Google Inc.

Georgia Tech Research Corporation

Emory University

Sehoon Ha

Jie Tan

Cheng-Yun Karen Liu

Yun Seong Song

Lena Ting

Peng Xu

Sergey Levine

Aleksandra Faust

Tingnan Zhang

Xinlei Pan

Brian Andrew Ichter

Zhenyu Tan

Techniques are disclosed that enable training a plurality of policy networks, each policy network corresponding to a disparate robotic training task, using a mobile robot in a real world workspace. Various implementations include selecting a training task based on comparing a pose of the mobile robot to at least one parameter of a real world training workspace. For example, the training task can be selected based on the position of a landmark, within the workspace, relative to the pose. For instance, the training task can be selected such that the selected training task moves the mobile robot towards the landmark.

System and methods for training robot policies in the real world

In a method for recycling energy from stairs, a step platform is moveable between an upper and lower position. An energy storage device, coupled to the platform, stores energy when a downward force is applied thereto, causing the step platform to move to the lower position. The energy storage device (also releases stored energy as the step platform moves from to the upper position. A controllable locking mechanism (locks the step platform in the lower position when the downward force has caused the step platform to move into the lower position. A sensor determines when a downward force has been applied to the next higher step platform. A controller signals the controllable locking mechanism to unlock the step platform when the step platform is in the lower position, and when the downward force has been applied to the next higher step platform.

Energy-efficient assistive stairs

A computer-implemented method is disclosed for training one or more machine-learned models. The method can include inputting a first image frame and a second image frame into a feature disentanglement model and receiving, as an output of the machine-learned feature disentanglement model, a state feature and a perspective feature. The method can include inputting the state feature and the perspective feature into a machine-learned decoder model and receiving, as an output of the machine-learned decoder model, the reconstructed image frame. The method can include comparing the reconstructed image frame with a third image frame corresponding with the location and the perspective orientation. The method can include adjusting one or more parameters of the machine-learned feature disentanglement model based on the comparison of the reconstructed image frame and the third image frame.

Systems and methods for training a machine learned model for agent navigation

In a mechanism for recycling energy from stairs (), a step platform () is moveable between an upper and lower position. An energy storage device (), coupled to the platform (), stores energy when a downward force is applied thereto, causing the step platform () to move to the lower position. The energy storage device () also releases stored energy as the step platform () moves from to the upper position. A controllable locking mechanism () locks the step platform () in the lower position when the downward force has caused the step platform () to move into the lower position. A sensor () determines when a downward force has been applied to the next higher step platform (). A controller () signals the controllable locking mechanism () to unlock the step platform () when the step platform () is in the lower position and when the downward force has been applied to the next higher step platform ().

Growing community of inventors

Atlanta, GA, United States of America

Sehoon Ha