Learning to Manipulate from Demonstration (to Escape from a Room)

Photo by Markus Spiske on Unsplash

Principal Investigators:

Oliver Brock

Team members

Xing Li (Doctoral researcher)

Teaching robots to physically operate locks and mechanisms

Research Unit 3, SCIoI Project 28

We will develop a novel approach to Learning from Demonstration for teaching a robot complex, contact-rich manipulation tasks. This approach will enable the robots to physically operate complex locks and other mechanisms. The scientific challenge consists of reliably operating such multi-degree-of-freedom mechanisms that require transitions between different multi-contact situations. Rather than programming these manipulation actions directly, we will have a human demonstrate motions to the robot. We will use our knowledge about the problem and a clever way to elicit demonstration behavior to produce robust and transferable manipulation strategies. These strategies will form an important building block for the one of the SCIoI demonstrators in which a mobile manipulator has to escape from an escape room. To develop and validate the proposed approach to Learning from Demonstration, we will construct escape rooms of increasing mechanical complexity, matching the newly posed challenges to further advances in our approach.

 

Related Publications

Li, X., & Brock, O. (2022). Learning from Demonstration Based On Environmental Constraints. IEEE Robotics and Automation Letters with IROS Option. https://doi.org/10.1109/LRA.2022.3196096