Overview of our method (IMAGE)

Journal Center of Harbin Institute of Technology

Overview of our method

Caption

(a) A robot whose kinematic and morphology model is unknown. (b) Taking photos of the robot under different joint configurations from different view directions. (c) Representing each part of the robot with a unique NeRF network and constructing a whole self-model. (d) Control the robot to achieve downstream tasks with the learned self-model.

Credit

The image may only be used with appropriate caption or credit

Usage Restrictions

This file may be viewed by all EurekAlert! users.

License

Original content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.