Research Statement

Research, for me, is a way of learning with consequence. I am drawn to questions that ask how a system works, where its assumptions begin to fail, and what we can learn by building it. Robotics gives these questions a physical setting. A model that is persuasive on paper must eventually meet imperfect measurements, uncertain dynamics, and a world that does not organize itself around our expectations.

I want my work to connect mathematical understanding with the judgment that comes from implementation and experimentation. I move between formulation, simulation, software, and physical systems because each reveals something the others can miss. Across these settings, I try to examine failure carefully and keep my claims proportional to the evidence.

At UCLA's Sensing & Robotics for Infrastructure Lab, my current research focuses on perception and planning for marine infrastructure inspection. Working with the Heron surface vehicle connects questions in state estimation and autonomy to the practical challenges of integrating LiDAR, cameras, inertial sensing, and sonar. Field testing keeps those questions close to the physical world. Through ORACLE, I also study how inspection requests can guide plans grounded in what a robot can observe.

My earlier research examined adaptive control for orbital rendezvous. I also studied coordinated spacecraft assembly under Dwaipayan Mukherjee's supervision. These projects brought together uncertainty modeling, optimization, allocation, and control, and led to conference papers and public simulation code. They continue to shape my interest in how robots can reason and act when models, measurements, and coordination are imperfect.

I think of my role as a translator and integrator. A useful idea can sit between control theory, sensing, software, and the practical knowledge of the people who use a system. Bringing those perspectives together requires understanding their assumptions and making the connections clear enough to test. I value the conversations that change how I understand a problem, as much as the tools that help me solve it.

As artificial intelligence becomes more capable, I find myself asking what I want my work as a researcher to stand for. Competing task by task with increasingly capable models does not seem to me a useful measure of a researcher's worth. Nor would I want to define my contribution by what AI cannot do today. I do not know where these capabilities will lead, and I want my approach to remain open to that uncertainty.

In the age of artificial intelligence, I want to be an “orchestrator of innovation”: someone who chooses worthwhile questions, connects ideas and people, and brings tools together with a clear sense of purpose. For me, that direction grows out of curiosity, lived experience, and attention to the people a system will affect. It means staying intellectually engaged in the work, whether I am developing an idea myself or working with an AI system.

That responsibility does not end when a tool produces an answer. I still need to understand the assumptions, test the result, and decide what the evidence justifies. I want to cultivate the judgment to know when AI helps, when another approach is more appropriate, and when the question itself needs to change.

My goal in research is to make sure that what I build helps answer a meaningful question and that others can understand and extend the work. I hope to contribute methods and systems that make difficult problems more tractable, while staying attentive to the people and communities those systems are intended to serve. That is the standard I want my research to meet.