Jiwon Park
Biography
Passionate student specializing in robot perception, with a talent for developing robotic system software. Dedicated to advancing research in robust perception techniques to enhance the reliability of outdoor mobile robots.
Recent graduate from Kyung Hee University with a dual major in Electronic Engineering and Software Convergence, Robot Vision Track(August 2024). Gained industry experience through internships at NAVER LABS and ROBROS.
Interests
- Robot Perception
- Off-road Mobile Robot
- Traversable Region Segmentation
Education
M.S. in Robotics, Upcoming student starts from 2025
KAIST
B.S. in Electronic Engineering, Software Convergence (Robot Vision Track), 2019 - 2024
Kyung Hee University
Projects
Publications
*
(2024).
Transformer based Collision Detection Approach by Torque Estimation using Joint Information.
The Journal of Korea Robotics Society, Vol.19, No.3.
(2023).
BumpyPatch: Heightmap-based Outdoor Point Cloud Segmentation to Find Less Bumpy Road.
IEEE IRC 2023.
(2023).
Outdoor visual SLAM and Path Planning for Mobile-Robot.
In IEEE IRC 2022 CHARMS workshop.
Experience
Intern
- Developed software of inference board for robot, enhancing on-board processing capabilities
- Implemented deep learning model optimization techniques for low-cost NPUs
- Integrated third-party robots into multi-robot control system (ARC)
- Developed monitoring applications for robot experiments
ROS1 / ROS2 / RKNN / gRPC / MQTT / CMake / Flutter
Intern
- Developed experimental environments for robotic arms using MuJoCo simulation platform
- Proposed a transformer-based model for detecting collisions in robotic arms without external sensors, relying on joint position and torque controller
ROS 1 / MuJoCo / PyTorch
Student Intern
- Conducted research on outdoor mobile visual SLAM (Simultaneous Localization and Mapping)
- Focused on enhancing accuracy and reliability of visual navigation for mobile robots
Mobile Robot / ROS / ORB-SLAM2
Student Intern
- Advisor : Professor Hyoseok Hwang
