Shahbaz Abdul Khader

I am a postdoctoral researcher at the Robotics, Perception, and Learning lab, where I work with Prof. Danica Kragic. I am interested in applying machine learning for solving control and decision making problems. My research is centered around how robots can autonomously acquire complex manipulation skills in uncertain and unstructured environments.

One of the main challenges in robotic manipulation is the possibility of physical contact. I believe that managing robot-environment physical interaction is an integral part of robotic manipulation, without which it is hard for robots to be productive in general settings. This aspect is often called "contact-rich" or "last inch" manipulation. My recent research addresses the exciting problem of provable safety guarantees in deep reinforcement learning (RL) when applied to contact-rich manipulation.

My work has so far explored imparting control stability (Lyapunov sense) in deep RL, structuring deep policies based on physics-based prior, and learning data-efficient models of contact dynamics. See the portfolio pages for more information.

My PhD thesis can be accessed here.