Control method for controlling wearable device on basis of exercise data of user and electronic device for porforming same

Abstract

Disclosed are a control method for controlling a wearable device on the basis of exercise data of a user, and an electronic device for performing same. The control method includes the steps of: receiving a user input about an exercise goal; determining a torque parameter, associated with achieving the exercise goal, using a neural network model that uses feature data, associated with an exercise performed by a user wearing a wearable device, as input, and outputs an estimated exercise performance indicator; and transmitting a control signal including the determined torque parameter to the wearable device so that the wearable device generates torque on the basis of the determined torque parameter.

Description

Control method for controlling a wearable device based on a user's exercise data and an electronic device performing the same

[0001] Certain embodiments relate to a control method for controlling a wearable device based on a user's exercise data and an electronic device performing the same.

[0002] In general, a walking assistance device is a device or apparatus that helps patients who cannot walk on their own due to various diseases or accidents to perform walking exercises for rehabilitation treatment, and / or a device or apparatus that can be used for exercise. Recently, as the aging society deepens, the number of people who have difficulty walking normally or complain of discomfort when walking due to leg joint problems is increasing, and interest in walking assistance devices is also increasing. Walking assistance devices are attached to the user's body and can assist the user's muscle strength required for walking, for example, and guide the user's walking so that the user can walk with a normal walking pattern. These walking assistance devices can also perform functions that assist the user with various leg exercises (e.g., power walking, jogging, stair climbing, lunges, stretching).

[0003] A control method for controlling a wearable device based on a user's exercise data according to an exemplary aspect may include an operation of receiving a user input for an exercise goal, an operation of determining a torque parameter related to achievement of the exercise goal using a neural network model that inputs feature data related to an exercise performed by a user while wearing the wearable device and outputs an estimated exercise performance index, and an operation of transmitting a control signal including the determined torque parameter to the wearable device so that the wearable device generates a torque based on the determined torque parameter.

[0004] An electronic device according to an exemplary aspect may include one or more processors, an input circuit for receiving user input, and a communication circuit for communicating with a wearable device. In response to receiving a user input regarding an exercise goal through the input circuit, the one or more processors may determine a torque parameter associated with achievement of the exercise goal using a neural network model that inputs feature data related to an exercise performed by a user while wearing the wearable device and outputs an estimated exercise performance index. The one or more processors may control the wearable device to generate a torque based on the determined torque parameter by causing the communication circuit to transmit a control signal including the determined torque parameter to the wearable device.

[0005] These and / or other aspects, features and advantages will become apparent and more readily understood from the following description of exemplary embodiments taken in conjunction with the accompanying drawings.

[0006] FIG. 1 is a drawing for explaining an overview of a wearable device worn on a user's body according to various embodiments.

[0007] FIG. 2 is a drawing for explaining an exercise assistance system according to various embodiments.

[0008] FIG. 3 illustrates a rear schematic diagram of a wearable device according to various embodiments.

[0009] FIG. 4 illustrates a left side view of a wearable device according to various embodiments.

[0010] FIG. 5 is a diagram illustrating configurations of a wearable device according to various embodiments.

[0011] FIG. 6 is a diagram illustrating interaction between a wearable device and an electronic device according to various embodiments.

[0012] FIG. 7 is a diagram illustrating configurations of an electronic device according to various embodiments.

[0013] FIG. 8 is a diagram illustrating a system for controlling a wearable device based on a user's exercise data according to various embodiments.

[0014] FIGS. 9 and 10 are diagrams for explaining input / output data and learning operations of a neural network model according to various embodiments.

[0015] FIG. 11 is a flowchart illustrating operations of a control method for controlling a wearable device based on user exercise data according to various embodiments.

[0016] FIG. 12 is a diagram illustrating an operation of determining a torque parameter related to the achievement of an exercise goal using a neural network model according to various embodiments.

[0017] FIG. 13 is a flowchart illustrating operations of a control method for controlling a wearable device based on user exercise data according to various embodiments.

[0018] FIGS. 14A, 14B, and 14C are drawings illustrating a cable-based wearable device and determining a torque parameter for a cable-based wearable device according to various embodiments.

[0019] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Therefore, the actual implementation is not limited to the specific embodiments disclosed, and the scope of this specification includes modifications, equivalents, or alternatives within the technical concepts described in the embodiments.

[0020] Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "has" should be understood to indicate the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0021] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art. Terms defined in commonly used dictionaries should be interpreted to have a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0022] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted.

[0023]

[0024] FIG. 1 is a drawing for explaining an overview of a wearable device worn on a user's body according to various embodiments.

[0025] Referring to FIG. 1, in one embodiment, a wearable device (100) may be a device worn on a user's (110) body to assist the user's (110) walking, exercising, and / or working. The wearable device (100) may also be used to measure the user's (110) physical ability (e.g., walking ability, exercise ability, exercise posture). In embodiments, the term 'wearable device' may be replaced with 'wearable robot', 'walking assistance device', or 'exercise assistance device'. The user (110) may be a person who wears the wearable device (100) and walks, exercises, or works.

[0026] A wearable device (100) may be worn on a user's (110) body (e.g., lower body (legs, ankles, knees, etc.) and / or upper body (torso, arms, wrists, etc.)) to apply external forces, such as assistance force and / or resistance force, to the body movements of the user (110). Assistance force refers to a force applied in the same direction as the body movement direction of the user (110), and represents a force that assists the body movements of the user (110). Resistance force refers to a force applied in the opposite direction to the body movement direction of the user (110), and represents a force that hinders the body movements of the user (110). The term 'resistance force' may also be referred to as 'exercise load'.

[0027] In one embodiment, the wearable device (100) may operate in a walking assistance mode to assist the walking of the user (110). In the walking assistance mode, the wearable device (100) may assist the walking of the user (110) by applying an assistive force generated from the driving module (120) of the wearable device (100) to the body of the user (110). The wearable device (100) may assist the force required for the walking of the user (110), thereby enabling the user (110) to walk independently or to walk for a long time, thereby expanding the walking ability of the user (110). The wearable device (100) may also help improve the walking of a user with abnormal walking habits or walking posture.

[0028] In one embodiment, the wearable device (100) may operate in an exercise assistance mode to enhance the exercise effect of the user (110) or to provide various exercise experiences to the user (110). The exercise assistance mode may include a resistance mode and an assistance mode. The resistance mode of the exercise assistance mode refers to a mode that impedes the body movement of the user (110) or provides resistance to the body movement of the user (110) by applying a resistance force generated from the driving module (120) to the body of the user (110). If the wearable device (100) is a hip-type wearable device worn on the waist (or pelvis) and legs (e.g., thighs) of the user (110), the wearable device (100) may provide an exercise load to the leg movement of the user (110) while being worn on the legs in the resistance mode, thereby further enhancing the exercise effect on the legs of the user (110). The assist mode of the exercise assistance mode refers to a mode in which an assistive force is applied to the body of the user (110) to assist the body movement of the user (110). In the assist mode, an assistive force, which is a force in the same direction as the body movement, is provided to the user (110). For example, when a disabled person or an elderly person wears a wearable device (100) and exercises, the wearable device (100) may provide an assistive force to assist the body movement. In the assistive mode, the wearable device (100) may provide a force in the same direction as the leg movement direction of the user (110), and the user (110) may perform an exercise with less force through the force provided from the wearable device (100). In an exercise program performed using the wearable device (100), the resistance mode and the assistive mode may be operated in combination. For example, the wearable device (100) may provide an assistive force and a resistance force in combination for each exercise section or time section, such as providing an assistive force in some exercise sections and a resistance force in other exercise sections.In the exercise assistance mode, various exercise programs can be operated according to the exercise purpose and / or the physical ability of the user (110). The exercise program is exercise content that the user (110) performs using the wearable device (100), and may include, for example, aerobic exercise, strength training, postural balancing exercise, or any combination thereof. The type of exercise program is not limited thereto and may vary. Depending on the exercise program performed by the wearable device (100), the resistance mode and the assistance mode may be appropriately operated in an alternating manner, and a target exercise speed that matches the appropriate physical condition (e.g., heart rate) of the user (110) while performing the exercise may be guided to the user.

[0029] In one embodiment, the wearable device (100) may operate in a physical ability measurement mode for measuring the physical ability of a user (110). The wearable device (100) may measure movement information of the user (110) using a sensor (e.g., an angle sensor (125)) or an inertial measurement unit (IMU) (135)) provided in the wearable device (100) while the user (110) walks and / or exercises, and may evaluate the physical ability of the user (110) based on the measured movement information. For example, the gait index (e.g., number of steps, total walking distance, stride) or the exercise ability index (e.g., muscle strength, exercise endurance, postural balance) of the user (110) may be estimated through the movement information of the user (110) measured by the wearable device (100).

[0030] In a specific embodiment, for convenience of explanation, a hip-type wearable device (100) as illustrated in FIG. 1 is used as an example, but is not limited thereto. As described above, the wearable device (100) may also be worn on other body parts (e.g., upper arms, lower arms, hands, calves, or feet) other than the waist and thighs. The shape and configuration of the wearable device (100) may vary depending on the body part on which it is worn.

[0031] The wearable device (100) may include a support frame (e.g., a waist support frame (20) of FIGS. 3 and 4) for supporting the body of the user (110) when the wearable device (100) is worn on the body of the user (110), a drive module (120) for generating a torque applied to the legs of the user (110) (e.g., a first drive module (45) and a second drive module (35) of FIG. 3), a torque transmission frame for transmitting the torque generated by the drive module (120) to the legs of the user (110) (e.g., a first torque transmission frame (55) and a second torque transmission frame (50) of FIG. 3), a sensor circuit including one or more sensors for obtaining sensor data including movement information on the body movement of the user (110) (e.g., leg movement, upper body movement), and a control circuit (130) for controlling the operation of the wearable device (100) (e.g., a control circuit (510) of FIG. 5). there is.

[0032] In one embodiment, the wearable device (100) may include an angle sensor (125) and an inertial sensor (135). The angle sensor (125) may measure a rotation angle of a torque transmission frame of the wearable device (100) corresponding to a hip joint angle of the user (110). The angle sensor (125) may include, for example, an encoder and / or a hall sensor. In one embodiment, the angle sensor (125) may be positioned near a motor included in the drive module (120) connected to the torque transmission frame. The inertial sensor (135) may include an acceleration sensor and / or an angular velocity sensor, and may measure changes in acceleration and / or angular velocity according to movements of the user (110). The inertial sensor (135) can measure, for example, a movement value of a waist support frame (e.g., waist support frame (20) of FIG. 3) or a base body (e.g., base body (80) of FIG. 3) of a wearable device (100). The movement value of the waist support frame or base body measured by the inertial sensor (135) can correspond to a waist movement value (or upper body movement value) of a user (110).

[0033] In one embodiment, the control circuit (130) and the inertial sensor (135) may be placed within a base body of the wearable device (100) (e.g., the base body (80) of FIG. 3). The base body may be positioned at the waist area of ​​the user (110) while the user (110) is wearing the wearable device (100). The base body may be formed or attached to the outside of the waist support frame of the wearable device (100). The base body may support the lumbar region of the user (110).

[0034]

[0035] FIG. 2 is a drawing for explaining an exercise assistance system according to various embodiments.

