysis of the sensed aspects of a user’s activities, the computing device 130, and specifically the learning module 330, may determine a current tendencies and capabilities of a user 110.

[0047] The sensed actions and capabilities of a user 110 in step 402 may be taken into account to determine a task, or series of tasks, that are to be conducted by the user. The determination of a task to be conducted by a user prompts the computing device 130 to evaluate the environment around a user, in step 404, to determine if a tool, or equipment, is present that may aid in the completion of the task from step 402. The identification of available equipment and / or tools may be employed by the computing device 130, andspecifically the instruction module of the 130, to generate instructions in step 406 that are directed to efficiently and accurately completing at least portions of a task.

[0048] The instructions provided to the user in step 406 may be unique to the user, based on sensed user tendencies and capabilities, or may be generic to the task at hand. The instructions provided in step 406 may be continually, routinely, or sporadically tracked by the sensors 120 of the computing device 130 to determine if the user has achieved a milestone. That is, the execution of instructions from step 406 may be monitored by the computing device 130 in step 408 to determine if a predetermined milestone has been reached. If so, the computing device 130 may transition to the next aspect of task completion. If not, step 406 may alter the instructions and / or communication mode in an effort to efficiently and accurately achieving the next milestone toward completing a task or activity.

[0049] Once the predetermined milestones of a task are achieved from step 408, the computing device 130 may advance to a different task. In addition, the computing device 130 may take sensed information during user action from one or more instructions to derive alterations to one or more prescribed task instructions, known user reactions, or effective communication tactics. As such, the computing device 130 may continuously learn and evolve an understanding of the user, effective instructions, and / or efficient communication modes.

[0050] FIG. 5 conveys a perspective view of portions of a sensor assembly 500 that may be employed in a training system in various embodiments. The sensor assembly 500 may be constructed of any number, and type, of materials to be worn by a user during assorted activities, such as motion and actions associated with completing a task. For instance, the sensor assembly 500 may have a body portion 510 that is structurally configured to fit around a designated aspect of a user, such as a head, arm, wrist, leg, waist, or chest. Various embodiments of the body portion 510 provide a continuous loop, halo (as shown), or unit that may be worn in more than one place on a user’s body depending on the adjustment characteristics of the body portion 510. Hence, it is contemplated that the body portion is adjustable for fit, comfort, and operation.

[0051] In accordance with some the body portion 510 provides a semi- rigid structure that supports a number of comfort features and electrical components. Such semi-rigid structure may be constructed of metal, polymers, or combinations thereof to provide a lightweight and flexible configuration that may be adapted to a variety of different user body sizes, such as head shape, chest size, and arm length while being sleek and comfortable. While not required, the body portion 510 may be lined with a comfort portion 512 that may be one or more materials that aid fit, comfort, and secure placement during user activities. For instance, the comfort portion 512 may be a fabric, foam, gel, or pneumatic structure that promotes secure placement of the body portion 510 despite user movement, sweating, external force, or encountered environmental conditions. It is contemplated that the body portion 510 may have one or more fitment features, such as additional bands, straps, and clasps, that may selectively add to the secure placement of the body portion 510

[0052] The body portion 510 may have a size and structural integrity conducive to the storage and accurate operation of a variety of different electronic components. Although not limiting, embodiments of the body portion 510 may enclose a power source, such as a battery or capacitor, that is electrically connected to assorted sensors 120 located around the body portion 510. Such sensors 120 may be configured to operate concurrently, sequentially, or randomly to collect information about the user, the environment around the user, and the user’s activity over time. The sensors 120 incorporated into the body portion 510, for example, may detect any variety of conditions, such as EEG, EKG, blood pressure, heart rate, brain activity, body temperature, air temperature, elevation, barometric pressure, wind speed, the location of the user’s appendages, and motion of the assorted aspects of a user over time.

