Google Inc.

Daniel Ho

Yunfei Bai

Eric Jang

Alexander Herzog

Chuyuan Fu

Wenlong Lu

Yu Qing Du

C Karen Liu

Mohi Khansari

Wenhao Yu

Zhuo Xu

Implementations disclosed herein relate to mitigating the reality gap through feature-level domain adaptation in training of a vision-based robotic action machine learning (ML) model. Implementations mitigate the reality gap through utilization of embedding consistency losses and/or action consistency losses during training of the action ML model.

Mitigating reality gap through feature-level domain adaptation in training of vision-based robot action model

Utilization of past dynamics sample(s), that reflect past contact physics information, in training and/or utilizing a neural network model. The neural network model represents a learned value function (e.g., a Q-value function) and that, when trained, can be used in selecting a sequence of robotic actions to implement in robotic manipulation (e.g., pushing) of an object by a robot. In various implementations, a past dynamics sample for an episode of robotic manipulation can include at least two past images from the episode, as well as one or more past force sensor readings that temporally correspond to the past images from the episode.

Utilizing past contact physics in robotic manipulation (e.g., pushing) of an object

Growing community of inventors

Fremont, CA, United States of America

Daniel Ho