[0036] Referring to FIG. 2, the exercise assistance system (200) may include a wearable device (100), an electronic device (210), another wearable device (220), and a server (230). In the exercise assistance system (200), at least one of the devices other than the wearable device (100) (e.g., the electronic device (210), another wearable device (220), or the server (230)) may be omitted, or one or more other devices (e.g., a dedicated controller device for the wearable device (100)) may be added.

[0037] In one embodiment, the wearable device (100) may be worn on the user's body in a walking assistance mode to assist the user's movements. For example, the wearable device (100) may be worn on the user's leg to generate an assistive force to assist the user's leg movements, thereby assisting the user's walking.

[0038] In one embodiment, the wearable device (100) may generate and apply to the user's body a resistance force to hinder the user's body movement and / or an assistive force to assist the user's body movement in order to enhance the user's exercise effect in the exercise assistance mode. In the exercise assistance mode, the user may select an exercise program (e.g., aerobic exercise such as power walking and outdoor walking, strength training such as squats, split lunges, dumbbell squats, and lunge and knee ups, stretching, postural balancing exercise, or any combination thereof) and / or an exercise intensity to be applied to the exercise program via the electronic device (210). The wearable device (100) may control a driving module (e.g., a driving module (120) of FIG. 1) of the wearable device (100) according to the exercise program and / or exercise intensity selected by the user. For example, the wearable device (100) can adjust the strength of the resistance and / or assist force generated by the drive module according to the exercise intensity selected by the user. The wearable device (100) can control the drive module to generate a resistance force corresponding to the exercise intensity selected by the user. As the exercise intensity increases, the magnitude of the resistance force applied to the user can also increase.

[0039] The wearable device (100) can transmit sensor data measured through a sensor (e.g., an angle sensor (125) or an inertial sensor (135) of FIG. 1) to an electronic device (210) and receive a control signal for controlling the operation of the wearable device (100) from the electronic device (210).

[0040] The electronic device (210) can communicate with the wearable device (100) via wireless communication (e.g., Bluetooth communication) or wired communication, and can remotely control the wearable device (100) or provide the user with status information regarding the status of the wearable device (100) (e.g., booting status, charging status, exercise program operation status, error status). The electronic device (210) can recommend an exercise program using the wearable device (100) to the user and analyze the exercise performed by the user. The electronic device (210) can receive sensor data acquired by a sensor of the wearable device (100) from the wearable device (100), and can estimate the user's current exercise status, exercise result, exercise posture, and / or physical ability based on the received sensor data. The electronic device (210) can provide the user with the estimated current exercise status, exercise result, exercise posture, and / or physical ability of the user through a graphical user interface (GUI).

[0041] In one embodiment, a user may execute a program (e.g., an application) on an electronic device (210) to control a wearable device (100), and the user may adjust the operation or setting values ​​(e.g., the torque intensity output from the motor of the drive module, the volume of audio output from an audio output circuit (e.g., the audio output circuit (550) of FIG. 5), the brightness of a lighting unit (e.g., the lighting unit (85) of FIG. 3)) of the wearable device (100) through the program. The program executed on the electronic device (210) may provide a graphical user interface for interaction with the user. The electronic device (210) may be a variety of devices. For example, the electronic device (210) may include, but is not limited to, a portable communication device (e.g., a smartphone), a computer device, a portable multimedia device, or a home appliance device (e.g., a television, an audio device, a projector device).

[0042] According to one embodiment, the electronic device (210) may be connected to the server (230) using short-range wireless communication or cellular communication. The server (230) may receive user profile information of a user using the wearable device (100) from the electronic device (210), and store and manage the received user profile information. The user profile information may include, for example, information on at least one of name, age, gender, height, weight, medical history, or body mass index (BMI). The server (230) may receive exercise history information regarding exercise performed by the user from the electronic device (210), and store and manage the received exercise history information. The server (230) may provide the electronic device (210) with various exercise programs or physical ability measurement programs that may be provided to the user. In one embodiment, the server (230) may be connected to the wearable device (100). The server (230) can receive sensor data measured by the wearable device (100) from the wearable device (100) and transmit control signals and / or exercise program-related data for controlling the operation of the wearable device (100) to the wearable device (100). In one embodiment, the server (230) can be a cloud server.

[0043] According to one embodiment, the wearable device (100) and / or the electronic device (210) may be connected to another wearable device (220). The user's exercise result information, physical ability information, and / or exercise motion evaluation information determined by the electronic device (210) may be transmitted to the other wearable device (220) and provided to the user through the other wearable device (220). Status information of the wearable device (100) may also be transmitted to the other wearable device (220) and provided to the user through the other wearable device (220). In one embodiment, the wearable device (100), the electronic device (210), and the other wearable device (220) may be connected to each other through wireless communication (e.g., Bluetooth communication, Wi-Fi communication). Other wearable devices (220) may be, for example, wireless earphones (222), a smartwatch (or a wearable device in the form of a watch) (224), or smartglasses (a wearable device in the form of glasses or goggles) (226), but are not limited to the aforementioned devices.

[0044] In one embodiment, the wireless earphones (222) may be wirelessly connected to the electronic device (210) and / or the wearable device (100) to output guide voices, music, and / or sound effects related to an exercise program. The wireless earphones (222) may provide the user with information related to the exercise program (e.g., an introduction to the exercise program, remaining exercise time) or may inquire about the user's selection through the guide voices. The wireless earphones (222) may include a microphone, and the microphone may receive a user's voice input. The voice input received through the microphone may be transmitted to the electronic device (210), and voice recognition may be performed on the voice input in the electronic device (210).

[0045] In one embodiment, the smartwatch (224) may include a biosensor (e.g., a heart rate sensor, an electromyography sensor) that measures a biosignal including heart rate information of the user, and may transmit the biosignal measured by the biosensor to the electronic device (210) and / or the wearable device (100). The electronic device (210) may estimate the heart rate information (e.g., current heart rate, maximum heart rate, average heart rate) and / or electromyography information of the user based on the biosignal received from the smartwatch (224), and may provide the estimated heart rate information and / or electromyography information to the user. The heart rate information and / or electromyography information may be used to determine the haptic intensity of the haptic feedback provided through the wearable device (100).

[0046] In one embodiment, the smartwatch (224) may include an inertial sensor for measuring user movement information and / or a position sensor for measuring user location information, and may transmit the user movement information and / or location information to the electronic device (210) and / or the wearable device (100). The smartwatch (224) may include a communication circuit (e.g., a short-range communication circuit) for communicating with another device (e.g., the electronic device (210), the wearable device (100)). In one embodiment, the smartwatch (224) may provide an exercise program related interface through a display. The exercise program related interface may be implemented through a separate application installed on the smartwatch (224). The user may also control the wearable device (100) through the smartwatch (224).

[0047] In one embodiment, the smart glasses (226) can provide information to the user through a glass-shaped display. For example, in exercise mode, the smart glasses (226) can output information such as current exercise speed, target exercise speed, current exercise volume achieved, exercise time, and biometric information through the display. Additionally, the smart glasses (226) can output a screen to guide the user on their exercise route.

[0048]

[0049] FIG. 3 illustrates a rear schematic diagram of a wearable device according to various embodiments. FIG. 4 illustrates a left side view of a wearable device according to various embodiments.

[0050] Referring to FIGS. 3 and 4, a wearable device (100) according to one embodiment may include a base body (80), a waist support frame (20), a driving module (35, 45), a torque transmission frame (50, 55), a thigh fastening part (1, 2), and a waist fastening part (60). The base body (80) may include a lighting unit (85). In one embodiment, at least one of these components (e.g., the lighting unit (85)) may be omitted from the wearable device (100), or one or more other components may be added.

[0051] The base body (80) can be positioned on the lumbar or stomach of the user while the user is wearing the wearable device (100). In one embodiment, the base body (80) can be mounted on the lumbar of the user to provide a cushioning feeling to the user's waist and support the user's waist. The base body (80) can be hung over the user's buttocks (hip area) to prevent the wearable device (100) from falling downward due to gravity while the user is wearing the wearable device (100). The base body (80) can distribute a portion of the weight of the wearable device (100) to the user's waist while the user is wearing the wearable device (100). The base body (80) can be connected to the waist support frame (20). Waist support frame connection elements (not shown) that can be connected to the waist support frame (20) can be provided at both ends of the base body (80).

[0052] In one embodiment, a lighting unit (85) may be provided on the outer surface of the base body (80). The lighting unit (85) may include a light source (e.g., a light emitting diode (LED)). The lighting unit (85) may emit light under the control of a processor (not shown) (e.g., a processor (512) of FIG. 5) of the wearable device (100). According to an embodiment, the lighting unit (85) may be controlled so that visual feedback corresponding to the status of the wearable device (100) may be provided (or output) through the lighting unit (85).

[0053] In one embodiment, a display (not shown) may be provided on the outer surface of the base body (80). The display may provide a screen for various visual information related to the wearable device (100) (e.g., status information of the wearable device (100)) and a user interface.

[0054] The waist support frame (20) can support the user's body (e.g., waist) when the wearable device (100) is worn on the user's body. The waist support frame (20) can extend from both ends of the base body (80). The user's waist can be accommodated on the inside of the waist support frame (20). The waist support frame (20) can include at least one rigid body beam. Each beam can have a curved shape with a preset curvature so as to surround the user's waist. A waist fastening part (60) can be connected to an end of the waist support frame (20). A first driving module (45) and a second driving module (35) can be directly or indirectly connected to the waist support frame (20).

[0055] In one embodiment, a processor, a memory (e.g., a memory (514) of FIG. 5), an inertial sensor (e.g., an inertial sensor (135) of FIG. 1, an inertial sensor (522) of FIG. 5), a communication circuit (e.g., a communication circuit (516) of FIG. 5), an audio output circuit (e.g., an audio output circuit (550) of FIG. 5), and a battery (not shown) may be disposed inside the base body (80). The base body (80) may protect the components disposed inside. The processor may generate a control signal that controls the operation of the wearable device (100). The processor may control a motor (or actuator) of each of the first driving module (45) and the second driving module (35) that generates torque based on electric energy stored in the battery.

[0056] In one embodiment, the wearable device (100) may include one or more sensors. The wearable device (100) may include one or more sensors that acquire sensor data including movement information of the user and / or movement information of components of the wearable device (100). For example, the one or more sensors may include, but are not limited to, an inertial sensor (e.g., the inertial sensor (135) of FIG. 1 and the inertial sensor (522) of FIG. 5) for measuring a movement value of the user's upper body or a movement value of the lumbar support frame (20) and / or an angle sensor (e.g., the angle sensor (125) of FIG. 1 and the first angle sensor (524) and the second angle sensor (524-1) of FIG. 5) for measuring a movement value of the user's hip joint or a movement value of the torque transmission frame (50, 55). For example, the one or more sensors may further include at least one of a position sensor, a torque sensor, a pressure sensor, a temperature sensor, a biosignal sensor (e.g., a heart rate sensor, an electrocardiogram sensor), a distance sensor, or a proximity sensor.

[0057] The waist fastening member (60) can be directly or indirectly connected to the waist support frame (20) and can secure the waist support frame (20) to the user's waist. The waist fastening member (60) can include, for example, a pair of belts.