[0053] It is noted that the various electrical components may be connected to a local processor, such as processor 210 of FIG. 2, as well as to one or more remote processors, such as processor 302 of FIG. 3. The ability to connect to local and / or remote processors via wired and / or wireless signal pathways allows for the seamless flow of sensed information from the assorted sensors to a computing device 130 configured to provide intelligent translation of sensed information into task determination, instruction generation,and communication mode selection. Some of the sensor assembly 500 operate autonomously or in response to user activity. However, other embodiments provide on- board controls for the user to manipulate the operation of aspects of the sensor assembly 500. For instance, user gestures, manual actions, or external voice prompts may activate, deactivate, or alter operation of one or more electronic components, which may allow for customized collection of data and energy usage.

[0054] While a single sensor assembly 500 may be employed to accurately detect user activity to allow intelligent generation of task-specific instructions and communication modes, some embodiments utilize multiple separate sensor assemblies worn on different portions of a user’s body to collect information that is input into the computing device 130 of FIG. 3. Regardless of the number, type, and location of sensor assemblies 500 worn by a user, information collected from the assorted sensors 120 may allow for intelligent task identification. FIG. 6 illustrates a flowchart for a task determination routine 600 that may be conducted with at least one sensor assembly 500 as part of a training system along with a computing device 130.

[0055] Although not limiting or required, the various aspects of the task determination routine 600 may be conducted with a variety of different sensors, such as optical, pressure, acoustic, mechanical, and measurement detectors, that are respectively connected to, and controlled by a computing device. In step 610, the computing device may activate one or more sensors worn by the user to detect any number, type, and location of conditions. That is, sensors may be activated concurrently, or sequentially, by the computing device to continuously, or sporadically, determine the state and condition of the user and the environment proximal the user. For instance, the activity of step 610 may involve activating a light detection and ranging (LiDAR) sensor to detect objects, surfaces, and people followed by confirmation of the presence, range, and depth of the detected aspects with one or more optical sensors, such as cameras positioned on a user’s head, chest, and / or hand.

[0056] The detection of conditions with assorted sensors provide the computing device with ample data for evaluation, in step 620, to determine if a known task, or activity, is being currently undertaken, imminent, or likely in the future. The presence of equipment,or tools, along with previous user actions, movements, may be taken into consideration by the computing device to determine if there is a task to be considered. If so, decision 630 translates the intelligently evaluated sensor information from step 620 into a determination of the presence of a previously known task. It is noted that a task, or activity, may be characterized as known if the computing device has instructions and / or milestones associated with a user beginning, conducting, and completing the task.

[0057] The identification of a known task, or series of tasks that comprise an activity, advances to step 640 where the presence of any equipment and / or tools are evaluated. However, in the event user actions and conditions indicate a task that is not known by the computing device from decision 630, step 632 proceeds to relate the currently sensed conditions to other, known, tasks in an attempt to choose a default series of instructions that may be customized for the current task at hand. It is contemplated that step 632 may result in no similar task being found, which may prompt the computing device to randomly choose a task as a default template or simply advance without default instructions.

[0058] Regardless of what default task instructions that are chosen from step 632, the known aspects of the user are incorporated to customize the default instructions in step 634. Such customization may be conducted over time, in response to user behavior, along with previously known user characteristics, such as tendencies, reactions, and capabilities, to produce instructions that are most likely to produce accurate and efficient execution of portions of a task, such as sub-steps defined between operational milestones. It is noted that the customization of instructions for the user and / or environment of the task are not restricted to unknown tasks from decision 630 and known tasks may have one or more instructions altered to optimize efficiency, accuracy, and / or safety.

