About

 

This is Jason Jangho Choi’s personal website.

I am an assistant professor at the UCLA Electrical and Computer Engineering Department, and the principal investigator of the Safety and Collective Intelligence (SCI) Autonomy Lab.

I finished my Ph.D. studies in Mechanical Engineering at UC Berkeley in 2025, where I was co-advised by Professor Koushil Sreenath and Professor Claire Tomlin for my Ph.D. thesis, “Safety guarantees for uncertain dynamical systems—combining model-based analysis with data-driven approaches.” I completed my undergraduate degree in Mechanical Engineering at Seoul National University. Before and during my Ph.D. studies, I had the opportunity to work at various companies advancing robot and aviation autonomy, including Joby Aviation, Naver Labs, and Mars Auto. 

For more details about my lab’s research agenda, please visit my lab webpage. To see my previous research in detail, please see my google scholar.

Please reach out to me by email (jjhchoi at g dot ucla dot edu).

(Pronouns: he/him/his)

Selected Honors & Awards

Robotics: Science and Systems (RSS) Pioneers 2024

Berkeley Fellowship for Graduate Study, Graduate Division, 2019—2021

Kwanjeong Educational Foundation Scholarship, 2019-2025

1st Place Winner, ACM/IEEE Human Robot Interaction Student Design Competition, 2018

News

Nov 2025 I gave a talk at the Global Technology Research (GTR) Division of Samsung Electronics, on “Safety & Collective Intelligence for Next-Generation Robot Autonomy”.

Nov 2025 I gave a talk in the seminar series of the KAIST Cho Chun Shik Graduate School of Mobility, on “Data-Driven Frameworks for Safety Verification and Applications in Advanced Air Mobility”.

Nov 2025 I gave a talk at the Seoul National University GoGE Workshop on Optimization, Decision and AI, on “Data-Driven Safety Frameworks—Indirect vs Direct Approaches”.

Nov 2025 My appointment at UCLA has started! Please visit our lab website for more details.

Jun 2025 I started my visit at Stanford University as a visiting scholar with the Safe and Intelligent Autonomy Lab. 

Jun 2025 I attended the annual meeting of the NASA University Leadership Initiative on Safe Aviation Autonomy at Stanford University. My collaborators and I presented our work on “Data-Driven Hamiltonian for Direct Construction of Safe Set from Trajectory Data” and “Resolving Conflicting Constraints in Multi-Agent Reinforcement Learning with Layered Safety“.

Apr 2025 I gave a talk in the seminar series of the Graduate School of Data Science at Seoul National University, on “AI and Control Theory for Safety in Aviation Autonomy”.

Apr 2025 I visited Safe and Intelligent Autonomy Lab (Professor Somil Bansal’s group) at Stanford University and gave a talk on “Towards Generalizable Safety for Advanced Air Mobility”.

Apr 2025 I successfully finished my Ph.D. dissertation talk! Check out the recording here!

Apr 2025 Our new work, “Resolving Conflicting Constraints in Multi-Agent Reinforcement Learning with Layered Safety“, is accepted to RSS 2025. In this work, we present a new way to inform the multi-agent reinforcement learning policy with safety, combined with the layered safety filter architecture.

Apr 2025 Our new work, “Data-Driven Hamiltonian for Direct Construction of Safe Set from Trajectory Data“, is submitted to IEEE CDC 2025, and is published on Arxiv. This new framework allows data-driven computation of safe sets without the need for explicit learning or characterization of the system’s dynamics. 

Jan 2025 I attended the NSF Workshop on Reinforcement Learning (RL) at Harvard and explored the latest innovations in the field. Professor Andrew Barto, one of the founders of modern RL, pointed out safety assurance for RL as one of the most important future research directions, noting that current RL works by “crash and burn while you adapt and learn”.

Dec 2024 I attended the IEEE CDC in Milan to present my work “Gait Switching and Enhanced Stabilization of Walking Robots with Deep Learning-based Reachability: A Case Study on Two-link Walker“. I also presented my research during the Meet the Faculty Candidates poster session.

Dec 2024 I visited Dufour Aerospace in Zurich, Switzerland to discuss safety frameworks for eVTOL autonomous flight control systems and flight test procedures.

Dec 2024 I visited the Institute for Dynamic Systems and Control (Professor Melanie Zeilinger’s group) at ETH Zurich, and presented on “Towards Generalizable Safety Frameworks for Robot and Aviation Autonomy”.

Nov 2024 I presented on “Effective Abstractions for Data-driven Safety Analysis in Robot and Aviation Autonomy” at the UC Berkeley Mechanical Engineering Control Seminar.

Nov 2024  Our lab (Hybrid Robotics Group) was invited to tour Figure AI‘s headquarters and explore their latest innovations in humanoid robotics.

Oct 2024  I have been announced as the Local Chair for the RSS Pioneers Workshop 2025.

Oct 2024  I presented on “Effective abstractions for direct data-driven safety analysis” at the Bay Area Robotics Symposium 2024.

July 2024  Our paper “Gait Switching and Enhanced Stabilization of Walking Robots with Deep Learning-based Reachability: A Case Study on Two-link Walker” is accepted to the IEEE Conference on Decision and Control (CDC) 2024.

July 2024  I was invited to lecture at the “Safety Filters for Control” tutorial session at the Learning for Dynamics & Control Conference (L4DC) 2024, with Melanie Zeilinger, Kim Wabersich, and Max Cohen.

July 2024  I presented my work, “Safety filters for black-box dynamical systems by learning discriminating hyperplanes” with Will Lavanakul, at Learning for Dynamics & Control Conference (L4DC) 2024 in Oxford.

July 2024  I am very honored to be selected as RSS Pioneers 2024.

June 2024  I conducted visiting research at Cranfield University, hosted by Professors Namhoon Cho and Hyo-Sang Shin. We developed a direct data-driven Hamilton-Jacobi reachability framework for black-box systems.

Mar 2024  I was invited to give a guest lecture at UC San Diego’s MAE 248 course (Safety for Autonomous Systems) on “Safety Filters for Uncertain Dynamical Systems: Control Theory & Data-driven Approaches”. The lecture is available online. 

Feb 2024  I was invited to give a guest lecture at CMU’s 16-886 course (Models & Algorithms for Interactive Robotics) on “Safety Filters for Uncertain Dynamical Systems: Control Theory & Data-driven Approaches”. The lecture is available online.

If you’re curious about my more whimsical projects, you might want to meet a cute robot tortoise who made it to IEEE Spectrum!