公开(公告)号：US20250018560A1

公开(公告)日：2025-01-16

申请号：US18610285

申请日：2024-03-20

Applicant: KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Hyun MYUNG ,  I Made Aswin NAHRENDRA ,  Byeongho YU

IPC: B25J9/16 ,  B25J13/08 ,  B62D57/032

Abstract: Disclosed is technology for controlling deep reinforcement learning-based legged robot locomotion by inferring implicit terrain information. A legged robot control method may include inferring an action of a quadrupedal robot from proprioception through a deep reinforcement learning-legged robot model, and a locomotion policy that implicitly infers properties of terrains through which the quadrupedal robot moves may be learned in the legged robot model.

公开(公告)号：US20230368016A1

公开(公告)日：2023-11-16

申请号：US17973408

申请日：2022-10-25

Applicant: Korea Advanced Institute of Science and Technology

Inventor： Hyun MYUNG ,  Sungwon HWANG

IPC: G06N3/08 ,  G06V10/82

CPC classification number: G06N3/08 ,  G06V10/82

Abstract: Disclosed are a method and apparatus for learning equivariant and invariant representations for rotation of an image based on a graph convolutional network. The method of learning an equivariant and invariant representation for rotation of an image based on a graph convolutional network performed by a computer device includes learning an equivariant representation for rotation of an image by using a self-weighted nearest neighbors graph convolutional network (SWN-GCN); and finally obtaining the equivariant representation for the rotation of the image obtained from the self-weighted nearest neighbors graph convolutional network as an invariant representation of the rotation of the image by using permutation invariance of global average pooling (GAP).