[0059] Once the initial set of task instructions are set from decision 630, or step 634, step 640 determines if one or more objects, such as tools or equipment, may enhance the efficiency and / or accuracy of task completion. The determination of available objects in step 640 may differ from the evaluation of conditions in step 620 in that objects detected in step 620 may be actually present while objects in step 640 are contemplated as present. That is, step 620 determines if equipment and / or tools are currently present while step 640 evaluates if any possible equipment and / or tools may optimize the execution of portions ofa task. If equipment and / or tools could in the performance of a task, step 640 may request a user to obtain such. Conversely, if a present object is not necessary for optimal task execution, step 640 may request removal of an object to avoid distraction or confusion.

[0060] Next, step 650 utilizes the initial task completion instructions from decision 630 or step 634, along with the availability of equipment / tools, to generate specific user instructions to achieve the next predetermined task milestone. The instruction generation of step 650 may involve any amount of real-time sensor data that may alter preexisting instructions to better cater the instructions to the task, and user. In accordance with some embodiments, step 650 translates the initial task-specific instructions into sub-steps associated with completion of the next milestone. Hence, step 650 may truncate and / or customize instructions to guide a user to complete less than all of a task, such as the actions and movements necessary to achieve a predetermined milestone event.

[0061] In response to delivery and execution of instructions from step 650 by a user, step 660 monitors subsequent user behavior, such as movement, verbal reaction, actions, and gestures. Through the monitoring of user behavior in step 660, a computing device may consistently learn and adapt to the user and the environment of the task. That is, decision 670 may react to a user’s behavior in response to instructions delivered in step 650 by evaluating if the instructions resulted in intended user actions. If so, no change to an instruction is necessary and step 660 is revisited as the user achieves the next task milestone. In the event a user’s reaction to instructions detected in step 660, and evaluated in decision 670, is not conducive to efficient and accurate task execution, step 672 then alters at least one aspect of an instruction before sending the revised instruction to the user. It is contemplated that the alteration of an instruction in step 672 involves changing the communication mode in which the instruction is conveyed, such as textual to graphical or video to graphical, as illustrated in FIG. 2.

[0062] FIG. 7 is a flowchart of an example instruction routine 700 that may be carried out by a computing device and sensor assembly as part of a training system in accordance with various embodiments. The instruction routine 700 may be conducted at any time and with any aspect of a training system, such as in conjunction with the task determination routine 600 of FIG. 6. Initially, step 710 generates one or more instructions, based oninformation detected from system to guide a system user to complete at least portions of a task. The generation of instructions in step 710 may coincide with various task determination evaluations and decisions along with assorted evaluations about the user as well as the environment in which a task may be executed.

[0063] An understanding about the task, user, and environment allows a computing device, such as device 130 of FIG. 4, to create instructions catered for the capabilities and tendencies of the user along with the available objects and aspects of the user’s proximal environment. Such customization of information may extend, in step 720 to how instructions and system information is conveyed to a user, which may generally be characterized as a communication strategy.

[0064] Although not required or limiting, a communication strategy generated in step 720 may provide multiple different manners of conveying information to a user along with intelligence in what manner the computing device is to select. That is, the communication strategy may set forth both different ways information may be communicated as well as a progression of communication manners catered to the user’s preferences, tendencies, and / or behavior. It is contemplated, in some embodiments, that the communication strategy prescribes communication alterations in response to specific user behavior or events. For instance, a communication strategy may direct a change, or supplement, to how instructions and / or information is conveyed to a user in response to detected changes in the environment, the user performing instructions quickly, or the user failing to reach milestones within a predetermined timeframe.

[0065] With the task completion instructions and communication strategy respectively generated, which may not coincide with the user performing any task actions, step 730 may proceed to selectively, or concurrently, activate system sensors, such as sensors worn by the user, to detect real-time conditions associated with the user. The sensing of real-time conditions in step 730 may allow a system computing device to select the most efficient and / or accurate instructions and communication manner to begin task completion. In accordance with some embodiments, the sensing of real-time conditions in step 730 may provide information about the user and / or environment that prompts for therevision of the task completion in step 710 and / or the communication strategy generated in step 720.

