Who I am:
ICRA Workshop: We are organizing a workshop on interactive robot learning at ICRA 2020. Please consider contributing! More details here.
[CLOSED!] Spring 2020: I am looking for motivated M.S. students interested in robot learning and multi-agent systems! More details here.
Fall 2019: I am teaching Georgia Tech’s CS 3630: Introduction to Robotics and Perception (syllabus).
What I do:
I spend most of my time trying to trick people (and sometimes myself!) into thinking robots are smart. To aid this illusion, I study the areas of robot learning, human-robot interaction, and multi-agent systems. I aim to build collaborative systems that can learn to effectively operate in dynamic and unstructured environments, while leveraging and enhancing their interactions with humans.
I received my Ph.D. from the University of Connecticut (UConn) in 2018, where I worked with Ashwin Dani and was a Graduate Fellow of the UTC Institute for Advanced Systems Engineering (UTC-IASE). As part of my dissertation, I developed computational methods for (i) learning robot movement primitives from demonstrations with strong theoretical guarantees, and (ii) probabilistic inference of the intentions of human partners during close-proximity physical collaboration. I received my M.S. from the Department of Electrical and Computer Engineering at the University of Florida, and my B.E. in Instrumentation and Control Engineering from Anna University, Chennai, India.
In other news…
- Our paper on a framework for geometrically-consistent learning and combination of sub-task policies is accepted for presentation at the Conference on Robot Learning (CoRL) 2019. [paper]
- We won the FetchIt! mobile manipulation challenge at ICRA (May 2019). [link]
- Traveling to ICRA to present our paper on multi-coordinate cost balancing for learning from demonstration (May 2019). [paper] [preprint]
- I received the college of computing’s Outstanding Post-Doctoral Research Award at Georgia Tech (April 2019).
- I will be in Austin, TX. during Jan 15-18, 2019, to present our recent work on multi-agent task assignment at the Army Science Planning and Strategy Meeting.