公开(公告)号：US20240017405A1

公开(公告)日：2024-01-18

申请号：US18222858

申请日：2023-07-17

Applicant: GOOGLE LLC

Inventor： Alexander Toshev ,  Fereshteh Sadeghi ,  Sergey Levine

IPC: B25J9/16 ,  G05B13/02 ,  G06N3/084 ,  G06N3/044 ,  G06N3/045

CPC classification number: B25J9/163 ,  B25J9/1697 ,  G05B13/027 ,  G06N3/084 ,  G06N3/044 ,  G06N3/045 ,  G05B2219/33056 ,  G05B2219/39391 ,  G05B2219/40499 ,  G05B2219/42152

Abstract: Training and/or using a recurrent neural network model for visual servoing of an end effector of a robot. In visual servoing, the model can be utilized to generate, at each of a plurality of time steps, an action prediction that represents a prediction of how the end effector should be moved to cause the end effector to move toward a target object. The model can be viewpoint invariant in that it can be utilized across a variety of robots having vision components at a variety of viewpoints and/or can be utilized for a single robot even when a viewpoint, of a vision component of the robot, is drastically altered. Moreover, the model can be trained based on a large quantity of simulated data that is based on simulator(s) performing simulated episode(s) in view of the model. One or more portions of the model can be further trained based on a relatively smaller quantity of real training data.

公开(公告)号：US20230281422A1

公开(公告)日：2023-09-07

申请号：US18140366

申请日：2023-04-27

Applicant: GOOGLE LLC

Inventor： Krishna Shankar ,  Nicolas Hudson ,  Alexander Toshev

IPC: G06N3/008 ,  G06N3/04 ,  G06N3/08 ,  B25J9/16 ,  G06N3/084 ,  G06F18/40 ,  G06V10/778 ,  G06V10/82

CPC classification number: G06N3/008 ,  B25J9/1605 ,  B25J9/161 ,  B25J9/1671 ,  G06F18/41 ,  G06N3/04 ,  G06N3/08 ,  G06N3/084 ,  G06V10/7788 ,  G06V10/82 ,  Y10S901/03

Abstract: Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

公开(公告)号：US20240094736A1

公开(公告)日：2024-03-21

申请号：US18240124

申请日：2023-08-30

Applicant: GOOGLE LLC

Inventor： Catie Cuan ,  Tsang-Wei Lee ,  Anthony G. Francis, JR. ,  Alexander Toshev ,  Soeren Pirk

IPC: G05D1/02 ,  G06V10/80 ,  G06V10/82 ,  G06V40/20

CPC classification number: G05D1/0221 ,  G05D1/0248 ,  G05D1/0274 ,  G06V10/80 ,  G06V10/82 ,  G06V40/28

Abstract: Training and/or utilizing a high-level neural network (NN) model, such as a sequential NN model. The high-level NN model, when trained, can be used to process a sequence of consecutive state data instances (e.g., N most recent, including a current state date instance) to generate a sequence of outputs that indicate a sequence of position deltas. The sequence of position deltas can be used to generate an intermediate target position for navigation and, optionally, an intermediate target orientation that corresponds to the intermediate target position. The intermediate target position and, optionally, the intermediate target orientation, can be provided to a low-level navigation policy, such as an MPC policy, and used by the low-level navigation policy as its goal position (and optionally goal orientation) for a plurality of iterations (e.g., until a new intermediate target position (and optionally new target orientation) is generated using the high-level NN model.

公开(公告)号：US20250061302A1

公开(公告)日：2025-02-20

申请号：US18939168

申请日：2024-11-06

Applicant: GOOGLE LLC

Inventor： Krishna Shankar ,  Nicolas Hudson ,  Alexander Toshev

IPC: G06N3/008 ,  B25J9/16 ,  G06F18/40 ,  G06N3/04 ,  G06N3/08 ,  G06N3/084 ,  G06V10/778 ,  G06V10/82

Abstract: Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

公开(公告)号：US12159210B2

公开(公告)日：2024-12-03

申请号：US18140366

申请日：2023-04-27

Applicant: GOOGLE LLC

Inventor： Krishna Shankar ,  Nicolas Hudson ,  Alexander Toshev

IPC: G06N3/008 ,  B25J9/16 ,  G06F18/40 ,  G06N3/04 ,  G06N3/08 ,  G06N3/084 ,  G06V10/778 ,  G06V10/82

Abstract: Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

公开(公告)号：US20220331962A1

公开(公告)日：2022-10-20

申请号：US17642325

申请日：2020-09-09

Applicant: GOOGLE LLC

Inventor： Soeren Pirk ,  Seyed Mohammad Khansari Zadeh ,  Karol Hausman ,  Alexander Toshev

IPC: B25J9/16

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.

公开(公告)号：US20210397195A1

公开(公告)日：2021-12-23

申请号：US17291540

申请日：2019-11-27

Applicant: GOOGLE LLC

Inventor： Alexander Toshev ,  Marek Fiser ,  Ayzaan Wahid

IPC: G05D1/02

Abstract: Training and/or using both a high-level policy model and a low-level policy model for mobile robot navigation. High-level output generated using the high-level policy model at each iteration indicates a corresponding high-level action for robot movement in navigating to the navigation target. The low-level output generated at each iteration is based on the determined corresponding high-level action for that iteration, and is based on observation(s) for that iteration. The low-level policy model is trained to generate low-level output that defines low-level action(s) that define robot movement more granularly than the high-level action—and to generate low-level action(s) that avoid obstacles and/or that are efficient (e.g., distance and/or time efficiency).

公开(公告)号：US20250058475A1

公开(公告)日：2025-02-20

申请号：US18936720

申请日：2024-11-04

Applicant: GOOGLE LLC

Inventor： Soeren Pirk ,  Seyed Mohammad Khansari Zadeh ,  Karol Hausman ,  Alexander Toshev

IPC: B25J9/16 ,  G06N3/045 ,  G06V10/147 ,  G06V10/44 ,  G06V10/82 ,  G06V20/10 ,  G06V20/13 ,  G06V20/17

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.

公开(公告)号：US12134199B2

公开(公告)日：2024-11-05

申请号：US17642325

申请日：2020-09-09

Applicant: GOOGLE LLC

Inventor： Soeren Pirk ,  Seyed Mohammad Khansari Zadeh ,  Karol Hausman ,  Alexander Toshev

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.

公开(公告)号：US12061481B2

公开(公告)日：2024-08-13

申请号：US17291540

申请日：2019-11-27

Applicant: GOOGLE LLC

Inventor： Alexander Toshev ,  Marek Fiser ,  Ayzaan Wahid

IPC: G05D1/02 ,  G05D1/00

CPC classification number: G05D1/0221

Abstract: Training and/or using both a high-level policy model and a low-level policy model for mobile robot navigation. High-level output generated using the high-level policy model at each iteration indicates a corresponding high-level action for robot movement in navigating to the navigation target. The low-level output generated at each iteration is based on the determined corresponding high-level action for that iteration, and is based on observation(s) for that iteration. The low-level policy model is trained to generate low-level output that defines low-level action(s) that define robot movement more granularly than the high-level action—and to generate low-level action(s) that avoid obstacles and/or that are efficient (e.g., distance and/or time efficiency).