[0066] Specifically, decision 740 evaluates the information from step 730 compared to the instructions generated in step 710 to determine if an alteration to one or more instructions may allow for an optimal execution of aspects of a task. If a task instruction may be revised in decision 740 to improve the efficiency, accuracy, or safety, for example, of the execution of a portion of a task, step 742 changes an existing instruction from step 710 in response to the real-time conditions sensed in step 730. It is noted that the alteration of an instruction in step 742 may involve adding clarity to an instruction, replacing an instruction, and / or replacing the entirety of an instruction. Similarly, a determination that an improvement to the way an instruction, or sensed information, is communicated to the user from decision 750 prompts step 752 to alter, supplement, or replace a portion of the communication strategy generated in step 720 to improve the efficiency and / or accuracy of information delivery.

[0067] By evaluating, and potentially adapting, the instructions and manner of communicating the instructions, steps 742 and 752 may adapt to real-time user and environmental conditions to provide the optimal content in the most efficient manner to the user in step 760. As the user receives the task instructions in the selected communication manner, step 770 monitors user action and compares such action to predetermined task milestones. That is, step 770 monitors how the user reacts to the instructions delivered in step 760 to determine if an established task completion milestone is likely to be met. If the instructions are being understood and promptly followed in step 770 to likely reach one or more predetermined milestones, additional instructions of the set generated in step 710 are revisited in view of real-time user and environmental conditions by returning to step 730

[0068] In the event the instructions provided to the user in step 760 are not effective at producing user actions that are likely to reach a predetermined task milestone in a set timeframe, decision 740 is revisited to determine if a revision to the prior instruction, and / or an altered communication manner from decision 750, may improve the efficiency, accuracy, quickness, and / or safety of performing actions and movements conducive with accomplishing an established task milestone. Through the establishment of initialinstructions and a communication followed by potential adaptation of the instructions and / or communication manner in response to real-time user and environmental conditions, a training system can maintain optimal interaction and guidance despite changing conditions as well as inefficient instructions and / or ineffective manners of communication.

[0069] FIG. 8 illustrates a learning routine 800 that may be conducted by a computing device, such as a learning module 330 of FIG. 3, in various embodiments. The assorted aspects of the learning routine 800 may be performed at any time, and in conjunction with, the routines 600 / 700 of FIGS. 6 and 7. Execution of the learning routine 800 may allow a training system to utilize intelligence and analysis to improve the generation of future task completion instructions and communication strategies. In response to an encountered condition, step 810 identifies that some aspect of the condition is unknown.

[0070] That is, a computing device in step 810 recognizes from one or more measurements from at least one sensor that an unknown situation is present. An unknown situation may be characterized as user position, action, movement, or gesture that is not recognized to correspond, initially, with an environment proximal to the user. Another, non- limiting example of an unknown condition is the presence of equipment and / or tools that are foreign to the computing device or any known tasks.

[0071] Regardless of the type and extent of the unknown condition identified by the training system in step 810, step 820 attempts to correlate the various detected aspects of a user and the user’s environment with a similar, known condition. For instance, the decision 820 may compare aspects of a sensed user, and / or environmental, condition, such as a user position, available equipment, and current user behavior, to one or more existing tasks, conditions, or activities in an attempt to find a suitable template to build a set of instructions, and communications strategy, efficiently and accurately. In other words, the decision 820 is not evaluating the unknown condition compared to known tasks to find a set of instructions, or communication strategy, that may be employed, but instead attempting to correlate the unknown condition with a suitable starting set of instructions that may be efficiently adapted to provide a predetermined set of instructions for the next time the condition of step 810 occurs.

