University of Birmingham > Talks@bham > IRLab Seminars: Robotics, Computer Vision & AI > Action Representations: from Shared Control to Supervised Autonomy

Action Representations: from Shared Control to Supervised Autonomy

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Martin Rudorfer.

In robotics, the appropriate level of autonomy depends on the requirements of the specific mission and task. Therefore, the Institute of Robotics and Mechatronics investigates a whole spectrum of robotic autonomy, including telepresence, shared control, supervised autonomy, and extended autonomy. In this presentation, I will focus on the action representations that the Department of Cognitive Robotics develops to be able to span this spectrum. First, I will present Action Templates, which provide a hybrid planning framework that is used for both autonomous planning and supervised autonomy. Second, the Shared Control Templates approach allows for seamless transitions between autonomous task completion and shared control with low-dimensional human input commands. Finally, I will argue that we should stop playing “hunt the thimble” in robotic reinforcement learning, and that the action representation above will enable robots to learn efficiently during operation.

This talk is part of the IRLab Seminars: Robotics, Computer Vision & AI series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Talks@bham, University of Birmingham. Contact Us | Help and Documentation | Privacy and Publicity.
talks@bham is based on talks.cam from the University of Cambridge.