[0058] The first driving module (45) and the second driving module (35) can generate an external force (or torque) applied to the user's body based on a control signal generated by the processor. For example, the first driving module (45) and the second driving module (35) can generate an assistive force or a resistance force applied to the user's leg. In one embodiment, the first driving module (45) can be positioned corresponding to the user's right hip joint position, and the second driving module (35) can be positioned corresponding to the user's left hip joint position. The first driving module (45) can include a first actuator and a first joint member, and the second driving module (35) can include a second actuator and a second joint member. The first actuator can provide power transmitted to the first joint member, and the second actuator can provide power transmitted to the second joint member. The first actuator and the second actuator may each include a motor that receives power from a battery and generates force (or torque). When powered and driven, the motor may generate force to assist the user's body movements (assistive force) or force to impede the user's body movements (resistive force). In one embodiment, the processor may control the strength and direction of the force generated by the motor by adjusting the voltage and / or current supplied to the motor.

[0059] In one embodiment, the first joint member and the second joint member can receive power from the first actuator and the second actuator, respectively, and apply an external force to the user's body based on the received power. In one embodiment, the first joint member and the second joint member can be disposed at positions corresponding to the user's joints, respectively. One side of the first joint member can be directly or indirectly connected to the first actuator, and the other side can be directly or indirectly connected to the first torque transmission frame (55). The first joint member can be rotated by the power received from the first actuator. An encoder or a hall sensor that can act as an angle sensor for measuring a rotation angle of the first joint member or the first torque transmission frame (55) (corresponding to the user's joint angle) can be disposed on one side of the first joint member. One side of the second joint member can be connected to the second actuator, and the other side can be connected to the second torque transmission frame (50). The second joint member can be rotated by power transmitted from the second actuator. An encoder or hall sensor that can function as an angle sensor for measuring the rotation angle of the second joint member or the second torque transmission frame (50) can also be arranged on one side of the second joint member.

[0060] In one embodiment, the first actuator may be disposed laterally of the first joint member, and the second actuator may be disposed laterally of the second joint member. The rotational axis of the first actuator and the rotational axis of the first joint member may be disposed to be spaced apart from each other, and the rotational axis of the second actuator and the rotational axis of the second joint member may also be disposed to be spaced apart from each other. However, the present invention is not limited thereto, and the actuator and the joint member may share a rotational axis. In one embodiment, each actuator may be disposed to be spaced apart from the joint member. In this case, the drive module (35, 45) may further include a power transmission module (not shown) that transmits power from the actuator to the joint member. The power transmission module may be a rotating body such as a gear, or a longitudinal member such as a wire, a cable, a string, a spring, a belt, or a chain. However, the scope of the embodiment is not limited by the positional relationship between the actuator and joint member and the power transmission structure described above.

[0061] In one embodiment, the first torque transmission frame (55) and the second torque transmission frame (50) can transmit the torque generated by the first driving module (45) and the second driving module (35) to the user's body (e.g., the leg) when the wearable device (100) is worn on the user's leg. The transmitted torque can act as an external force applied to the user's leg movement. One end of each of the first torque transmission frame (55) and the second torque transmission frame (50) can be directly or indirectly connected to a joint member and rotated. The other end of each of the first torque transmission frame (55) and the second torque transmission frame (50) is directly or indirectly connected to the first thigh fastening portion (2) and the second thigh fastening portion (1), so that the first torque transmission frame (55) and the second torque transmission frame (50) can support the user's thigh while transmitting the torque generated by the first driving module (45) and the second driving module (35) to the user's thigh. For example, the first torque transmission frame (55) and the second torque transmission frame (50) can push or pull the user's thigh. The first torque transmission frame (55) and the second torque transmission frame (50) can extend along the length of the user's thigh and can be bent to wrap at least a portion of the user's thigh circumference. The first torque transmission frame (55) can be a torque transmission frame for transmitting torque to the user's right leg, and the second torque transmission frame (50) can be a torque transmission frame for transmitting torque to the user's left leg.

[0062] The first thigh fastening part (2) and the second thigh fastening part (1) are directly or indirectly connected to the first torque transmission frame (55) and the second torque transmission frame (50), respectively, and can fasten the wearable device (100) to the user's leg (particularly, the thigh). The first thigh fastening part (2) may be a thigh fastening part for fastening the wearable device (100) to the user's right thigh, and the second thigh fastening part (1) may be a thigh fastening part for fastening the wearable device (100) to the user's left thigh.

[0063] In one embodiment, the first thigh fastening unit (2) may include a first cover, a first fastening frame, and a first strap, and the second thigh fastening unit (1) may include a second cover, a second fastening frame, and a second strap. The first cover and the second cover may apply torque generated from the first driving module (45) and the second driving module (35) to the user's thigh, respectively. The first cover and the second cover may be disposed on one side of the user's thigh, respectively, to push or pull the user's thigh. The first cover and the second cover may be disposed along the circumferential direction of the user's thigh. The first cover and the second cover may extend in both directions with respect to the other end of the first torque transmission frame (55) and the second torque transmission frame (50), respectively, and may include a curved surface corresponding to the user's thigh. One end of each of the first cover and the second cover may be directly or indirectly connected to the first fastening frame and the second fastening frame, respectively. The other end of each of the first cover and the second cover can be directly or indirectly connected to the first strap and the second strap.

[0064] The first fastening frame and the second fastening frame may be arranged to, for example, surround at least a portion of the user's thigh, thereby preventing the user's thigh from being detached from the wearable device (100) or reducing the possibility of detachment. The first fastening frame may have a fastening structure connecting the first cover and the first strap, and the second fastening frame may have a fastening structure connecting the second cover and the second strap.

[0065] The first strap may encircle the user's right thigh, the remaining portion not covered by the first cover and the first fastening frame, and the second strap may encircle the user's left thigh, the remaining portion not covered by the second cover and the second fastening frame. The first strap and the second strap may comprise, for example, an elastic material (e.g., a band).

[0066]

[0067] FIG. 5 is a diagram illustrating configurations of a wearable device according to various embodiments.

[0068] Referring to FIG. 5, the wearable device (100) may include a control circuit (510), a communication circuit (516), one or more sensors (e.g., an inertial sensor (522, a first angle sensor (524), a second angle sensor (524-1)), a drive module (530, 530-1), an input circuit (540), an audio output circuit (550) including a speaker, and a haptic circuit (560).

[0069] The drive module (530) may include a motor (534) and a motor driver circuit (532) for driving the motor (534), and the drive module (530-1) may include a motor (534-1) and a motor driver circuit (532-1) for driving the motor (534-1). In the embodiment of FIG. 5, two drive modules are illustrated, but this is merely an example, and the number of drive modules may be one or three or more. The drive module (530) including the motor driver circuit (532) and the motor (534) may correspond to the first drive module (45) of FIG. 3, and the drive module (530-1) including the motor driver circuit (532-1) and the motor (534-1) may correspond to the second drive module (35) of FIG. 3.

[0070] One or more sensors may include one or more sensors that acquire sensor data (or sensed values). The one or more sensors may transmit the acquired sensor data to the control circuit (510). The one or more sensors may include, for example, an inertial sensor (522), a first angle sensor (524), and / or a second angle sensor (524-1). Each of these sensors may be present in multiples, and some may be omitted.

[0071] The inertial sensor (522) can measure the movement value of the user's body. The inertial sensor (522) can sense the acceleration of the X-axis, Y-axis, and Z-axis and the angular velocity of the X-axis, Y-axis, and Z-axis according to the user's movement. The inertial sensor (522) can measure, for example, the movement value of the user's upper body. The movement value of the user's upper body can correspond to the movement value of the waist support frame of the wearable device (100) (e.g., the waist support frame (20) of FIGS. 3 and 4). In one embodiment, the inertial sensor (522) can be located on a printed circuit board present inside the base body (80) of the wearable device (100), and can measure a signal indicating the degree of inclination of the wearable device (100) and / or the acceleration of the wearable device (100).

[0072] The first angle sensor (524) and the second angle sensor (524-1) can measure the hip joint angle according to the user's leg movement. The first angle sensor (524) can sense the hip joint angle of the user's right leg, and the second angle sensor (524-1) can sense the hip joint angle of the user's left leg. Each of the first angle sensor (524) and the second angle sensor (524-1) can include, for example, an encoder and / or a hall sensor. The hip joint angle of the right leg sensed by the first angle sensor (524) may correspond to a movement value (e.g., a rotation angle value) of the first torque transmission frame of the wearable device (e.g., the first torque transmission frame (55) of FIG. 3), and the hip joint angle of the left leg sensed by the second angle sensor (524-1) may correspond to a movement value (e.g., a rotation angle value) of the second torque transmission frame of the wearable device (e.g., the second torque transmission frame (50) of FIG. 3).

[0073] In one embodiment, the one or more sensors may further include a torque sensor for sensing a torque value, a position sensor for obtaining a position value of the wearable device (100), a proximity sensor for detecting the proximity of an object, a biosignal sensor for detecting a biosignal of a user, a distance sensor for measuring a distance to an object, a pressure sensor for measuring a pressure value, and / or a temperature sensor for measuring an ambient temperature.

[0074] The input circuit (540) can receive commands or data to be used in a component of the wearable device (100) (e.g., a processor (512)) from an external source (e.g., a user) of the wearable device (100). The input circuit (540) can include, for example, keys (e.g., buttons) and / or a touch screen.

[0075] The audio output circuit (550) can output an audio signal to the outside of the wearable device (100). The audio output circuit (550) can include a speaker that plays a guide audio signal (e.g., a driving start sound, an operation error notification sound), music content, or a guide voice.

[0076] In one embodiment, the wearable device (100) may further include a battery (not shown) for supplying power to each component of the wearable device (100) and a power management circuit (not shown) for controlling the power supply. The wearable device (100) may convert the power of the battery to an operating voltage of each component of the wearable device (100) and supply the converted power to each component.

[0077] The drive module (530, 530-1) can generate an external force applied to the user's leg under the control of the control circuit (510). The drive module (530, 530-1) is located at a location corresponding to the user's hip joint position and can generate a torque applied to the user's leg based on a control signal generated by the control circuit (510). The control circuit (510) can transmit the control signal to the motor driver circuit (532, 532-1), and the motor driver circuit (532, 532-1) can control the operation of the motor (534, 534-1) by generating a current signal (or voltage signal) corresponding to the control signal and supplying it to the motor (534, 534-1). Depending on the control signal, the current signal may not be supplied to the motor (534, 534-1). The motor (534, 534-1) can generate an assistive force that assists the user's leg movement or a resistive force that impedes the leg movement when a current signal is supplied to the motor (534, 534-1) and the motor is driven.

[0078] The control circuit (510) controls the overall operation of the wearable device (100) and can generate control signals for controlling each component of the wearable device (100). The control circuit (510) may include a processor (512) and a memory (514).

[0079] The processor (512) may, for example, execute software to control at least one other component (e.g., a hardware or software component) of the wearable device directly or indirectly connected to the processor (512), and may perform various data processing or calculations. According to one embodiment, as at least a part of the data processing or calculation, the processor (512) may store instructions or data received from another component (e.g., a communication circuit (516)) in the memory (514), process the instructions or data stored in the memory (514), and store the result data after the processing in the memory (514). The processor (512) may include one or more processors, and the operations of the wearable device (100) described in the present disclosure may be performed by one processor or by a combination of multiple processors.

[0080] According to one embodiment, the processor (512) may include at least one of a main processor (e.g., a central processing unit (CPU) or an application processor) and / or an auxiliary processor (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor) that can operate independently or in conjunction therewith. The processor (512) may also be implemented as a system on chip (SoC) or an integrated circuit (IC) that performs processing. The auxiliary processor may be implemented separately from the main processor or as a part thereof.

