Gunhee Shin
Ph.D. Student, Urban Robotics Lab, KAIST
School of Electrical Engineering, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea
gunhee_shin@kaist.ac.kr·Google Scholar·GitHub·gunheeshin.github.io

Education

Ph.D. Student, Electrical Engineering — KAIST
2025 – Present
  • Advisor: Prof. Hyun Myung — Urban Robotics Lab
M.S., Electrical Engineering — KAIST
2023 – 2025
  • GPA: 3.7/4.3
  • Advisor: Prof. Hyun Myung — Urban Robotics Lab
  • Thesis: "Uncertainty-aware Multiple LiDAR-Inertial Odometry for Outdoor Mobile Robots"
B.S., Electrical Engineering — Inha University
2017 – 2023
  • GPA: 3.9/4.5
  • Thesis: "A Fast Correspondence Matching for Iterative Closest Point Algorithm"

Publications

Journal Articles

  1. G. Shin, J. Choi, and K. Kim†. "A Fast Correspondence Matching for Iterative Closest Point Algorithm." J. Korea Robot. Soc., vol. 17, no. 3, pp. 373–380, 2022.

Conference Papers

  1. G. Shin, Y. Kim, C. Lee, W. Kim, M. Lee, S. Han, J. Woo, H. Chin, M. Park, and H. Myung†. "Online Target-less Radar-LiDAR-Camera Extrinsic Calibration via Joint Optimization." in Proc. Int. Conf. Control, Automat. Syst., 2026.
  2. G. Shin, S. Lee, J. Kong, Y. Seo, and H. Myung†. "SE(3)-LIO: Smooth IMU Propagation With Jointly Distributed Poses on SE(3) Manifold for Accurate and Robust LiDAR-Inertial Odometry." in Proc. IEEE Int. Conf. Robot. Automat., 2026.
  3. H. Lim, D. Kim, G. Shin, J. Shi, I. Vizzo, H. Myung†, J. Park†, and L. Carlone†. "KISS-Matcher: Fast and Robust Point Cloud Registration Revisited." in Proc. IEEE Int. Conf. Robot. Automat., 2025.
  4. G. Shin, S. Lee, M. Oh, D. Lee, J. Lee, Y. Seo, and H. Myung†. "NM-LIO: Multiple LiDAR-Inertial Odometry Addressing LiDAR Measurement Noise Discrepancy." in Proc. Int. Conf. Robot Intell. Tech. Appl., 2024.
  5. M. Oh, G. Shin, S. Jang, S. Lee, D. Lee, W. Song, B. Yu, H. Lim, J. Lee, and H. Myung†. "B-TMS: Bayesian Traversable Terrain Modeling and Segmentation Across 3D LiDAR Scans and Maps for Enhanced Off-Road Navigation." in Proc. IEEE Intell. Veh. Symp. Workshops, 2024.
  6. H. Lim, K. Han, G. Shin, G. Kim, S. Hong, and H. Myung†. "ORORA: Outlier-Robust Radar Odometry." in Proc. IEEE Int. Conf. Robot. Automat., 2023.
† Corresponding author

Patents

  1. H. Myung, G. Shin, S. Lee, and J. Kong. "LiDAR and Inertial-Based Position and Pose Estimation System and Method." Korean Patent Application No. 10-2026-0080491, May 2026.

Open-Source Software

Honors & Awards

Winner, COMFORT Localization Challenge — Team Lead · IROS 2026, Pittsburgh, USA
2026
  • Challenge: Estimate the trajectory of an ANYmal legged robot in six field scenes of the GrandTour dataset.
  • Role: Built the whole system: LiDAR-visual-inertial odometry and online drift correction.
Winner, Nothing Stands Still Challenge — CVPR 2026, Denver, USA
2026
  • Challenge: Align scans of a building site taken at different construction stages into one map.
  • Role: Graph optimization that aligns the scans across stages.
2nd Place, Quadruped Robot Challenge — ICRA 2025, Atlanta, USA
2025
  • Challenge: Legged robots cross rough-terrain lanes and complete inspection and manipulation tasks.
  • Role: Pose estimation for onboard navigation.
2nd Place, Quadruped Robot Challenge (Autonomous Track) — ICRA 2024, Yokohama, Japan
2024
  • Challenge: Legged robots run as many laps of a course as they can in 10 minutes, fully autonomously.
  • Role: Pose estimation for onboard navigation.
Winner, HILTI SLAM Challenge (LiDAR Single-Session) — ICRA 2023, London, UK
2023
  • Challenge: Estimate accurate trajectories from data recorded on construction sites, in narrow rooms, and in the dark.
  • Role: LiDAR-inertial odometry and merging of multiple sessions.

Projects

GNSS-Free Visual-Inertial Navigation for Drones — Hanwha Aerospace, Korea
2026 – Present
  • Role: Fused visual-inertial odometry with satellite-image localization.
Online Radar-LiDAR-Camera Calibration for Autonomous Vehicles — Project Lead · LG Innotek, Korea
2025 – Present
  • Role: Led development of online calibration between radar, LiDAR, and camera without targets.
Off-Road Autonomous Driving in Unstructured Terrain — Hanwha Aerospace, Korea
2023 – 2026
  • Role: Built LiDAR-inertial odometry (SE(3)-LIO) and fused it with satellite-image localization for driving without GNSS.
Mobility Intelligence Software for Legged Robots — KEIT, Korea
2022 – 2026
  • Role: Set up the multi-LiDAR sensor system and built pose estimation for onboard navigation.
Semantic SLAM for Spatial AI — LG Electronics, Korea
2023
  • Role: Ran 3D scene graph SLAM (Hydra) on real sensor data and added object-level loop closing.