Invention Grant
- Patent Title: Deep reinforcement learning for robotic manipulation
-
Application No.: US17878186Application Date: 2022-08-01
-
Publication No.: US11897133B2Publication Date: 2024-02-13
- Inventor: Sergey Levine , Ethan Holly , Shixiang Gu , Timothy Lillicrap
- 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
- Main IPC: G06F17/00
- IPC: G06F17/00 ; B25J9/16 ; G05B13/02 ; G06N3/08 ; G06N3/008 ; G06N3/045 ; G05B19/042

Abstract:
Implementations utilize deep reinforcement learning to train a policy neural network that parameterizes a policy for determining a robotic action based on a current state. Some of those implementations collect experience data from multiple robots that operate simultaneously. Each robot generates instances of experience data during iterative performance of episodes that are each explorations of performing a task, and that are each guided based on the policy network and the current policy parameters for the policy network during the episode. The collected experience data is generated during the episodes and is used to train the policy network by iteratively updating policy parameters of the policy network based on a batch of collected experience data. Further, prior to performance of each of a plurality of episodes performed by the robots, the current updated policy parameters can be provided (or retrieved) for utilization in performance of the episode.
Public/Granted literature
- US11845183B2 Deep reinforcement learning for robotic manipulation Public/Granted day:2023-12-19
Information query