E5, U. Waterloo
Announcement (July 2021): I have joined the University of Waterloo as a Tenure-track Assistant Professor in the ECE department. I am looking to hire (jointly with Sebastian Fischmeister) PhD students to work on problems in safety assurance of autonomous systems. Please send me an email with your CV if you are interested.
My research interests lie in using elements of Control Theory, Formal Methods, Machine Learning and Optimization to solve problems in Safe Autonomy of Robotic and other Cyber-Physical Systems, and Human-Robot Interaction. [Curriculum Vitae]
Prior to joining the University of Waterloo, I was a Postdoctoral Fellow in the EECS Department at UC Berkeley. As a part of the VeHICaL project, I was mentored by Professors Sanjit Seshia (primarily) and Bjoern Hartmann.
I received my PhD in Electrical Engineering from the University of Pennsylvania, where I worked on developing Robust Predictive Methods for Planning and Control of Autonomous Systems as a part of Prof. Rahul Mangharam’s Real-Time and Embedded Systems Lab.
Computationally tractable solutions, that potentially solve real problems, are a recurring theme across my research. More often that not, the algorithms we develop are implemented on actual Cyber-Physical Systems, some of which are: multiple Crazyflie 2.0 quad-rotors, a 1/10th scale autonomous car, a hex-rotor platform, a hardware-in-the-loop vehicle ECU simulator, a scaled down hybrid powertrain for Electric Vehicles, and a self-balancing, fully enclosed electric motorcycle.
|Jun 15, 2021||Our paper on a Signal Temporal Logic-based framework for safety of Autonomous UAVs in Urban Air Mobility missions has been accepted for publication in the Transportation Research Part C: Emerging Technologies, special issue on Embracing Urban Air Mobility.|
|Apr 29, 2021||Our paper on decentralized path planning with line-of-sight constrained communication between agents, led by Victoria Tuck, has been accepted at the IEEE Conference on Control Technology and Applications (CCTA), 2021.|
|Feb 4, 2021||Our paper on Learning-based Decentralized Multi-UAV Collision Avoidance will appear in the ACM Transactions on Cyber-Physical Systems.|
|Feb 1, 2021||Two of our papers have been accepted at the 2021 American Control Conference (ACC): one on co-designing the planner and controller for UAVs with Signal Temporal Logic objectives, and another on Counter-Example Guided Synthesis for systems with black-box learning-based perception modules.|
|Oct 18, 2020||Our paper (led by Kuk Jang) won the best paper in session (UAV conflict management) award at DASC 2020.|
- Fly-by-Logic: Control of Multi-Drone Fleets with Temporal Logic ObjectivesIn Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) 2018
- Anytime Computation and Control for Autonomous SystemsIEEE Transactions on Control Systems Technology 2020
- Learning-’N-Flying: A Learning-based, Decentralized Mission Aware UAS Collision Avoidance SchemeACM Transactions on Cyber-Physical Systems 2021