[0081] The memory (514) can store various data used by at least one component (e.g., the processor (512)) of the control circuit (510). The data can include, for example, input data or output data for software, sensor data, and commands related thereto. The memory (514) can include at least one instruction executable by the processor (512). The memory (514) can include one or more memories, and instructions for controlling the processor (512) to perform operations of the wearable device (100) described in the present disclosure can be stored in one memory or can be divided and stored in multiple memories. The memory (514) can include a volatile memory or a non-volatile memory.

[0082] The communication circuit (516) may support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between the control circuit (510) and other components of the wearable device (100) or an external electronic device (e.g., the electronic device (210) of FIG. 2 or another wearable device (220)), and the performance of communication through the established communication channel. The communication circuit (516) may, for example, transmit sensor data acquired by a sensor to an external electronic device (e.g., the electronic device (210) of FIG. 2) and receive a control signal from the external electronic device. In one embodiment, the communication circuit (516) may include one or more communication processors that operate independently from the processor (512) and support direct (e.g., wired) communication or wireless communication. In one embodiment, the communication circuit (516) may include a wireless communication circuit (e.g., a cellular communication circuit, a short-range wireless communication circuit, or a global navigation satellite system (GNSS) communication circuit) and / or a wired communication circuit. The wireless communication circuitry may communicate with other components of the wearable device (100) and / or external devices via, for example, Bluetooth, WiFi (wireless fidelity), ANT (advanced and adaptive network technology), IrDA (infrared data association), a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a local area network (LAN) or a wide area network (WAN).

[0083] The haptic circuit (560) can provide haptic feedback to a user under the control of the processor (512). The haptic circuit (560) can include one or more haptic actuators. The haptic actuators can include, for example, a piezo actuator, a bander type actuator, and / or a vibration motor-based actuator. The haptic actuators can be one or more. In one embodiment, the haptic actuator can be located in at least one of a base body (e.g., the base body (80) of FIG. 3), a torque transmission frame (e.g., the first torque transmission frame (75) of FIG. 3, the second torque transmission frame (70)), and a thigh fastening part (e.g., the first thigh fastening part (2) of FIG. 3, the second thigh fastening part (1)) of the wearable device (100).

[0084]

[0085] FIG. 6 is a diagram illustrating interaction between a wearable device and an electronic device according to various embodiments.

[0086] Referring to FIG. 6, a wearable device (100) can communicate with an electronic device (210). For example, the electronic device (210) may be a user terminal of a user using the wearable device (100). According to one embodiment, the wearable device (100) and the electronic device (210) may be connected to each other via short-range wireless communication (e.g., Bluetooth communication, Wi-Fi communication).

[0087] In one embodiment, the electronic device (210) may execute an application for checking the status of the wearable device (100) or controlling or operating the wearable device (100). By executing the application, a screen of a user interface (UI) for controlling the operation of the wearable device (100) or determining the operation mode of the wearable device (100) may be displayed on the display (212) of the electronic device (210). The UI may be, for example, a graphical user interface (GUI).

[0088] In one embodiment, a user may input a command (e.g., a command to execute a walking assistance mode or an exercise assistance mode) for controlling the operation of the wearable device (100) or change the settings of the wearable device (100) through a GUI screen on a display (212) of the electronic device (210). In addition, the user may set an exercise goal and change a torque parameter to be applied to the wearable device (100) through the GUI screen. The torque parameter may include, for example, a first parameter that controls the intensity of a torque generated by a motor of the wearable device (100) (e.g., motor (534) or motor (534-1) of FIG. 5) and / or a second parameter that controls the timing of application of the torque. In various embodiments of the present disclosure, the term 'torque parameter' may be replaced with the term 'parameter', 'robot parameter', or 'control parameter'.

[0089] The electronic device (210) can generate a control command (or control signal) corresponding to a motion control command or setting change command input by a user, and transmit the generated control command to the wearable device (100). In one embodiment, the control command may include a torque parameter set by the user. The wearable device (100) can operate according to the received control command, and transmit a control result according to the control command and / or sensor data measured by a sensor module of the wearable device (100) to the electronic device (210). The electronic device (210) can analyze the control result and / or sensor data to provide the user with result information (e.g., current exercise status information, exercise result information, exercise posture evaluation information, physical ability evaluation information) through a GUI screen.

[0090]

[0091] FIG. 7 is a diagram illustrating configurations of an electronic device according to various embodiments.

[0092] Referring to FIG. 7, the electronic device (210) may include a processor (710), a memory (720), a communication circuit (730), a display circuit (740), an audio output circuit (750), and an input circuit (760). In one embodiment, the electronic device (210) may omit at least one of these components (e.g., an audio output circuit (750)), or may have one or more other components added (e.g., a sensor circuit, a haptic circuit, a battery).

[0093] The processor (710) may control at least one other component (e.g., hardware or software component) of the electronic device (210) and perform various data processing or calculations. According to one embodiment, as at least a part of the data processing or calculation, the processor (710) may store a command or data received from another component (e.g., communication circuit (730)) in the memory (720), process the command or data stored in the memory (720), and store the resulting data in the memory (720). The processor (710) may include one or more processors, and the operations of the electronic device (210) described in the present disclosure may be performed by one processor or by a combination of multiple processors.

[0094] According to one embodiment, the processor (710) may include at least one of a main processor (e.g., a central processing unit or an application processor) or an auxiliary processor (e.g., a graphics processing unit, a neural network processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor) that can operate independently or in conjunction with the main processor. The processor (512) may also be implemented as a system on a chip (SoC) or an integrated circuit that performs processing.

[0095] The memory (720) can store various data used by at least one component (e.g., the processor (710) or the communication circuit (730)) of the electronic device (210). The data can include, for example, input data or output data for a program (e.g., an application) and instructions related thereto. The memory (720) can include at least one instruction executable by the processor (710). The memory (720) can include one or more memories, and instructions for controlling the processor (710) to perform operations of the electronic device (210) described in the present disclosure can be stored in one memory or can be divided and stored in multiple memories. The memory (720) can include a volatile memory or a non-volatile memory.

[0096] The communication circuit (730) may support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between the electronic device (210) and another electronic device (e.g., wearable device (100), another wearable device (220), server (230)), and the performance of communication through the established communication channel. The communication circuit (730) may include a communication circuit for performing a communication function. The communication circuit (730) may operate independently from the processor (710) (e.g., application processor) and may include one or more communication processors that support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication circuit (730) may include a wireless communication circuit (e.g., a Bluetooth communication circuit, a cellular communication circuit, a Wi-Fi communication circuit, or a GNSS communication circuit) or a wired communication circuit (e.g., a LAN communication circuit or a power line communication circuit) that performs wireless communication. The communication circuit (730) may, for example, transmit a control command to the wearable device (100) and receive at least one of sensor data including body movement information of a user wearing the wearable device (100), status data of the wearable device (100), or control result data corresponding to the control command from the wearable device (100).

[0097] The display circuit (740) can visually provide information to an external device (e.g., a user) of the electronic device (210). The display circuit (740) can include, for example, an LCD or OLED display, a holographic device, or a projector device. The display circuit (740) can further include a control circuit for controlling display operation. In one embodiment, the display circuit (740) can further include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of a force generated by a touch. The display circuit (740) can output a user interface screen for controlling the wearable device (100) or providing various information (e.g., exercise evaluation information, setting information of the wearable device (100).

[0098] The audio output circuit (750) can output an audio signal to the outside of the electronic device (210). The audio output circuit (750) can include a speaker that plays a guide audio signal (e.g., a driving start sound, an operation error notification sound), music content, or a guide voice based on the status of the wearable device (100).

[0099] The input circuit (760) can receive commands or data to be used in a component of the electronic device (210) (e.g., a processor (710)) from an external source (e.g., a user) of the electronic device (210). The input circuit (760) can include an input component circuit and can receive user input. The input circuit (760) can include, for example, a key (e.g., a button) and / or a touch recognition circuit for recognizing a touch on a screen.

[0100] In one embodiment, the electronic device (210) can control the operation of the wearable device (100) based on the user's exercise data. The electronic device (210) can not only measure the exercise ability and exercise status of the user exercising using the wearable device (100), but also directly provide the exercise intensity to be applied to the user's exercise. The provision of the exercise intensity can be implemented by controlling the torque output from the motor of the wearable device (100) (e.g., the motor (534) and the motor (534-1) of FIG. 5 ).

[0101] When a user wears a wearable device (100) and performs exercise (e.g., walking exercise), the user can set an exercise goal through a UI provided by the electronic device (210). For example, the user can set at least one of an exercise type related to a target maximum heart rate, a target walking speed, a target stride length, and a target exercise time. With respect to the exercise type, the user can select any one of low-intensity exercise, weight control exercise, aerobic exercise, anaerobic exercise, and maximum heart rate exercise as the exercise type. The electronic device (210) can actively control the wearable device (100) to achieve the exercise goal set by the user. The electronic device (210) can control a torque parameter related to a torque output from a motor of the wearable device (100) based on the exercise goal. The torque parameter can include a first parameter that controls the intensity of the torque output from the motor and / or a second parameter that controls the timing of application of the torque. Depending on the first parameter, the greater the torque, the greater the force received by the user. Depending on the second parameter, the timing at which the torque of the wearable device (100) is transmitted to the user or the response speed of the wearable device (100) may vary.

[0102] The electronic device (210) may automatically adjust the torque intensity and / or torque application timing of the wearable device (100) to achieve a target heart rate zone, target stride length, and / or target walking speed defined in an exercise type set by the user. The electronic device (210) may determine the torque parameter of the wearable device (100) for each exercise section of the exercise performed by the user based on the exercise goal set by the user, and control the torque output of the wearable device (100) based on the determined torque parameter. The torque intensity or torque application timing may vary for each exercise section. According to various embodiments described in the present disclosure, the electronic device (210) may help the user efficiently achieve the exercise goal and may personalize the operation of the wearable device (100) based on the user's exercise data. Since the user does not have to manually adjust the torque parameter of the wearable device (100) to achieve the exercise goal, the convenience of use may be improved.

[0103] An electronic device (210) according to one embodiment may include one or more processors (710), an input circuit (760) for receiving user input, and a communication circuit (730) for communicating with a wearable device (100). The electronic device (210) may provide a user interface (UI) related to an exercise program to a user through a display included in a display circuit (740). The user may set an exercise type and / or an exercise goal through the UI. User input (e.g., touch input) for the exercise goal may be received through the input circuit (760). One or more processors (512) may, in response to receiving user input regarding an exercise goal through an input circuit (760), determine a torque parameter related to the achievement of the exercise goal using a neural network model (e.g., a neural network model (830) of FIG. 8, a neural network model (900) of FIG. 9, a neural network model (1200) of FIG. 12) that inputs feature data related to an exercise performed by a user wearing a wearable device (100) and outputs an estimated exercise performance index.

[0104] In one embodiment, the neural network model may be learned through the learning process described below with reference to FIG. 9. The neural network model may be a model that learns and models the relationship between the user's exercise data, the torque parameters of the wearable device (100), and the user's exercise performance indicators. Using the neural network model, the torque parameters of the wearable device (100) for achieving exercise goals may be determined. Various neural network models may be used to determine the torque parameters. For example, neural network models may exist for each exercise type, exercise difficulty, and / or exercise interval.

