METHOD AND APPARATUS FOR EXTRACTING RESULT INFORMATION USING MACHINE LEARNING

    公开(公告)号:US20240331360A1

    公开(公告)日:2024-10-03

    申请号:US18616389

    申请日:2024-03-26

    CPC classification number: G06V10/774 G06V10/7715 G06V10/82 G06V40/10

    Abstract: A method of operating a neural network device for extracting result information using machine learning according to an embodiment of the present disclosure includes: extracting feature data from an image frame; storing a pre-trained first training model generated by performing machine learning on the feature data and including first training data; storing a pre-trained second training model that is generated by performing machine learning on the first training data and includes second training data generated according to the four arithmetic operations based on the first training data; and generating two-dimensional vector information on the result information by performing a probability-based operation on the second training data in the image frame based on the second training model.

    APPARATUS AND METHOD FOR VIRTUAL DEVICE MIRRORING BETWEEN EDGES

    公开(公告)号:US20240320028A1

    公开(公告)日:2024-09-26

    申请号:US18586329

    申请日:2024-02-23

    Abstract: Disclosed herein are an apparatus and method for virtual device mirroring between edges. An edge server includes memory configured to store at least one program and a processor configured to execute the program, wherein the program is configured to create a virtual device based on specification information of an actual device and provide the specification information used to create the virtual device in response to a request received from an additional edge server, or acquire specification information used to create a virtual device of the additional edge server in response to a request of the actual device and create a virtual device based on the acquired specification information, wherein the virtual device synchronizes data with the actual device and the virtual device of the additional edge server in real time.

    Method and apparatus for object following robot using UWB and odometry-based relative position estimation

    公开(公告)号:US12093055B2

    公开(公告)日:2024-09-17

    申请号:US17523501

    申请日:2021-11-10

    Inventor: Jeong Gi Yang

    CPC classification number: G05D1/028 G05D1/027 G05D1/0272

    Abstract: Disclosed are a method and apparatus for enabling a robot to follow an object by using distance measurement data and odometry data through ultra-wideband (UWB) to estimate a relative position of a target in a robot center coordinate system. The method comprises initializing a robot's own object following algorithm according to an object following request; transmitting a single-sided two-way ranging (SS-TWR) poll message; receiving a single response message in response to the SS-TWR poll message from the object, the single response message including second odometry information of a second odometry measurement device of the object; estimating a distance to the object based on a round trip delay calculated by an SS-TWR method; predicting a position of the object based on the second odometry information and first odometry information of the first odometry measurement device; and correcting the position of the object based on the estimated distance.

Patent Agency Ranking