[0072] It is noted that the criteria what designates a suitable initial template of instructions to be deemed a correlation may differ between tasks, activities, and available computing resources. Hence, if no efficient correlation exists in decision 820, routine may proceed with step 822 choosing, and assigning, a default task profile to the current, unknown condition from step 810. The default profile may have any number, type, and timing of instructions collectively directed to guiding a user through the execution of movements, actions, and gestures that perform the intended task, in the aggregate. With the default task profile chosen in step 822, a set of separate user instructions, communication strategy, and milestones may be present that provide a framework of actions, sensed aspects, and movement of a user that is necessary to accomplish a task.

[0073] While deviations from the default instructions, or communications strategy, may be implemented to accommodate the unknown condition from step 810, various embodiments conduct one or more test instruction in step 824 that supplements the instructions of the default profile to detect assorted aspects of the user. For instance, a training system computing device may generate any number, and type, of instructions, such as reach as far as possible, pinch fingers, or tilt a head, which are delivered to, and executed by, a user to allow the body worn sensors to detect assorted tendencies, reactions, and capabilities of the user. It is contemplated that the sensed aspects from step 826 may gauge the mental and / or physical potential of a user so that the computing device may assign behavioral characteristics to the default profile in step 828.

[0074] Through the adaptation of the default profile with respect to sensed aspects of a user, an unknown condition may have a relative accuracy and efficiency for performing a task, or sub-steps of a task marked by milestones. In contrast, applying a default profile to the unknown condition from step 810 without customization to detected aspects of a user may result in inefficiencies, errors, and safety risks due to providing instructions to a user that are outside the user’s capabilities or are not mentally processed efficiently.

[0075] In the event the characteristics of the unknown condition from step 810 are correlated to an actual task, activity, milestone, or user action, step 832 proceeds to evaluate the unknown condition in an effort to aid in future assignment of task completion instructions and a communication strategy without the evaluation of the correlation decision820. The characterization of the unknown in step 832 is not limited to a particular analysis or set of assigned characteristics, but may involve logging and associating a variety of different sensor readings. Such assigned characteristics are then utilized in step 834 to adapt one or more known task profiles, such as instructions and communications strategies, to accommodate the assorted aspects of the unknown condition.

[0076] As a non-limiting example, steps 832 and 834 may sequentially identify the availability of an unknown tool, piece of equipment, or user behavior before adapting the instructions for a particular task to utilize the newly identified aspects. At the conclusion of the adapting of a default task completion profile in step 828 or the incorporation of unknown condition characteristics to a known task completion profile in step 834, step 840 selects the best possible communication mode and task completion instruction to advance the unknown condition of step 810 to the next milestone.

[0077] Through the building of a task completion profile from decision 820 that is customized to the user, environment, and task to be completed, step 840 may intelligently select a communication mode that provides the best possible chance of efficient user understanding. The selected communication mode from step 840 may be employed to deliver any number, and type, of task completion instructions over time, which may accomplish one or more predetermined milestones. In some embodiments, the communication mode changes in response to detected user, or environmental conditions. Similarly, step 850 may adapt task completion instructions based on detected conditions and the user profile developed from decision 820 to cater and optimize how the training system utilizes sensors worn by the user to provide training interactions to elicit accurate and safe user execution of a task.

[0078] It is noted that the generation of test instructions and subsequent evaluation of sensed user reactions to the test instructions may be employed in combination with steps 832 and 834. The ability to selectively deliver user instructions directed at identifying the real-time behavior and capabilities of the user, in step 824 or after step 832, allows a training system to collect accurate information about the user and the environment around the user, which may translate into optimized instruction and / or communication mode modifications that are the most likely to result in efficient, accurate, and safe taskcompletion by the user. With the learn about the user, task, and environment both passively, via monitoring user activity, and actively, via monitoring the execution of specific, non-task generated user instructions.