[0105] In one embodiment, the feature data related to exercise input to the neural network model may include feature values ​​for exercise segments performed by the user (e.g., exercise segment timestamp, exercise segment index). The feature data related to exercise may further include feature values ​​for at least one of an exercise elapsed time (e.g., an exercise session timestamp) and whether a torque parameter has been changed by a user input. The estimated exercise performance index, which is output data of the neural network model, may include, for example, at least one of the user's heart rate, walking speed, and stride length. The neural network model may estimate the user's exercise performance index based on the feature data related to exercise as input data and the torque parameter applied to the wearable device (100) and output the estimated exercise performance index.

[0106] In one embodiment, an exercise performed by a user may include multiple exercise segments. The start and end of the exercise may be defined as a single exercise session, and a single exercise session may include multiple exercise segments. A rest period of a certain length may exist between the exercise segments. The torque parameters applied to the wearable device (100) may be the same within a single exercise segment. The torque parameters may be the same or different between exercise segments.

[0107] In one embodiment, one or more processors (710) may use a neural network model to determine torque parameters to be applied to each exercise segment. Determining the torque parameters may be performed prior to the start of exercise and / or during a rest period between exercise segments. Upon completion of one exercise segment, one or more processors (710) may determine torque parameters to be applied to the next exercise segment during the rest period.

[0108] In one embodiment, when the first exercise section is completed, one or more processors (710) can train a neural network model based on data collected up to the first exercise section, and use the trained neural network model to determine a torque parameter to be applied to the second exercise section, which is the next exercise section of the first exercise section.

[0109] One or more processors (710) can estimate exercise-related feature data for the next exercise segment. The one or more processors (710) can estimate feature data for the next exercise segment based on feature values ​​for exercise segments completed so far. For example, the exercise session timestamp, exercise segment timestamp, and exercise segment index for the next exercise segment can be estimated based on the exercise session timestamp, exercise segment timestamp, and exercise segment index performed so far. The one or more processors (710) can input the estimated feature data for the next exercise segment and a torque parameter corresponding to a reference value into a neural network model, and obtain an estimated exercise performance index from the neural network model. Here, the torque parameter corresponding to the reference value may be, for example, a torque parameter having a predefined value. The one or more processors (710) can determine a torque parameter to be applied to the next exercise segment based on the estimated exercise performance index obtained from the neural network model and an exercise goal index for achieving an exercise goal. The exercise goal index is an index representing an exercise goal set by a user input. Exercise goal metrics may include, for example, target maximum heart rate, target walking speed, and / or target stride length.

[0110] In one embodiment, one or more processors (710) may search for torque parameters that satisfy a set condition based on the difference between the estimated exercise performance indicator and the exercise target indicator obtained from the neural network model. One or more processors (710) may set the difference between the estimated exercise performance indicator and the exercise target indicator as a loss in machine learning, and search for torque parameters that minimize the loss or torque parameters that make the loss below a threshold.

[0111] In one embodiment, the one or more processors (710) can control the wearable device (100) to generate torque based on the determined torque parameter by causing the communication circuit (730) to transmit a control signal including the determined torque parameter to the wearable device (100). The one or more processors (710) can control the communication circuit (730) to transmit a control signal including the found torque parameter to the wearable device (100) in response to a torque parameter being found that satisfies a set condition based on a difference between an estimated exercise performance index obtained from a neural network model and an exercise target index, so that the wearable device (100) generates torque based on the found torque parameter in a next exercise section.

[0112] In one embodiment, when the entire exercise section of the exercise is completed, learning (machine learning) for a neural network model may be performed based on sensor data collected during the exercise, torque parameters applied during the exercise, and feature data related to the exercise. One or more processors (710) may update model parameters of a neural network model based on exercise performance indices measured based on sensor data collected during the exercise and the estimated exercise performance indices output from the neural network model when the entire exercise section of the exercise is completed. Learning (machine learning) for a neural network model may be performed based on sensor data collected during the exercise, torque parameters applied during the exercise, and feature data related to the exercise.

[0113] Sensor data collected during the exercise may include, for example, a user's heart rate measured by a heart rate sensor, a hip joint angle value measured by an angle sensor of the wearable device (100) (e.g., the first angle sensor (524) and the second angle sensor (524-1) of FIG. 5), and / or a movement value measured by an inertial sensor (522) of the wearable device (100). One or more processors (710) may measure the user's maximum heart rate based on the heart rate measured by the heart rate sensor. The heart rate sensor may be provided in, for example, a smartwatch (e.g., the smartwatch (224) of FIG. 2) and / or the wearable device (100). One or more processors (710) may measure the user's walking speed and / or stride length based on the hip joint angle value measured by the angle sensor of the wearable device (100). One or more processors (710) may measure the user's walking speed based on movement values ​​measured through the inertial sensors of the wearable device (100). The one or more processors (710) may use the exercise performance index measured based on the sensor data as label data for machine learning. The one or more processors (710) may calculate the difference between the estimated exercise performance index output by the neural network model and the exercise performance index measured based on the sensor data as a loss, and update the model parameters (e.g., connection weights and biases between artificial neurons) of the neural network model so as to minimize the calculated loss. Here, a backpropagation algorithm that updates the model parameters based on the loss may be used. The neural network model for which learning is completed may be stored and used as the initial neural network model used at the start of the user's next exercise. The initial neural network model may be used to determine the torque parameter to be applied to the first exercise section of the next exercise.

[0114]

[0115] FIG. 8 is a diagram illustrating a system for controlling a wearable device based on a user's exercise data according to various embodiments.

[0116] Referring to FIG. 8, the system may include a wearable device (100), a control module (810), and an interaction module (840). The control module (810) may include a movement data extractor (812) and a torque parameter determiner (814). The control module (810) is a module that controls movement performed by a user and the wearable device (100), and may be implemented by a processor (e.g., a processor (710) of FIG. 7), a memory (e.g., a memory (720) of FIG. 7), and a communication circuit (e.g., a communication circuit (730) of FIG. 7) of the electronic device (210). The interaction module (840) is a module that performs interaction with a user, and may be implemented by a display circuit (e.g., a display circuit (740) of FIG. 7) and an input circuit (e.g., an input circuit (760) of FIG. 7) of the electronic device (210). The exercise data database (820) and neural network model (830) may be stored in the electronic device (210) or in a device other than the electronic device (210) (e.g., the server (230) of FIG. 2).

[0117] The exercise data database (820) can store the user's exercise data. For example, the exercise data database (820) can store exercise data that records the type of exercise performed by the user, the type of exercise (e.g., low-intensity exercise, weight-control exercise, aerobic exercise, anaerobic exercise, and maximum heart rate exercise), exercise time, exercise intensity, heart rate measured during exercise, walking speed, and stride length, but the types of exercise data stored are not limited thereto.

[0118] The neural network model (830) may be a machine learning model that learns relationships between exercise data. The operation of the neural network model (830) may be determined by model parameters (e.g., connection weights and biases between artificial neurons) of the neural network model (830). The model parameters are learnable parameters and may be determined through a machine learning process. Various neural network models may exist depending on the type of learning data learned by the neural network model (830). For example, there may be a neural network model learned based on the user's accumulated overall exercise data, a neural network model learned based on the user's exercise data for a day, a neural network model learned based on the user's exercise data for a specific time period, or a neural network model learned based on exercise data collected for each exercise type.

[0119] The control module (810) can control the overall operation of the wearable device (100). The movement data extractor (812) of the control module (810) can extract the user's movement data from sensor data acquired through a sensor of the wearable device (100) or a sensor of another device (e.g., a smartwatch (224) of FIG. 2). The movement data extractor (812) can acquire data on the user's heart rate based on the heart rate measured through the heart rate sensor. The movement data extractor (812) can acquire data on the user's walking speed and / or stride based on a hip joint angle value measured through an angle sensor of the wearable device (100) (e.g., the first angle sensor (524) and the second angle sensor (524-1) of FIG. 5). The movement data extractor (812) can obtain data on the user's walking speed based on the movement value measured through the inertial sensor of the wearable device (100).

[0120] The torque parameter determiner (814) of the control module (810) can determine a torque parameter that determines the driving characteristics of the wearable device (100). The torque parameter can include a first parameter that controls the intensity of the torque output from the motor of the wearable device (100) (e.g., the motor (534) and the motor (534-1) of FIG. 5) and a second parameter that controls the timing at which the torque is applied. The control module (810) can determine the torque parameter to be applied to the wearable device (100) using the neural network model (830) according to the method described in FIG. 7. The torque parameter determiner (814) can determine the torque parameter for each exercise section according to the user's exercise goal and transmit the torque parameter determined for each exercise section to the wearable device (100). Even if the user does not specify specific torque parameters for each exercise segment, the torque parameter determiner (814) can determine torque parameters for each exercise segment to achieve the user's exercise goal and control the wearable device (100) in each exercise segment based on the determined torque parameters. The torque parameters may be the same or different between exercise segments.

[0121] The interaction module (840) can interact with a user. For example, the interaction module (840) can provide a UI screen to the user through a display and receive user input. The user can set an exercise goal through user input, and the interaction module (840) can display a torque parameter (torque parameter determined by the torque parameter determiner (814)) for achieving the set exercise goal to the user through the UI screen. The user can adjust the torque parameter through user input, and the interaction module (840) can transmit the torque parameter adjusted by the user input to the torque parameter determiner (814). The torque parameter determiner (814) can control the wearable device (100) based on the torque parameter adjusted by the user input.

[0122]

[0123] FIGS. 9 and 10 are diagrams for explaining input / output data and learning operations of a neural network model according to various embodiments.

[0124] Referring to FIG. 9, a neural network model (900) (e.g., the neural network model (830) of FIG. 8) may output an estimated exercise performance index of a user as output data when input data is input. The exercise performance index output by the neural network model (900) may not be an actually measured value, but an estimated value based on the input data. The neural network model (900) may be a model designed to output an estimated value of a user's exercise performance index when feature data and torque parameters related to exercise are input.

[0125] The input data of the neural network model (900) may include feature data related to exercise and torque parameters representing the operating characteristics of the wearable device (100). The feature data related to exercise may be time series data up to a specific time interval. The feature data related to exercise may include feature values ​​related to timestamps of exercise performed by the user. For example, the feature data may include feature values ​​for an exercise session timestamp and an exercise segment timestamp. The exercise session timestamp represents the elapsed exercise time measured after the start of exercise, and the exercise segment timestamp represents the elapsed time of the exercise segment measured after the start of the corresponding exercise segment. The feature data related to exercise may further include feature values ​​for at least one of an exercise segment index and whether a torque parameter has been changed by a user input. The exercise segment index represents an index (or order) of the corresponding exercise segment among all exercise segments. The feature data related to exercise may be composed of, for example, {an exercise session timestamp, an exercise segment timestamp, an exercise segment index, and a value indicating whether a torque parameter has been changed by a user input}. The value indicating whether the torque parameter has been changed by the user input may be designated as '1' if the user has changed the torque parameter through the user input, or '0' if the torque parameter has not been changed, but is not limited thereto. The torque parameter of the wearable device (100) input to the neural network model (900) may be a torque parameter (e.g., a parameter for the strength of the torque, a parameter for the timing of application of the torque) applied to the wearable device (100) at the exercise section timestamp.

[0126] The neural network model (900) can estimate and output a user's exercise performance indicator from input data. The estimated exercise performance indicator may include, for example, at least one of the user's heart rate, walking speed, and stride length.

