公开(公告)号：US11170526B2

公开(公告)日：2021-11-09

申请号：US16725672

申请日：2019-12-23

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Tarik Tosun ,  Eric Mitchell ,  Ben Eisner ,  Jinwook Huh ,  Bhoram Lee ,  Daewon Lee ,  Volkan Isler ,  Sebastian Seung ,  Daniel Lee

IPC: G06T7/73 ,  G06T7/38

Abstract: An apparatus for estimating a trajectory of a tool may include: a memory storing instructions; and a processor configured to execute the instructions to: receive a task to be performed by the tool on a target object; receive a grayscale image and a depth image of the target object; and estimate a tool trajectory for performing the task, from the grayscale image and the depth image, via a pixels-to-plans neural network that is trained based on a labeled tool trajectory that is generated from a point cloud model of the target object.

公开(公告)号：US20200311969A1

公开(公告)日：2020-10-01

申请号：US16725672

申请日：2019-12-23

Applicant: Samsung Electronics Co., Ltd.

Inventor： Tarik Tosun ,  Eric Mitchell ,  Ben Eisner ,  Jinwook Huh ,  Bhoram Lee ,  Daewon Lee ,  Volkan Isler ,  Sebastian Seung ,  Daniel Lee

IPC: G06T7/73 ,  G06T7/38

Abstract: An apparatus for estimating a trajectory of a tool may include: a memory storing instructions; and a processor configured to execute the instructions to: receive a task to be performed by the tool on a target object; receive a grayscale image and a depth image of the target object; and estimate a tool trajectory for performing the task, from the grayscale image and the depth image, via a pixels-to-plans neural network that is trained based on a labeled tool trajectory that is generated from a point cloud model of the target object.

公开(公告)号：US11710301B2

公开(公告)日：2023-07-25

申请号：US16749658

申请日：2020-01-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Riley Simmons-Edler ,  Ben Eisner ,  Eric Mitchell ,  Daniel Dongyuel Lee ,  Sebastian Seung

IPC: G06V10/82 ,  G06N5/043 ,  G06N3/08 ,  G06N3/045 ,  G06N3/006 ,  G06V10/764 ,  G06V10/778 ,  G06V10/80

CPC classification number: G06V10/82 ,  G06N3/006 ,  G06N3/045 ,  G06N3/08 ,  G06N5/043 ,  G06V10/764 ,  G06V10/7788 ,  G06V10/811

Abstract: An apparatus for performing continuous actions includes a memory storing instructions, and a processor configured to execute the instructions to obtain a first action of an agent, based on a current state of the agent, using a cross-entropy guided policy (CGP) neural network, and control to perform the obtained first action. The CGP neural network is trained using a cross-entropy method (CEM) policy neural network for obtaining a second action of the agent based on an input state of the agent, and the CEM policy neural network is trained using a CEM and trained separately from the training of the CGP neural network.