My research focuses on the design and analysis of algorithms for efficient learning and planning in sequential decision-making problems under uncertainty. Most of my work is motivated by real-world problems (e.g., navigation of planetary rovers in unknown environments, optimal routing for resource-constrained systems in stochastic networks) and involves using mathematical tools from the areas of control theory, optimization, motion planning, and machine learning.
While my recent work is theoretical at its core, often involving providing formal guarantees on performance and safety and validation using simulations, I strongly believe that deployment on real-world systems is essential to ensure the theory is honest, relevant, and to assess its practical utility and hidden flaws. I enjoy working with hardware and am planning to spend more time with physical systems during my PhD. I am in the processing of updating this space to add details about my recent projects, hardware experience, and a list of publications. Stay tuned for more updates!