Minseong Choi

Prof. Minseong Choi

Korea Advanced Institute of Science and Technology
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 August 2023 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
Minseong ChoiMinseong Choi
Minseong Choi
SYSeunghoon Yang
Seunghoon Yang
SHSeungho Han
Seungho Han
YLYeongseok Lee
Yeongseok Lee
MLMinyoung Lee
Minyoung Lee
KKKyung-Soo Kim
Kyung-Soo Kim
Recently, several 4D radar datasets have been published due to its high resolution and robustness in extreme weather conditions. However, most of the recent 4D radar dataset publications have insufficient odometry sensors, which is disadvantageous for 4D radar odometry research. To tackle this problem, this letter presents a 4D radar dataset with various odometry sensors based on a robot operating system (ROS) framework called MSC-RAD4R , which stands for Motivated for SLAM in City, ROS-based Automotive Dataset with 4D Radar . Our dataset at a glance includes 98,786 pairs of stereo images, 60,562 frames of LiDAR data, 90,864 frames of 4D radar data, 60,570 frames of RTK-GPS, 60,559 frames of GPS, 1,211,486 frames of IMU data and 6,057,276 wheel data covering approximately 51.6 km and 100 minutes of automotive data in various environments including day, night, snow and smoke. In particular, our setup includes a high-resolution 79 GHz Adaptive PDM FMCW Oculii 4D radar, which provides approximately 5,000–20,000 points per frame and operates at a long range of 400 meters with a frequency of 15 Hz. It is expected that the proposed dataset will be useful for researchers working on 4D radar SLAM.

19 November 2021 in Sensors
Verified
Journal Article
Open access
PDF file
SHSungchul Hong
Sungchul Hong
ABAntyanta Bangunharcana
Antyanta Bangunharcana
JPJaemin Park
Jaemin Park
Minseong ChoiMinseong Choi
Minseong Choi
HSHyu-Soung Shin
Hyu-Soung Shin
With the recent discovery of water-ice and lava tubes on the Moon and Mars along with the development of in-situ resource utilization (ISRU) technology, the recent planetary exploration has focused on rover (or lander)-based surface missions toward the base construction for long-term human exploration and habitation. However, a 3D terrain map, mostly based on orbiters’ terrain images, has insufficient resolutions for construction purposes. In this regard, this paper introduces the visual simultaneous localization and mapping (SLAM)-based robotic mapping method employing a stereo camera system on a rover. In the method, S-PTAM is utilized as a base framework, with which the disparity map from the self-supervised deep learning is combined to enhance the mapping capabilities under homogeneous and unstructured environments of planetary terrains. The overall performance of the proposed method was evaluated in the emulated planetary terrain and validated with potential results.

21 December 2023 in IEEE Access
Verified
Journal Article
Open access
SYSeunghoon Yang
Seunghoon Yang
Minseong ChoiMinseong Choi
Minseong Choi
SHSeungho Han
Seungho Han
KCKeun Ha Choi
Keun Ha Choi
KKKyung-Soo Kim
Kyung-Soo Kim
The integration of cameras and millimeter-wave radar into sensor fusion algorithms is essential to ensure robustness and cost effectiveness for vehicle pose estimation. Due to the low resolution of traditional radar, several studies have investigated 4D imaging radar, which provides range, Doppler, azimuth, and elevation information with high resolution. This paper presents a method for robustly estimating vehicle pose through 4D radar and camera fusion, utilizing the complementary characteristics of each sensor. Leveraging the single-view geometry of the detected vehicle bounding box, the lateral position is derived based on the camera images, and the yaw rate is calculated through feature matching between consecutive images. The high-resolution 4D radar data are used to estimate the heading angle and forward velocity of the target vehicle by leveraging the position and Doppler velocity information. Finally, an extended Kalman filter (EKF) is employed to fuse the physical quantities obtained by each sensor, resulting in more robust pose estimation results. To validate the performance of the proposed method, experiments were conducted in foggy environments, including straight and curved driving scenarios. The experimental results indicate that the performance of the camera-based method is reduced due to frame loss in visually challenging scenarios such as foggy environments, whereas the proposed method exhibits superior performance and enhanced robustness.

29 August 2023 in International Journal of Control, Automation, and Systems
Verified
Journal Article
Closed access
PDF file
SHSeungho Han
Seungho Han
Minseong ChoiMinseong Choi
Minseong Choi
MCMinsu Cho
Minsu Cho
JPJi-il Park
Ji-il Park
KKKyung-Soo Kim
Kyung-Soo Kim

16 July 2025
Verified
Book Chapter
Closed access
PDF file
Hyun-Bin KimHyun-Bin Kim
Hyun-Bin Kim
CKChanseok Kim
Chanseok Kim
BHByeong-Il Ham
Byeong-Il Ham
JKJeonguk Kang
Jeonguk Kang
Minseong ChoiMinseong Choi
Minseong Choi
KKKyung-Soo Kim
Kyung-Soo Kim

4 September 2025 in IEEE Robotics and Automation Letters
Verified
Journal Article
Open access
HJHyun-Yong Jeon
Hyun-Yong Jeon
Minseong ChoiMinseong Choi
Minseong Choi
YLYeongseok Lee
Yeongseok Lee
SOSangyoon Oh
Sangyoon Oh
SYSeunghoon Yang
Seunghoon Yang
KKKyung-Soo Kim
Kyung-Soo Kim
This letter proposes an algorithm for full-velocity estimation by fusing radial velocity vectors obtained from 4D radar with optical flow vectors extracted from camera images. The full-velocity algorithm consists of a preprocessing step and a full-velocity vector estimation step. In preprocessing, radar noise is removed and ego-vehicle velocity estimation is enhanced using a Hampel filter for improved robustness in dynamic environments. In the full-velocity estimation stage, the algorithm estimates full-velocity vectors using a formulation derived from mathematical equations that incorporate multiple constraints. To evaluate the proposed method, an embedded system is implemented on a real vehicle, and datasets are collected under various scenarios. Experimental results show that the proposed algorithm significantly improves object velocity estimation performance. (error rate: baseline 81% $\rightarrow$ proposed 31%).