Abstract: A method and apparatus for recognizing a motion of a user are provided. The apparatus may map each of right leg angle information and left leg angle information to context information by comparing each of the right leg angle information and left leg angle information to a comparison value and/or preset threshold, may generate a motion event corresponding to the context information based on a criterion, and may recognize the motion of the user based on the motion event and a previous motion of the user.

Abstract: A balance training method using a wearable device, and the wearable device are disclosed. The balance training method using the wearable device configured to provide a walking assist function includes executing a balance training mode of the wearable device and supplying an irregular pattern torque to an actuator of the wearable device at a time point or in a time period in the balance training mode.

Abstract: A wearable device capable of outputting a torque on a user is provided. The wearable device includes a motor, a motor driver circuit, a communication circuit for receiving movement information about a first user from a server or an electronic device, a frame connected to the motor, and worn on a body part of a second user so as to support the lower body, a sensor, and a processor for controlling the motor driver circuit so that motion information about the second user is obtained using the sensor, the difference between the obtained motion information and the received motion information is calculated, torque strength is determined based on the calculated difference, and a torque of a determined torque strength is output by the motor.

Abstract: A method and device for controlling a walking assist device is disclosed. The method includes determining a first state variable for a gait state of a user wearing the walking assist device based on a gait of the user, obtaining a second state variable which is smoothed and time-delayed from the first state variable, obtaining a third state variable by applying a torque control variable to the second state variable, and determining an assist torque to be provided by the walking assist device based on the obtained third state variable.

Abstract: A tendency of an action of a robot may vary based on learning data used for training. The learning data may be generated by an agent performing an identical or similar task to a task of the robot. An apparatus and method for updating a policy for controlling an action of a robot may update the policy of the robot using a plurality of learning data sets generated by a plurality of heterogeneous agents, such that the robot may appropriately act even in an unpredicted environment.

Abstract: A method and apparatus for controlling an operation of a service robot is disclosed. The method includes measuring, by processing circuitry, an evaluation index of the service robot based on sensor data in a service mode; determining, by the processing circuitry, an operation mode of the service robot from a set of at least two operation modes based on the measured evaluation index; selecting, by the processing circuitry, a behavior to be applied to the operation of the service robot from a set of at least two behaviors based on the operation mode; and controlling, by the processing circuitry, the operation of the service robot based on the behavior.

Abstract: A method and device for controlling a walking assist device is disclosed. The method includes determining a first state variable for a gait state of a user wearing the walking assist device based on a gait of the user, obtaining a second state variable which is smoothed and time-delayed from the first state variable, obtaining a third state variable by applying a torque control variable to the second state variable, and determining an assist torque to be provided by the walking assist device based on the obtained third state variable.

Abstract: Sensor devices closely attached to a body of a user, and walking assist devices such sensor devices may be provided. For example, a sensor device including a sensor configured to sense physical information of the user, and a support configured to provide an elastic force to the sensor such that the sensor closely attached to the body regardless pf a movement of the user, and enable the sensor to sense the physical information of the user with relative accuracy may be provided.

Abstract: An autonomous driving apparatus for accompanied driving in an environment that includes a companion and an obstacle includes a sensor, processing circuitry, and a driver. The sensor may generate sensor data. The processing circuitry may define a current state of the autonomous driving apparatus based on processing the sensor data to determine respective positions of the companion and the obstacle in the environment and select a first tracking point of a plurality of tracking points at least partially surrounding the position of the companion in the environment based on the current state, a position of each tracking point of the plurality of tracking points in the environment defined by the position of the companion in the environment. The driving apparatus drive mechanism may move the autonomous driving apparatus to the first tracking point to cause the autonomous driving apparatus to accompany the companion in the environment.

Abstract: Apparatuses and methods for generating a local motion of a mobile object based on a machine learning are disclosed. The methods may include determining, by processing circuitry, a local window corresponding to a current position of the mobile object in a global area; generating, by processing circuitry, a local cost map indicating a probability of a collision of the mobile object with an obstacle in the local window; generating, by processing circuitry, a local goal map indicating a local path in the local window between the current position and a local goal in the local window; and determining, by processing circuitry, a target velocity of the mobile object based on output data of a neural network, the output data based on an input of input data to the neural network, the input data including the local cost map, the local goal map, and a current velocity of the mobile object.

Abstract: A method and apparatus for controlling a walking assistance apparatus are provided. The apparatus may include a detector configured to detect a first step of a user, based on measured right and left hip joint angle information, a reconstructor unit configured to reconstruct knee joint information matched to the right and left hip joint angle information based on knee joint trajectory information in response to the user's steps, and a torque generator configured to generate a first torque applied to a first leg corresponding to the first step. The torque generator may generate the first torque, based on a second torque applied to a second leg that is opposite to the first leg and that corresponds to a second step preceding the first step.

Abstract: A method and apparatus for controlling a walking assistance apparatus are provided. The apparatus may include a detector configured to detect a first step of a user, based on measured right and left hip joint angle information, a reconstructor unit configured to reconstruct knee joint information matched to the right and left hip joint angle information based on knee joint trajectory information in response to the user's steps, and a torque generator configured to generate a first torque applied to a first leg corresponding to the first step. The torque generator may generate the first torque, based on a second torque applied to a second leg that is opposite to the first leg and that corresponds to a second step preceding the first step.

Abstract: A gait assistance apparatus including at least one sensor configured to sense a step motion of a user, and an operator configured to determine assistance forces of a right step and a left step of the user based on step motion information of the user and adjust the assistance force for at least one of the right step or the left step of the user based on a result of proportionally adding together the assistance forces of the right step and the left step based on a variable ratio may be provided.

Abstract: A gait assistance apparatus including at least one sensor configured to sense a step motion of a user, and an operator configured to determine assistance forces of a right step and a left step of the user based on step motion information of the user and adjust the assistance force for at least one of the right step or the left step of the user based on a result of proportionally adding together the assistance forces of the right step and the left step based on a variable ratio may be provided.

Abstract: An exercise assist method includes receiving a user input signal comprising exercise setting information from an external device; measuring, based on the user input signal, a joint angle of a hip joint of a user wearing an exercise assist device configured to assist a user in performing a leg movement; and controlling torque of the exercise assist device to be applied to a leg of the user based on the measured joint angle and the exercise setting information.

Abstract: A balance training method using a wearable device, and the wearable device are disclosed. The balance training method using the wearable device configured to provide a walking assist function includes executing a balance training mode of the wearable device and supplying an irregular pattern torque to an actuator of the wearable device at a time point or in a time period in the balance training mode.

Abstract: A standing-up assistance method and apparatus may measure a pressure applied to a part of a body of a user, acquire information about a torque corresponding to the measured pressure, and provide an assistance force to the body of the user based on the information about the torque.

Abstract: A walking assistance method may include receiving at least one abnormal gait type selected from a plurality of abnormal gait types, and differently controlling a walking assistance apparatus based on the at least one abnormal gait type.

Abstract: A walking assistance method may include receiving at least one abnormal gait type selected from a plurality of abnormal gait types, and differently controlling a walking assistance apparatus based on the at least one abnormal gait type. Therefore, the walking assistance method may effectively assist various types of abnormal gaits.

Abstract: A method and apparatus for helping a user walk may verify whether the user is in a standing state and calculate a torque that controls balance of the user, and an assistance force for controlling balance may be provided for the user based on the torque generated by an actuator.