The goal of the Shared Autonomy for Teleoperation project is to help robots quickly learn from teleoperator controls what an operator would like the robot to do. After a robot determines the operator's goal, it can autonomously work toward completing it. The researchers approached this project by adding active learning components to robot movement planning to help the robot infer operator goals more efficiently from subsequent operator inputs.
- Connor Brooks and Dan Szafir. 2019. Balanced Information Gathering and Goal-Oriented Actions in Shared Autonomy. In Proceedings of the 2019 ACM/IEEE International Conference on Human-Robot Interaction (HRI '19). (Daegu, South Korea – March 11-14, 2019).