Leonard Jung

EXP 720

815 Columbus Ave

Boston, MA 02115

Hello World!

I’m a current PhD Candidate at Northeastern University, where I am lucky enough to be advised by Michael Everett in the NAIL Lab. I generally am interested in

Safe-ifying our current robots, through both neural network verification and control-theoretic tools
Scalable, real-time algorithms for long-term/multi-robot SLAM
Fun/cool optimization techniques (especially utilizing structure/geometry)

I also am drawn to anything with very pretty math (current obsession: Riemannian Optimization)!

Previously, I earned my B.S. in Aerospace Engineering from Purdue University, where, I among other things, was fortunate enough to work in the Autonomy and Optimization LaB under Ran Dai working on solving QCQPs for optimal planning and controls and the design of origami-inspired robots.

Feel free to drop me an email; I advise undergrad/masters students regularly, and additionally get nerd-sniped very easily!

news

Dec 11, 2025	My labmate Alenna Spiro presented our co-authored work on Learning Koopman Operators at CDC 2025! Thanks Alenna and Alex!
Jun 05, 2025	I presented my work on Contingency Constrained MPPI at L4DC 2025!

selected publications

Practical and Performant Enhancements for Maximization of Algebraic Connectivity

Leonard Jung, Alan Papalia, Kevin Doherty, and 1 more author

arXiv preprint arXiv:2511.08694, 2025

arXiv
Verification of Visual Controllers via Compositional Geometric Transformations

Alexander Estornell, Leonard Jung, and Michael Everett

arXiv preprint arXiv:2507.04523, 2025

arXiv
A Hybrid Framework for Efficient Koopman Operator Learning

Alexander Estornell^*, Leonard Jung^*, Alenna Spiro^*, and 2 more authors

arXiv preprint arXiv:2504.18676, 2025

arXiv
Contingency Constrained Planning with MPPI within MPPI

Leonard Jung, Alexander Estornell, and Michael Everett

In Proceedings of the 7th Annual Learning for Dynamics & Control Conference, 04–06 jun 2025

Abs arXiv PDF

For safety, autonomous systems must be able to consider sudden changes and enact contingency plans appropriately. State-of-the-art methods currently find trajectories that balance between nominal and contingency behavior, or plan for a singular contingency plan; however, this does not guarantee that the resulting plan is safe for all time. To address this research gap, this paper presents Contingency-MPPI, a data-driven optimization-based strategy that embeds contingency planning inside a nominal planner. By learning to approximate the optimal contingency-constrained control sequence with adaptive importance sampling, the proposed method’s sampling efficiency is further improved with initializations from a lightweight path planner and trajectory optimizer. Finally, we present simulated and hardware experiments demonstrating our algorithm generating nominal and contingency plans in real time on a mobile robot.