Invention Grant
- Patent Title: Determining environment-conditioned action sequences for robotic tasks
-
Application No.: US17642325Application Date: 2020-09-09
-
Publication No.: US12134199B2Publication Date: 2024-11-05
- Inventor: Soeren Pirk , Seyed Mohammad Khansari Zadeh , Karol Hausman , Alexander Toshev
- Applicant: GOOGLE LLC
- Applicant Address: US CA Mountain View
- Assignee: GOOGLE LLC
- Current Assignee: GOOGLE LLC
- Current Assignee Address: US CA Mountain View
- Agency: Gray Ice Higdon
- International Application: PCT/US2020/049851 WO 20200909
- International Announcement: WO2021/050488 WO 20210318
- Main IPC: B25J9/16
- IPC: B25J9/16 ; G06N3/045 ; G06V10/147 ; G06V10/44 ; G06V10/82 ; G06V20/10 ; G06V20/17 ; G06V20/13

Abstract:
Training and/or using a machine learning model for performing robotic tasks is disclosed herein. In many implementations, an environment-conditioned action sequence prediction model is used to determine a set of actions as well as a corresponding particular order for the actions for the robot to perform to complete the task. In many implementations, each action in the set of actions has a corresponding action network used to control the robot in performing the action.
Public/Granted literature
- US20220331962A1 DETERMINING ENVIRONMENT-CONDITIONED ACTION SEQUENCES FOR ROBOTIC TASKS Public/Granted day:2022-10-20
Information query
IPC分类: