公开(公告)号：US20220204038A1

公开(公告)日：2022-06-30

申请号：US17502769

申请日：2021-10-15

Applicant: Electronics and Telecommunications Research Institute

Inventor： SEONGHUN SEO ,  Eun Hye KIM

IPC: B60W60/00 ,  G06K9/62 ,  G06K9/00 ,  G08G1/16 ,  G01S17/931 ,  G01S17/89 ,  G01P15/18

Abstract: A method and system for detecting a collision by an autonomous robot based on a multi-sensor long short-term memory (LSTM) are disclosed. The method includes generating an input of an LSTM model by combining an input image received from an autonomous robot, light detection and ranging (LiDAR) distance information, and acceleration information, learning a collision alert situation by inputting the input to the LSTM model, and determining a collision situation using an output of the LSTM model and a fully connected neural network (FNN) model.

公开(公告)号：US20240013539A1

公开(公告)日：2024-01-11

申请号：US18215925

申请日：2023-06-29

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： SEONGHUN SEO ,  Dae Hee KIM

IPC: G06V20/50 ,  G06V10/774 ,  G06V10/82 ,  G06T7/60 ,  B65G43/00

CPC classification number: G06V20/50 ,  G06V10/774 ,  G06V10/82 ,  G06T7/60 ,  B65G43/00 ,  B65G2203/041

Abstract: Disclosed are an insertion automation method and system based on deep-learning parcel recognition. The insertion automation method includes determining an initial gradient change angular velocity of a tipper based on a total weight of parcels in the tipper, recognizing a loading state of the parcels in the tipper by inputting images of the parcels in the tipper to an object recognition model, and redetermining a gradient change angular velocity and gradient angle of the tipper based on the recognized loading state, in which the loading state of the parcels in the tipper includes at least one of a position, size, and packing material of the parcels in the tipper.