Welcome

Thank you for visiting my profile page.

I am a M.S. candidate in Applied Artificial Intelligence at Myongji University (MJU), advised by Prof. Jaehee Jung .

My master's research focused on enhancing dynamic obstacle avoidance performance in mobile robot navigation by leveraging deep learning models to process semantic information and generate human-like trajectories based on imitation learning.

In the future, my primary research interests will focus on

Reinforcement Learning for Robotic Manipulation

Vision-Language-Action model

Multimodal learning

News

[Oct 2025] Started research on graph-based pedestrian trajectory prediction.
[Oct 2025] The first open-source robot project under Gradus, DiffDriveRobot, has been released.
[Oct 2025] Launched the open-source robotics project Gradus.
[Sep 2025] One paper on imitation-learning-based dynamic obstacle avoidance was accepted for publication in IEEE Access.
[Sep 2025] One paper on hierarchical misbehavior classification for V2X communication systems was accepted for publication in IEEE Access.
[Mar 2025] Submitted a paper to IROS 2025. (Rejected in Jun 2025)
[Oct 2024] Submitted a paper to IEEE RA-L. (Rejected in Dec 2024)
[Apr 2024] I have been selected as the principal investigator for a research project funded by the National Research Foundation's "석사과정생연구장려금지원사업" (₩12,000,000).

Research

(* denotes equal contribution)

1. HADP: Hybrid A*-Diffusion Planner for Robust Navigation in Dynamic Obstacle Environments

[Paper]

Mingyu Kim*, Chanyeong Heo*, