Learning robotic reactive behavior from human demonstration via dynamic behavior trees

Investigation of a new method for robot learning from demonstration problems, which will model the human demonstrations using dynamic behavior trees. This new paradigm would allow a robot to reason not only in terms of task accomplishment, but also in terms of sensory motor couplings to implement behaviors at runtime; resulting in increase of the robustness of the robot

Thumb ticker sm portrait 3
Max Planck Institute for Intelligent Systems