[0127] Actual exercise performance indicators measured based on sensor data can be used as label data for machine learning. Label data can be selected from exercise data extracted by, for example, the exercise data extractor (812) of FIG. 8. Label data can include data on the user's heart rate, walking speed, and / or stride length measured based on sensor data. Label data is used as ground truth in machine learning.

[0128] An example of feature data, torque parameters, and label data related to exercise is illustrated in FIG. 10. Referring to FIG. 10, a table is illustrated showing data (1010) collected to determine whether a torque parameter has changed by an exercise session timestamp, an exercise segment timestamp, an exercise segment index, and a user input, data (1020) collected to determine a torque parameter of a torque delay that controls the torque intensity and the timing of torque application at each exercise session timestamp applied to a wearable device (100), and data (1030) collected to determine a user's heart rate, walking speed, and stride length at each exercise session timestamp measured based on sensor data. Feature data related to exercise for input to a neural network model (900) can be extracted from data (1010), and torque parameters for input to a neural network model (900) can be extracted from data (1020). Label data to be compared with the output data of the neural network model (900) can be extracted from data (1030). The input data and output data of the neural network model (900) may be time-series data. For example, feature data related to exercise in a specific time interval and torque parameters applied to the wearable device (100) in the specific time interval may be used as input data for training the neural network model (900), and the user's exercise performance indicators measured in the specific time interval may be used as label data.

[0129] A processor of an electronic device (210) (e.g., processor (710) of FIG. 7) can train a neural network model (900) based on input / output data and label data of the neural network model (900). Through the training process, the neural network model (900) can learn from the user's exercise data how the user's exercise performance indicators change according to exercise-related feature data and torque parameters. Here, the exercise data may be time-series data acquired at a defined time period.

[0130] The neural network model (900) is given motion-related feature data as learning data at each time t. and torque parameters Based on the user's exercise performance indicators can be estimated. The processor outputs the estimated exercise performance index output by the neural network model (900). and measured exercise performance indicators shown in the label data. The difference between the two can be calculated as a loss. For example, the estimated exercise performance index and measured exercise performance indicators The mean squared error (MSE) between the two can be calculated as a loss. The processor can continuously update the model parameters (e.g., connection weights and biases between artificial neurons) of the neural network model (900) in a direction that reduces the calculated loss.

[0131] In one embodiment, the learning process of the neural network model (900) described above may be performed between exercise intervals or after the exercise is completed. When learning is performed between exercise intervals, learning is performed based on exercise data collected up to the previous exercise interval. When learning is performed after the exercise is completed, learning may be performed based on exercise data collected throughout the entire exercise interval. The trained neural network model (900) may be used to estimate torque parameters to be applied to the wearable device (100) when the user performs the next exercise.

[0132]

[0133] FIG. 11 is a flowchart illustrating operations of a control method for controlling a wearable device based on a user's exercise data according to various embodiments. In one embodiment, at least one of the operations in FIG. 11 may be performed simultaneously or in parallel with another operation, and the order of the operations may be changed. Furthermore, at least one of the operations may be omitted, and another operation may be additionally performed.

[0134] Referring to FIG. 11, in operation (1110), an input circuit of an electronic device (210) (e.g., input circuit (760) of FIG. 7) may receive user input regarding an exercise goal. The input circuit may receive user input including, for example, a selection of at least one of an exercise type, a target walking speed, and a target stride length associated with a target maximum heart rate. The user may set or input an exercise goal through a UI screen provided through a display of the electronic device (210) before starting an exercise.

[0135] In operation (1120), a processor of an electronic device (210) (e.g., processor (710) of FIG. 7) may determine a torque parameter related to the achievement of an exercise goal by using a neural network model (e.g., neural network model (830) of FIG. 8, neural network model (900) of FIG. 9, neural network model (1200) of FIG. 12) that inputs feature data related to an exercise performed by a user wearing a wearable device (100) and outputs an estimated exercise performance index. The neural network model may be a model that learns and models the relationship between the user's exercise data, the torque parameter of the wearable device (100), and the exercise performance index of the user. The torque parameter of the wearable device (100) for achieving an exercise goal may be determined by using the neural network model. The processor may determine at least one of a first parameter that determines the strength of a torque output from a motor of the wearable device (100) (e.g., motor (534) and motor (534-1) of FIG. 5) and a second parameter that determines the point in time at which the torque is applied.

[0136] In one embodiment, the feature data related to exercise input to the neural network model may include feature values ​​for exercise segments performed by the user (e.g., exercise segment timestamp, exercise segment index). The feature data related to exercise may further include feature values ​​for at least one of an exercise elapsed time (e.g., an exercise session timestamp) and whether a torque parameter has been changed by a user input. The estimated exercise performance index, which is output data of the neural network model, may include, for example, at least one of the user's heart rate, walking speed, and stride length. The neural network model may estimate the user's exercise performance index based on the feature data related to exercise as input data and the torque parameter applied to the wearable device (100) and output the estimated exercise performance index.

[0137] In one embodiment, a workout performed by a user may include multiple workout segments. In one embodiment, upon completion of a first workout segment, the processor may train a neural network model based on data collected up to the first workout segment, and use the trained neural network model to determine torque parameters to be applied to a second workout segment, which is the next workout segment after the first workout segment.

[0138] In one embodiment, the processor can estimate exercise-related feature data for the next exercise segment. The processor can estimate feature data for the next exercise segment based on feature values ​​for the exercise segment completed so far. For example, the processor can estimate the exercise session timestamp, exercise segment timestamp, and exercise segment index for the next exercise segment based on the exercise session timestamp, exercise segment timestamp, and exercise segment index that have been performed so far. The processor can input the estimated feature data for the next exercise segment and a torque parameter corresponding to a reference value into a neural network model, and obtain an estimated exercise performance index from the neural network model. Here, the torque parameter corresponding to the reference value may be, for example, a torque parameter having a predefined value. The processor can determine a torque parameter to be applied to the next exercise segment based on the estimated exercise performance index obtained from the neural network model and an exercise target index for achieving an exercise goal. The exercise target index may include, for example, a target maximum heart rate, a target walking speed, and / or a target stride.

[0139] In one embodiment, the processor may search for torque parameters that satisfy a set condition based on the difference between the estimated exercise performance indicator and the exercise target indicator obtained from the neural network model. The processor may set the difference between the estimated exercise performance indicator and the exercise target indicator as a loss in machine learning, and search for torque parameters that minimize the loss or torque parameters that reduce the loss to a threshold value.

[0140] In one embodiment, a display circuit of the electronic device (210) (e.g., display circuit (740) of FIG. 7) may output information about the determined torque parameter. When the input circuit receives a user input including a change input for the torque parameter, the processor, in response to receiving the user input, may determine the torque parameter changed by the change input as a torque parameter related to the achievement of the exercise goal.

[0141] In operation (1130), the communication circuit of the electronic device (210) (e.g., the communication circuit (730) of FIG. 7) may transmit a control signal including the determined torque parameter to the wearable device (100) so that the wearable device (100) generates torque based on the determined torque parameter under the control of the processor.

[0142] In one embodiment, the processor may control the communication circuit (730) to transmit a control signal including the found torque parameter to the wearable device (100) so that the wearable device (100) generates torque based on the found torque parameter in the next exercise section in response to the finding of a torque parameter that satisfies a set condition based on the difference between the estimated exercise performance index obtained from the neural network model and the exercise target index.

[0143] In one embodiment, when the entire exercise section of the exercise is completed, the processor may update the model parameters of the neural network model based on the exercise performance index measured based on sensor data collected during the exercise and the estimated exercise performance index output from the neural network model. Training of the neural network model may be performed based on the sensor data collected during the exercise, the torque parameters applied during the exercise, and feature data related to the exercise.

[0144]

[0145] FIG. 12 is a diagram illustrating an operation of determining a torque parameter related to the achievement of an exercise goal using a neural network model according to various embodiments.

[0146] Referring to FIG. 12, a neural network model (1200) (e.g., the neural network model (830) of FIG. 8 and the neural network model (900) of FIG. 9) may output an estimated exercise performance index of a user as output data when input data is input. The input data of the neural network model (900) may include feature data related to exercise and torque parameters representing the operating characteristics of the wearable device (100).

[0147] A processor of an electronic device (210) (e.g., processor (710) of FIG. 7) may perform the following operations to search for torque parameters that cause a neural network model (1200) to output an estimated exercise performance index that is identical or similar to a target exercise performance index indicated in the user's exercise goal data.

[0148] (1) The processor can estimate exercise-related feature data for the next exercise segment. In one embodiment, the processor can set the exercise-related feature data for the next exercise segment with an expected exercise session timestamp, an exercise segment timestamp, and an exercise segment index for the next exercise segment.

[0149] (2) The processor can set the torque parameter in the next exercise section to be input into the neural network model (1200) to a torque parameter corresponding to a reference value. For example, the torque parameter to be input into the neural network model (1200) can be set to a value of '0'. Within one exercise section, the torque parameter can be set to be the same over time.

[0150] (3) The processor inputs the feature data related to the exercise set in (1) and the torque parameter set in (2) into the neural network model (1200), and can obtain the estimated exercise performance index (e.g., maximum heart rate, walking speed, stride length) as output data of the neural network model (1200). The neural network model (1200) can estimate the exercise performance index through feed-forward computation.

[0151] (4) The processor sets exercise target data based on the user's exercise goals. The exercise target data may include, for example, a target maximum heart rate, a target walking speed, and / or a target stride.

[0152] (5) The processor can calculate the difference between the estimated exercise performance index obtained from the output data of the neural network model (1200) and the exercise target index indicated in the exercise target data as a loss of machine learning. For example, the difference between the maximum heart rate estimated by the neural network model (1200) and the user's target maximum heart rate, the difference between the walking speed estimated by the neural network model (1200) and the user's target walking speed, and the difference between the stride length estimated by the neural network model (1200) and the user's target stride length can be calculated as losses.

[0153] (6) The processor can determine whether the loss calculated in (5) is below the threshold. If the loss is not below the threshold, the processor can perform the following operation (7).

[0154] (7) The processor may search for torque parameters that minimize the calculated loss. In one embodiment, the processor may search for torque parameters that minimize the loss using a gradient descent algorithm based on the partial derivative of the loss with respect to the torque parameters. During the search process, the model parameters of the neural network model (1200) are not updated.

[0155] (8) The processor can transmit a control signal including the torque parameter searched in (7) to the wearable device (100) through a communication circuit of the electronic device (210) (e.g., the communication circuit (730) of FIG. 7). The wearable device (100) can control the output of the torque based on the torque parameter included in the control signal.

[0156] (9) The processor can perform the processes (1) to (8) described above again for the next exercise section.

[0157]

[0158] FIG. 13 is a flowchart illustrating operations of a control method for controlling a wearable device based on a user's exercise data according to various embodiments. In one embodiment, at least one of the operations in FIG. 13 may be performed simultaneously or in parallel with another operation, and the order of the operations may be changed. Furthermore, at least one of the operations may be omitted, and another operation may be additionally performed.

[0159] Referring to FIG. 13, in operation (1310), a processor of an electronic device (210) (e.g., a processor (710) of FIG. 7) may load a neural network model (e.g., a neural network model (830) of FIG. 8, a neural network model (900) of FIG. 9, a neural network model (1200) of FIG. 12). If there is a previously learned neural network model, the learned neural network model may be loaded, and if there is no previously learned neural network model, a neural network model operating with default model parameters may be loaded. The previously learned neural network model may be, for example, a neural network model learned based on the user's exercise data according to the learning process described in FIG. 9.

