I work on making robots that are safe, simple, and work together. Most of my projects ask: how little can a robot know or carry, and still do something useful? I build the controllers, prove they work, and test them on real hardware.
I am a postdoctoral research associate with Prof. Naomi Leonard at Princeton University. I obtained my PhD in robotics at Cornell University advised by Prof. Hadas Kress-Gazit. Prior to my doctoral studies, I earned my Bachelor's and Master's degrees in Aerospace Engineering from IIT Bombay.
I have collaborated with the Cohen Group, Apsel Lab, and Laboratory for Molecular Engineering to develop autonomous micron-scale origami robots active at time and length scales comparable to biological microorganisms.
Outside research, I enjoy running and hiking, and I am an avid reader with a particular interest in world affairs, human psychology, and philosophy of science.
I want robots that work in the real world: unpredictable environments, cheap sensors, no cloud connection. My approach is to design controllers with formal correctness guarantees, so the math tells you it works before you ever run it on hardware. The projects below span swarms, UAVs, manipulation, and bio-inspired micro-robots, but they share the same thread: doing more with less. Research statement →
Safe motion planning for resource-constrained UAVs using Reference Governor — validated on a Crazyflie in real-time.
Correct-by-construction swarm controllers for target encapsulation using only local noisy sensors — no communication, no localization.
Locomotion synthesis for electrically programmable micron-scale origami robots that fold, reconfigure, and swim in biocompatible solutions.
Provably correct task-and-motion planner for robot manipulation that satisfies complex temporal logic specifications.
Distributed consensus algorithm that achieves agreement despite persistent sensor biases with unknown bounds — no centralized coordinator needed.
Line-of-sight guidance law for autonomous spacecraft rendezvous and docking with Lyapunov-based safety guarantees.