I am a researcher in robotics and AI search at University of California, Irvine, working with Professor Sven Koenig in the Intelligent Decision-Making (IDM) group. Before this, I received my Ph.D. from the University of New Hampshire (UNH) in May 2026, where I conducted my research independently while being fortunate and honored to collaborate closely with Professor Oren Salzman of the Technion-Israel Institute of Technology.

My research centers on building algorithmic and theoretical foundations for planning with strong theoretical guarantees on optimality, completeness, robustness, and reliability for robotics and intelligent systems. Rather than treating trade-offs between computational efficiency, solution quality, and theoretical guarantees as inevitable, my research asks: Which computations are truly unnecessary to preserve these rigorous theoretical guarantees?

To answer this question, my research bridges discrete heuristic search and continuous motion planning, focusing on fundamental algorithmic principles that can be applied across different planning paradigms, from single-robot planning to multi-robot and multi-agent systems. My current work spans bidirectional and incremental heuristic search, optimal sampling-based motion planning, any-angle pathfinding, and multi-agent pathfinding, with broader interests in kinodynamic motion planning, task and motion planning, graph-based optimization, and optimal control. I am particularly into extending these algorithmic foundations to high-dimensional, nonholonomic, and kinodynamic systems, as well as dynamic and unknown environments.

Beyond developing individual planning algorithms, my long-term goal is to establish a unified algorithmic framework for efficient, scalable, and provably reliable autonomous decision-making, bridging theoretical insights with real-world applications.

I have also been a contributor to the Open Motion Planning Library (OMPL) since Oct 2025.

Recent Posts

Tighter Bidirectional Heuristic Search and Lazy Optimal Sampling-Based Motion Planning. Ph.D. Dissertation, May 2026, UNH.

Bidirectional Tight Informed Trees (BTIT*), Proceedings of ICCA 2026, 16-20 June, Almaty, Kazkhstan. Invited paper

Bidirectional Lazy Informed Trees (BLIT*), Proceedings of ICRA 2025, 19–23 May, Atlanta, USA. pdf

Meet-in-the-middle with Early and Efficient Termination (MEET), Proceedings of IJCAI 2025, August 16-22, Montreal, Canada.pdf

EMPC-Based Flight Control and Collision-Free Path Planning for A Quadrotor with Unbalanced Payload. Proceedings of IEEE/ASME Transactions on Mechatronics (TMECH) 2025. pdf

Honors and Awards

Special IJCAI 2025 Doctoral Consortium Travel Award, awarded by the National Science Foundation (NSF).

Selected to present a lightning talk at the IJCAI 2025 Doctoral Consortium as one of six lightning talks of selected papers from main conference track.

Open soruce contributions

We have released C++ implementation of BLIT* for comparison-version via github.

BLIT* is already merged into Open Motion Planning Library OMPL.

Highlights of My Work

(1) Bidirectional Lazy Informed Trees (BLIT*):

Oral presentation at ICRA 2025, May 20, Atlanta, GA, USA.

Poster Presentation: BLIT_poster.pdf

(2) Meet-in-the-middle with Early and Efficient Termination (MEET)

Main-Track oral presentation at IJCAI 2025, Aug 22, Montreal, Canada.

Lightning talk to the DC of IJCAI25 at IJCAI 2025, Aug 18, Montreal, Canada.

(3) EMPC-Based Flight Control and Collision-Free Path Planning.

An experimental demonstration video is available here.