[0160] In operation (1315), the input circuit of the electronic device (210) (e.g., the input circuit (760) of FIG. 7) may receive a user input regarding an exercise goal. The user may set an exercise goal for the current exercise through the user input. With respect to the exercise goal, the user may select an exercise type related to a maximum heart rate. In one embodiment, given the user's age, a reference maximum heart rate corresponding to the user's age may be determined, and various exercise types may be presented based on the determined reference maximum heart rate. For example, a low-intensity exercise may set an exercise goal such that the user's maximum heart rate falls within a heart rate zone corresponding to 50-60% of the reference maximum heart rate, and a weight control exercise may set an exercise goal such that the user's maximum heart rate falls within a heart rate zone corresponding to 60-70% of the reference maximum heart rate. Aerobic exercise can have a target heart rate of 70-80% of the user's maximum heart rate, while anaerobic exercise can have a target heart rate of 80-90% of the user's maximum heart rate. Maximum heart rate exercise can have a target heart rate of 90-100% of the user's maximum heart rate. For example, if a user's maximum heart rate is 220 and the user selects an anaerobic exercise type, the target maximum heart rate can be set between 176 and 198.

[0161] In one embodiment, a user may set exercise goals for a target walking speed and / or a target stride length in relation to a walking exercise. For example, the user may select a target walking speed that is slower than a reference walking speed, a target walking speed corresponding to the reference walking speed, or a target walking speed that is faster than the reference walking speed as the target walking speed. The reference walking speed may correspond to, for example, an average of previously collected measurements of the user's walking speed. The user may also select a target stride length that is narrower than the reference stride length, a target stride length corresponding to the reference stride length, or a target stride length that is wider than the reference stride length as the target stride length. The reference stride length may correspond to, for example, an average of previously collected measurements of the user's stride length.

[0162] In operation (1320), the processor may determine a torque parameter related to the achievement of the exercise goal using a neural network model (e.g., the neural network model (830) of FIG. 8 , the neural network model (900) of FIG. 9 , and the neural network model (1200) of FIG. 12 ). The processor may determine the torque parameter using the neural network model according to a process as described in FIG. 12 . The processor may repeatedly perform an estimation of the torque parameter that can achieve the user's exercise goal using the neural network model.

[0163] In operation (1325), the display circuit of the electronic device (210) (e.g., the display circuit (740) of FIG. 7) may output information regarding the determined torque parameter. The electronic device (210) may estimate a torque parameter that can satisfy the user's exercise goal using a neural network model and then provide feedback of the estimated torque parameter to the user. In one embodiment, the user may directly change the torque parameter through a UI screen if desired.

[0164] At operation (1330), the processor may determine whether a user input including a change input for a torque parameter has been received via the input circuit.

[0165] If it is determined that a user input including the above change input has been received (if 'Yes' in operation (1330)), in operation (1335) the processor can determine the torque parameter changed by the change input as a torque parameter related to the achievement of the exercise goal.

[0166] In operation (1340), the communication circuit of the electronic device (210) (e.g., the communication circuit (730) of FIG. 7) may transmit a control signal including the determined torque parameter to the wearable device (100) so that the wearable device (100) generates torque based on the determined torque parameter under the control of the processor.

[0167] In operation (1345), the processor and the wearable device (100) may start a scheduled exercise segment. If no exercise segment has been performed immediately before, the first exercise segment may start, and if a exercise segment has been performed immediately before, the next exercise segment may start. The wearable device (100) may control the torque output based on the torque parameter received from the electronic device (210). The wearable device (100) may apply the torque parameter received from the electronic device (210) to the torque output of the motor and start the exercise segment.

[0168] In operation (1350), the processor can determine whether the next exercise segment is the last exercise segment.

[0169] If the next exercise segment is determined to be not the last exercise segment ('No' in operation (1350)), the processor may update the parameters of the neural network model in operation (1355). The processor may perform a learning process for the neural network model based on exercise data acquired in the previous exercise segment. For example, the processor may update the parameters of the neural network model for the exercise segment using exercise data collected from one or more exercise segments performed by the user so far among the exercises performed today. The learning process for the neural network model may refer to the learning process described in FIG. 9.

[0170] If the next exercise segment is determined to be the last exercise segment (if 'Yes' in operation (1350)), in operation (1360), the processor may store exercise data collected up to the last exercise segment after the last exercise segment ends. The collected exercise data may be stored in an exercise data database (e.g., exercise data database (820) of FIG. 8).

[0171] In operation (1365), the processor may update the parameters of the neural network model based on the exercise data stored in operation (1360). The processor may train the neural network model based on exercise data collected from all exercise segments performed today. After training the neural network model is completed, the processor may store the model parameters of the trained neural network model. The trained neural network model may be used as the initial neural network model at the start of the next exercise.

[0172] In one embodiment, the neural network model learned in operation (1355) and the neural network model learned in operation (1365) may be different. The neural network model learned in operation (1355) may be a neural network model learned based on exercise data collected during the exercise section, and the neural network model learned in operation (1365) may be a neural network model learned using the user's entire exercise data.

[0173] In various embodiments described in the present disclosure, the operations for determining the torque parameters of the wearable device (100) may be performed by the processor of the wearable device (100) (e.g., the processor (512) of FIG. 5) as well as the electronic device (210).

[0174]

[0175] FIGS. 14A, 14B, and 14C are drawings illustrating a cable-based wearable device and determining a torque parameter for a cable-based wearable device according to various embodiments.

[0176] The wearable devices controlled by the electronic device (210) may be diverse. For example, there may be a hip-type wearable device (100) capable of leg assistance as shown in FIGS. 1, 2, 3, and 4, and there may be a cable-based wearable device (1400) worn on the user's upper body as shown in FIGS. 14a, 14b, and 14c.

[0177] Referring to FIGS. 14A, 14B, and 14C, a wearable device (1400) may be worn on the upper body (torso, waist, arms, wrists, hands, etc.) of a user (110) and may apply an external force to resist the body movement of the user (110). In one embodiment, the wearable device (1400) may apply a resistance force to the body movement of the user (110) by adjusting the tension of cables (1430, 1435) connected to body coupling components (1440, 1445) held by the user (110). The tension of cables (1430, 1435) represents the force (e.g., torque) generated by a motor of the wearable device (1400) applied to cables (1430, 1435).

[0178] In one embodiment, the wearable device (1400) can operate in a motion guide mode that guides the motion of the user (110). In the motion guide mode, the wearable device (1400) can estimate the position of the body coupling component (1440, 1445) of the wearable device (1400) and adjust the tension of the cable (1430, 1435) based on the position of the body coupling component (1440, 1445). The wearable device (1400) can estimate the position of the body coupling component (1440, 1445) as the position of the hand of the user (110) and determine whether the position of the user's hand (or a change in the position of the user's hand over time) corresponds to a desired motion. The body coupling component (1440, 1445) is a component for connecting or fixing the cable (1430, 1435) to a part of the body of the user (110). The body coupling component (1440, 1445) may have a shape such as a handle, a band, a frame, a bracelet, a ring, a string, a glove, etc., but is not limited thereto. In some embodiments, for the convenience of explanation, the body coupling component (1440, 1445) is described as having a shape of a handle, but the scope of the embodiment should not be understood as being limited thereto.

[0179] A desired exercise motion can be defined as a path along which a hand or arm must move, or a specific exercise posture. The wearable device (1400) can induce the user (110) to move the user's hand in the desired exercise motion by adjusting the tension of the cables (1430, 1435) to a low value when the change in the position of the user's hand over time corresponds to the desired exercise motion, and by adjusting the tension of the cables (1430, 1435) to a high value when the change in the position of the user's hand over time does not correspond to the desired exercise motion. A low value of tension causes the user (110) to feel little or no resistance from the cables (1430, 1435), and a high value of tension causes the user (110) to feel relatively great resistance from the cables (1430, 1435). The low and high values ​​of tension are relative relationships, and can indicate a value lower than a specific reference value and a value higher than the reference value, respectively.

[0180] In one embodiment, the wearable device (1400) may operate in a strength training assistance mode to strengthen the muscle strength of the user (110). In the strength training assistance mode, the wearable device (1400) may impede the body movement of the user (110) or provide resistance to the body movement of the user (110) by applying resistance to the body of the user (110) through the cables (1430, 1435). In one embodiment, the wearable device (1400) may provide exercise load to the arm movement of the user (110) to further enhance the exercise effect. The wearable device (1400) may set the resistance of the cables (1430, 1435) to be high when the user (110) moves in a desired exercise motion, and may set the resistance of the cables (1430, 1435) to be low when the user (110) moves in a direction different from the desired exercise motion. Through this, the user (110) can perform strength training with a desirable exercise motion.

[0181] In one embodiment, the wearable device (1400) may operate in a physical ability measurement mode for measuring the physical ability of a user (110). The wearable device (1400) may measure movement information of the user (110) using sensors (e.g., an inertial sensor, a force sensor, an angle sensor) included in the wearable device (1400) while the user (110) is exercising, and may evaluate the physical ability of the user (110) based on the measured movement information. For example, the exercise ability index (e.g., muscle strength, endurance, balance, exercise posture) of the user (110) may be estimated through the movement information of the user (110) measured by the wearable device (1400).

[0182] According to one embodiment, a wearable device (1400) may include a first wearing part (1410), a main body module (1420), a cable (1430, 1435) connected to an actuator of the main body module (1420), a body coupling component (1440, 1445) connected to the cable (1430, 1435) and connected or fixed to a body part (e.g., hand, wrist, arm, finger) of a user (110), and a second wearing part (1450, 1455). The wearable device (1400) may further include a sensor (not shown) that measures a movement of a user's (110) hand and generates movement data corresponding to the movement of the user's hand. Some of these components (e.g., the second wearable portion (1450, 1455)) may be omitted, and other components (e.g., the third wearable portion worn on the upper limb) may also be added to the wearable device (1400).

[0183] The first wearable part (1410) can be worn on the body of the user (110). For example, the first wearable part (1410) can be worn on the upper body of the user (110). The first wearable part (1410) can be in close contact with the upper body of the user (110) and support a portion of the upper body of the user (110). The first wearable part (1410) can close the main body module (1420) to the upper body of the user (110) so that the main body module (1420) fixed to the first wearable part (1410) does not shake due to the movement of the user (110). In one embodiment, the first wearable part (1410) can include an elastic material.

[0184] The body module (1420) may include a motor (not shown) connected to a cable (1430, 1435) for controlling tension applied to the cable (1430, 1435), a processor (not shown) for controlling the motor, and a battery (not shown). In one embodiment, the body module (1420) may guide a desirable exercise motion for the user (110) or assist in muscle training of the user (110) by controlling the magnitude of tension transmitted to the body coupling component (1440, 1445) of the user (110) through the cable (1430, 1435). The body module (1420) may control the tension applied to the cable (1430, 1435) by controlling an actuator based on the position of the hand (or arm) of the user (110). The tension applied to the cables (1430, 1435) can be transmitted to the body coupling components (1440, 1445) held by the user (110). The main body module (1420) can obtain sensor data related to the position of the hand of the user (110) through one or more sensors (not shown) and estimate the position of the hand and the posture of the arm of the user (110) based on the sensor data. The sensor data can include data measured about the movement and / or rotation of the user's hand. The one or more sensors can include, for example, an inertial measurement unit (IMU). The one or more sensors can be included in, for example, the second wearable unit (1450, 1455), the body coupling components (1440, 1445), a third wearable unit (not shown) worn on the upper arm, and / or another wearable device (e.g., a watch-type wearable device). The main body module (1420) and one or more sensors can be connected to each other wired or wirelessly.