[0079] In accordance with various embodiments, a training system may intelligently utilize a variety of sensors worn by a user to instruct the user how to conduct a vast array of tasks. The utilization of sensors worn by the user may allow the training system to recognize what task a user is about to perform and intelligently generate instructions, sub- steps, and operational milestones directed to guide the user through the completion of the task. Such guidance allows a user to have no knowledge and experience to conduct a task sufficiently or to optimize the efficiency, accuracy, and safety of task execution by a user with previous task experience. The presence of various sensors worn by a user further allows for real-time adaptations to task completion instructions as well as user activity communication modes to provide maximum efficiency for user understanding of what has been conducted and what is to be conducted in the future.

[0080] Various embodiments of a headband provide a shape and structure that is ergonomically designed to comfortably fit around the head while housing various sensors for biomechanical and physiological monitoring. The headband may be constructed of lightweight, durable materials, such as flexible polymers and / or silicone, to ensure comfort and flexibility during use. A headband, in some embodiments, has integrated sensors that detect assorted aspects of an athlete’s movement, and condition, over time. For instance, a headband may have biomechanical sensors to measure body movements, including head position, arm and hand orientation, finger spacing, and shot mechanics along with physiological sensors, such as photoplethysmogram (PPG) sensors, to measure heart rate and variability, respiratory sensors to track breath rate, and accelerometers for measuring movement and fatigue levels. A headband may employ neurofeedback sensors that monitor neurological states, including focus and mental clarity, using EEG-like measurements or similar technologies.

[0081] A headband may integrate with one or more software applications, or platforms, that use AI to analyze the collected data. Machine learning algorithms may be employed to process collected data in real time to provide insights on areas ofimprovement, such as posture, shot mechanics. A voice assistant may provide real-time feedback to guide athletes during practice sessions as well as offering personalized suggestions on technique adjustments. The headband may have a system that generates a "focus score" to reflect an athlete's mental clarity during practice or gameplay. Such focus score may be influenced by physiological factors, such as heart rate and breathing, as well as sleep quality and neurological state. It is contemplated that the headband, along with corresponding executed software, provides actionable insights into how to improve focus and performance by adjusting physical and mental conditions.

[0082] Embodiments of this disclosure present a revolutionary wearable headband that not only enhances athletic performance through biomechanical data collection but also uses AI-driven feedback to improve physical and mental readiness. A smart headband may bridge the gap between technology and athletic training by offering a personalized approach to skill development and performance optimization.

[0083] Through various embodiments, a smart headband may have a headband body and a plurality of sensors with the headband body having a size and shape to fit a user and the plurality of sensors positioned in the headband body. The plurality of sensors may measure biomechanical data of the user with the biomechanical data including body movement, finger spacing, and shot mechanics of the user. The plurality of sensors of the smart headband may measure physiological data of the user, such as heart rate, breath, and fatigue levels of the user. The plurality of sensors of the smart headband may track neurological capabilities of the athlete, such as focus and mental clarity of the user.

[0084] The plurality of sensors may be connected to a processor that executes a machine learning algorithm to process data from the plurality of sensors in real-time to provide personalized feedback to the user. The smart headband may have a feedback system that delivers audio or visual instructions to the user to optimize technique, improve mental clarity, and optimize physical effort. Such a feedback system may generate a focus score that correlates the user’s mental state with a performance metric, such as shooting, hitting, or otherwise scoring. The smart headband may have an audio / visual system that displays data from the plurality of sensors, such as real-time analytics, feedback, insights, and progress tracking to the user.

[0085] In accordance with some an athlete’s shooting performance may be trained with a headband body having a size and shape to fit an athlete. A plurality of sensors may be incorporated in the headband body and arranged to measure biomechanical data of the athlete, such as body movement, finger spacing, and shot mechanics of the athlete. The plurality of sensors may measure physiological data of the athlete, such as heart rate, breath, and fatigue levels of the athlete. The plurality of sensors may track neurological capabilities of the athlete, such as focus and mental clarity of the athlete. Data from the plurality of sensors may be analyzed with a machine learning algorithm in real-time to provide personalized feedback to the athlete. A feedback system of the headband body may generate a focus score that correlates the athlete’s mental state with a shooting performance prior to delivering instructions to the athlete with the feedback system to optimize technique, improve mental clarity, and optimize physical effort.

