RBE Colloquium Series Presents
Dr. Reza Ahmadzadeh
Trajectory Learning from Demonstration – Two New Approaches
Friday, March 29, 2019
2:00 p.m. - 3:00 p.m.
60 Gateway, Rm 1002
Abstract: Learning from demonstration (LfD) is a paradigm that aims to equip robots with the ability to learn efficiently from demonstrations provided by humans. This presentation focuses on the subject of trajectory learning from human-provided demonstrations by introducing two new approach. The first part of the talk discusses a novel geometric representation and illustrates how this representation can be used for encoding, reproduction, generalization, and refinement of trajectory-based skills as well as dealing with obstacles. The second part of the talk discusses a novel LfD approach that encodes trajectory-based skills taking into account multiple coordinates and illustrates the effect of such consideration on multiple skills as well as compares the results with the cases when only one coordinate has been used.
Bio: Reza Ahmadzadeh is an Assistant Professor with the Department of Computer Science at the University of Massachusetts Lowell. Previously, Reza was a Postdoctoral Research Fellow at Georgia Tech and prior to joining Georgia Tech, he was a Postdoctoral Research Fellow at Italian Institute of Technology. Reza received his Ph.D. in "Robotics, Cognition and Interaction Technologies" from the University of Genoa in collaboration with the Department of Advanced Robotics at Italian Institute of Technology (IIT). His research focuses on developing machine learning algorithms and their application to robot autonomy and physical Human-Robot Interaction.