[0185] When the first wearable part (1410) is worn on the upper body of the user (110), the main body module (1420) may be located on the back (e.g., back) of the upper body of the user (110), but is not limited thereto. For example, the main body module (1420) may also be located on the front (e.g., chest) of the upper body.

[0186] One end of the cable (1430, 1435) may be connected to a body coupling component (1440, 1445). In one embodiment, the body coupling component (1440, 1445) and the second wearable portion (1450, 1455) may be worn on the hand and wrist of the user (110), respectively. The body coupling component (1440, 1445) may be in a form that can be grasped by the hand of the user (110), but the form of the body coupling component (1440, 1445) is not limited to the described embodiment. The body coupling component (1440, 1445) may have a form that can be fixed to any part of the hand or arm of the user (110). Cables (1430, 1435) are connected to a motor within the main body module (1420) and can have their tension and / or length changed according to the movement of the user's (110) hand (or arm). Cables (1430, 1435) can be elastic cables, inelastic cables, or a combination thereof.

[0187] The second wearable part (1450, 1455) can allow one end of the cable (1430, 1435) to be positioned closely to the body of the user (110) even when the user (110) does not hold the body coupling component (1440, 1445). The second wearable part (1450, 1455) can include, for example, a sensor such as an inertial sensor and a biometric sensor, a communication module (not shown) for communication of the sensor, and a battery (not shown) for power supply. The biometric sensor can sense, for example, the heart rate of the user (110).

[0188] In the illustrated embodiment, the body coupling components (1440, 1445) and the second wearable portion (1450, 1455) are illustrated as having separate shapes. However, depending on the embodiment, the body coupling components (1440, 1445) and the second wearable portion (1450, 1455) may be manufactured to have a single shape (e.g., a shape of a glove) as an integrated body. In the wearable device (1400), the cables (1430, 1435), the body coupling components (1440, 1445), and the second wearable portion (1450, 1455) are included in two each for both arms of the user (110), but may also be included in only one for either arm. In addition, the wearable device (1400) may operate for both arms or for only one arm. For example, if the user (110) exercises only the right arm (or right hand), the user (110) can grip the first body coupling component (1440) and perform the exercise for the right arm based on the resistance transmitted from the first cable (1430).

[0189] A user (110) can wear a wearable device (1400) and train an exercise posture or perform various types of exercise using the wearable device (1400). For example, the user (110) can perform various exercises such as weight training, boxing, Pilates, stretching, or yoga while wearing the wearable device. In one embodiment, the wearable device (1400) can guide the user (110) to naturally move his / her arm in a preferred exercise posture by providing resistance through cables (1430, 1435) when the user (110) moves his / her arm in a posture other than a preferred exercise posture. Alternatively, the wearable device (1400) can provide high resistance when the user moves in a preferred exercise posture to help the user (110) perform muscle strength training in a preferred exercise posture.

[0190] In one embodiment, a processor of the electronic device (210) (e.g., processor (710) of FIG. 7) may control the tension of a cable (1430, 1435) of the wearable device (1400) in a manner similar to that described above. The tension parameter controlling the tension of the cable (1430, 1435) may correspond to, for example, the torque parameter described with reference to the preceding drawings. When the tension parameter is changed, the operating characteristics of the wearable device (1400) may change. For example, as the tension parameter is changed, it may become more difficult or easier for the user (110) to pull the cable (1430, 1435).

[0191] In one embodiment, the processor may determine tension parameters to achieve the user's exercise goals using a neural network model in a similar manner as described in FIG. 12. The neural network model may input feature data related to the user's exercise (e.g., exercise session timestamp, exercise interval timestamp, exercise interval index) and tension parameters for the next exercise segment. The neural network model may estimate the maximum speed of the user (110) while extending the cable (1430, 1435) in the forward direction and output the estimated maximum speed as output data. The processor may calculate the difference between the target maximum speed set by the user when extending the cable and the estimated maximum speed output by the neural network model as a loss and search for tension parameters that minimize the loss. The processor may control the tension of the wearable device (1400) in the next exercise segment based on the searched tension parameters. In the learning process of the neural network model, sensor data can be acquired using a sensor capable of measuring the moving speed of the cable (1430, 1435), and the maximum speed of the cable (1430, 1435) measured through the acquired sensor data can be used as label data to update the model parameters of the neural network model.

[0192]

[0193] The various embodiments of the present disclosure and the terminology used therein are not intended to limit the technical features described in the present disclosure to specific embodiments, but should be understood to include various modifications, equivalents, or substitutes of the embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more of the items, unless the context clearly indicates otherwise. In the present disclosure, each of the phrases "A or B," "at least one of A and B," "at least one of A or B," "A, B, or C," "at least one of A, B, and C," and "at least one of A, B, or C" can include any one of the items listed together in the corresponding phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish the corresponding component from other corresponding components and do not limit the corresponding components in any other respect (e.g., importance or order). When a component (e.g., a first) is referred to as "coupled" or "connected" to another (e.g., a second) component, with or without the terms "functionally" or "communicatively," it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through at least a third component(s).

[0194] The term "module" used in various embodiments of the present disclosure may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integrally formed component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC). Accordingly, each "module" in this specification may include a circuit.

[0195] Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to perform a desired operation or, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, or computer storage medium or device for interpretation by the processing device or for providing instructions or data to the processing device. The software may be distributed across networked computer systems and stored or executed in a distributed manner. The software and data may be stored on a computer-readable recording medium. Various embodiments of the present disclosure may be implemented as software comprising one or more instructions stored on a storage medium that can be read by a machine. For example, a processor of the device may recall at least one of the one or more instructions stored from the storage medium and execute it. This enables the device to operate to perform at least one function in accordance with the recalled at least one instruction. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' only means that the storage medium is a tangible device and does not contain signals (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0196] According to one embodiment, the method according to the embodiments may be provided as a computer program product. The computer program product may be traded between sellers and buyers as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or may be provided through an application store (e.g., Play Store). TM ) or directly between two user devices (e.g., smart phones), online distribution (e.g., downloading or uploading). In the case of online distribution, at least a portion of the computer program product may be at least temporarily stored or temporarily created in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

[0197] According to various embodiments, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to various embodiments, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0198] While this disclosure has been illustrated and described with reference to various embodiments, it will be understood that the various embodiments are illustrative and not limiting. It will be further understood by those skilled in the art that various changes in form and detail may be made without departing from the true spirit and scope of the present disclosure, including the appended claims and their equivalents. Furthermore, it will be understood that any embodiment(s) described herein may be used in conjunction with any other embodiment(s) described herein.

Claims

1. A control method for controlling a wearable device (100) based on a user's exercise data, An action that receives user input regarding exercise goals; An operation of determining a torque parameter related to the achievement of the exercise goal by using a neural network model that inputs feature data related to an exercise performed by a user wearing the wearable device (100) and outputs an estimated exercise performance index; and An operation of transmitting a control signal including the determined torque parameter to the wearable device (100) so that the wearable device (100) generates torque based on the determined torque parameter. A control method including:

2. In paragraph 1, The operation of determining the above torque parameters is: An action to estimate movement-related feature data for the next movement segment; An operation of inputting the torque parameters corresponding to the estimated feature data and reference values for the above next exercise section into the neural network model; and An operation of determining a torque parameter to be applied to the next exercise section based on the estimated exercise performance index obtained from the neural network model and the exercise goal index for achieving the exercise goal. A control method including:

3. In paragraph 2, The operation of determining the above torque parameters is: An operation of searching for the torque parameter that satisfies the set condition based on the difference between the estimated exercise performance index and the exercise target index. A control method including:

4. In paragraph 3, The operation of transmitting the above control signal is as follows: In response to the torque parameter satisfying the above-described set condition being searched, an operation of transmitting a control signal including the searched torque parameter to the wearable device (100) so that the wearable device (100) generates torque based on the searched torque parameter in the next exercise section. A control method including:

5. In any one of paragraphs 1 to 4, The exercise performed by the above user includes multiple exercise sections, The above control method is, When the first exercise section is completed, an operation of training the neural network model based on the data collected up to the first exercise section; and An operation of determining the torque parameters to be applied to the second exercise section, which is the next exercise section after the first exercise section, using the above-mentioned learned neural network model. A control method further comprising:

6. In any one of paragraphs 1 to 5, The operation of determining the above torque parameters is: An operation of determining at least one of a first parameter that determines the intensity of torque output from a motor (534; 534-1) of the wearable device (100) and a second parameter that determines the point in time at which the torque is applied. A control method including:

7. In any one of paragraphs 1 to 6, The action of receiving the above user input is: An operation of receiving user input including selection of at least one of an exercise type, a target walking speed, and a target stride length associated with a target maximum heart rate. A control method including:

8. In any one of paragraphs 1 to 7, An operation for outputting information about the determined torque parameters; and In response to receiving a user input including a change input for the above torque parameter, an operation of determining the torque parameter changed by the change input as a torque parameter related to the achievement of the exercise goal. A control method further comprising:

9. In any one of paragraphs 1 to 8, When the entire exercise section of the above exercise is completed, an operation of training the neural network model based on data collected during the performance of the above exercise. A control method further comprising:

10. In paragraph 9, The operation of training the above neural network model is as follows: An operation of updating model parameters of the neural network model based on the exercise performance index measured based on sensor data collected during the execution of the exercise and the estimated exercise performance index output from the neural network model. A control method including:

11. In any one of paragraphs 1 to 10, The feature data related to the above exercise includes at least one of a feature value for an exercise section performed by the user, a feature value for the elapsed exercise time, and a feature value for whether a torque parameter has been changed by user input. The above estimated exercise performance indicators are: including at least one of the user's heart rate, walking speed and stride length, Control method.

12. A computer-readable recording medium having recorded thereon instructions that, when executed by one or more processors (710), cause the one or more processors (710) to perform the method of any one of claims 1 to 11.

13. In the electronic device (210), One or more processors (710); an input circuit (760) for receiving user input; and Communication circuit (730) for communicating with a wearable device (100) Including, The above one or more processors (710) In response to receiving user input regarding exercise goals through the above input circuit (760), A torque parameter related to the achievement of the exercise goal is determined using a neural network model that inputs feature data related to the exercise performed by the user wearing the wearable device (100) and outputs an estimated exercise performance index. The wearable device (100) is controlled to generate torque based on the determined torque parameter by causing the communication circuit (730) to transmit a control signal including the determined torque parameter to the wearable device (100). Electronic device (210).

14. In paragraph 13, The above one or more processors (710) Estimate exercise-related feature data for the next exercise segment, The torque parameters corresponding to the estimated feature data and reference values for the above next exercise section are input into the neural network model, Determining the torque parameter to be applied to the next exercise section based on the estimated exercise performance index obtained from the neural network model and the exercise goal index for achieving the exercise goal. Electronic device (210).

15. In paragraph 13 or 14, The exercise performed by the above user includes multiple exercise sections, The above one or more processors (710) When the first exercise section is completed, the neural network model is trained based on the data collected up to the first exercise section, Using the above-mentioned learned neural network model, the torque parameter to be applied to the second exercise section, which is the next exercise section of the first exercise section, is determined. Electronic device (210).

Images