Claims

What is claimed is:

1. A method comprising: connecting a computing device to an array of sensors positioned proximal to a user; determining, with the computing device, a task to be completed by the user in response to information detected by the array of sensors; generating, with the computing device, a set of instructions directing the user to complete the task; generating, with the computing device, a communication strategy in response to the information detected by the array of sensors, the communication strategy prescribing different communication modes associated with different user actions during the completion of the task; sensing, with the array of sensors, real-time user conditions during the completion of the task; sensing, with the array of sensors, real-time environmental conditions during the completion of the task; determining, with the computing device, that the user will fail to complete a predetermined milestone of the task based on the real-time user conditions and the real-time environmental conditions; and adapting, with the computing device, at least one instruction of the set of instructions in response to the failure of the user to complete the predetermined milestone, the adaptation of the at least one instruction changing how the user is taught to perform a sub-step of the task.

2. The method of claim 1, wherein the array of sensors comprises at least one stationary sensor and at least one sensor positioned on the athlete.

3. The method of claim 1, wherein the plurality of different communication modes comprises audible, visual, and haptic cues.

4. The method of claim 3, wherein the communication strategy prescribes multiple manners of concurrently communicating training information to the user.

5. The method of claim 4, the communication strategy prescribes multiple different manners of conveying training information.

6. The method of claim 1, wherein the computing device generates a progression of milestones corresponding to the set of instructions to complete the task in response to the determination of the task.

7. The method of claim 6, wherein the progression of milestones corresponds with a series of sub-steps to complete the task.

8. The method of claim 6, wherein the computing device determines an availability of equipment for the user to generate the progression of milestones.

9. The method of claim 6, wherein the computing device detects user reaction to the instructions with the array of sensors.

10. The method of claim 9, wherein the computing device adapts the at least one instruction in response to the detected user reaction.

11. The method of claim 1, wherein the computing device adapts the at least one instruction in response to the sensed real-time environmental conditions.

12. The method of claim 1, wherein the computing device alters a communication mode prescribed by the communication strategy in response to sensed real- time user conditions.

13. The method of claim 1, wherein the computing device alters communication modes based on generated user profile.

14. The method of claim 13, wherein the user profile comprises user tendencies determined in response to real-time user conditions sensed by the array of sensors.

15. The method of claim 13, the user profile comprises user capabilities determined in response to real-time user conditions sensed by the array of sensors.

16. A training system comprising: a computing device; a first sensing assembly worn by a user; wherein the computing device comprises a processor and non-volatile memory; wherein the first sensing assembly houses an array of sensors respectively connected to the computing device; wherein the first sensing assembly is operable to provide the computing device with detected information about the user and an environment proximal the user; wherein a task module of the computing device comprises circuitry operable to identify a task to be completed by the user in response to information detected by the array of sensors; wherein an instruction module of the computing device comprises circuitry operable to generate a set of instructions directing the user to complete the task; wherein a learning module of the computing device comprises circuitry operable to generate a communication strategy in response to the information detected by the array of sensors, the communication strategy prescribing different communication modes associated with different user actions during the completion of the task; and wherein the computing device comprises circuitry operable to adapt the set of instructions in response to real-time information sensed by the array of sensors indicating the user will fail to complete a predetermined milestone of the task, the adaptation of the set of instructions changing how the user is taught to perform a sub-step of the task.

17. The training system of claim 16, wherein a video module of the computing device comprises circuitry operable to combine separate video feeds from the array of sensors into a single, compiled video.

18. The training system of wherein the array of sensors is operable to measure biomechanical aspects of the user.

19. The training system of claim 16, wherein the array of sensors is operable to measure physiological aspects of the user.

20. The training system of claim 16, wherein the array of sensors is operable to measure neurological aspects of the user.