About Jungseok
I am a Postdoctoral Associate in Marine Robotics Group with Prof. John Leonard in the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). I received a Ph.D. in Computer Science from the Department of Computer Science and Engineering at the University of Minnesota. I was a member of IRV LAB advised by Prof. Junaed Sattar.
My research advances perception and mapping for underwater and field robots, progressing from robust single-robot semantic understanding to cooperative, action-capable robot teams.
I design multimodal perception and SLAM algorithms that operate under degraded visual conditions and uncertainty, integrating object-centric and semantic reasoning into mapping. Building on this foundation, I will focus on uncertainty-aware cooperative mapping and multi-robot intervention, enabling underwater robots to understand their environments and collaboratively transform them.
Outside of my research hours, I like to be active (e.g., sailing, swim, and play tennis). I am also a sailor and certified scuba diver. Check out my research projects and publications.
Recent News
2026.07 Our workshop paper Loggerhead: Grounding Feed-forward Transformers with Multimodal Odometry for Metric-scale Reef Reconstructions won the Best Contribution Award ($5,000) at the RSS 2026 Workshop on Mapping the Reef: Underwater 3D Reconstruction for Coral Ecosystems, Vienna, Austria.
2026.07 We submitted a paper A Simulation and Field Study Analysis of Cooperative Range-Aided Localization Strategies for Compact, Low-Cost AUV Teams to IEEE AUV 2026, Southampton, UK.
2026.06 I gave a spotlight talk on our recent work, ReefMapGS, at the From Sea to Space: Advancing Perception in Harsh Domains workshop at ICRA 2026.
2026.05 Two workshops I am co-organizing, AQ²UASIM-V2 (Advancing Quantitative and QUAlitative SIMulators for marine applications) and AMBRE (Adaptive Manipulation: Bridging Rigid and Soft Robotics in Underwater Environments), have been accepted to IROS 2026.
2026.03 I gave an invited seminar at the MIT MechE department, Host: Domitilla Del Vecchio, presenting my research on embodied spatial intelligence.
2026.03 We submitted a paper ReefMapGS: Enabling Large-Scale Underwater Reconstruction by Closing the Loop Between Multimodal SLAM and Gaussian Splatting.
2026.02 I gave an invited seminar at Inha University, presenting our latests and ongoing research on Underwater SLAM and Mapping.
2026.01 In Srping 2026, I am co-teaching 2.679: Electronics for Mechanical Systems II at the MIT MechE department.
Released Datasets
 |
TrashCAN (2020.07) An Instance-Segmentation Labeled Dataset of Trash Observations. Download |
 |
Trash-ICRA19 (2020.07) A Bounding Box Labeled Dataset of Underwater Trash. Download |
Selected Invited Talks
2026.06 Underwater 3D Reconstruction by Interleaving Multimodal SLAM and Incremental Gaussian Splatting, From Sea to Space: Advancing Perception in Harsh Domains Workshop at ICRA 2026, Vienna, Austria.
2026.03 Embodied Spatial Intelligence, MIT MechE Department, Host: Domitilla Del Vecchio.
2025.08 Semantic SLAM in Challenging Environments, MIT SPARK Lab, Host: Luca Carlone (Lorenzo Shaikewitz).
Selected Honors and Awards
2026.07 RSS 2026 Workshop Best Contribution Award, Mapping the Reef: Underwater 3D Reconstruction for Coral Ecosystems, $5,000.
2024.10 IROS 2024 Workshop Best Paper Award, Autonomous Robotic Systems in Aquaculture, $2,125.
2019 UMII MnDRIVE Graduate Fellowship, University of Minnesota MnDrive, $50,000. Awarded in recognition of academic record and potential for future research productivity in Robotics.
Industry Experiences
2022.02-2022.09 Samsung AI Center New York Research intern. Worked on visual servoing and semantic mapping. Published an IROS 2022 paper and an ICRA 2023 paper; Obtained two patents (1, 2).
2020.05-2020.08 Sentera Deep learning intern. Developed a weed identification pipeline.
Professional Services
Professional Service Roles
- Session Chair, International Conference on Intelligent Robots and Systems (IROS), 2025
- Session Chair, IEEE International Conference on Robotics and Automation (ICRA), 2025
- External Evaluator, Ph.D. Thesis Proposal, Department of Aeronautics and Astronautics, MIT, 2025
International Journals/Conferences Reviewer
IEEE-TRO |
2024 |
IEEE-JOE |
2022 |
IEEE RA-L |
2022-2025 |
ICRA |
2020 - 2026 |
IROS |
2019 - 2025 |
RSS |
2025 - 2026 |
CRV |
2019 - 2021 |
CRV |
Program Committee Member for 2021 |
Faculty Hiring Committee 2020-2021
- Student Representative, Department of Computer Science and Engineering, University of Minnesota
Favorite Quotes
“Do it fast, but do it correctly.”
“Documentation is a Love Letter That You Write to Your Future Self.” —Damian Conway
“Everything you learn, Learn the thing that is the building block for the thing you just learned.” —John Mayer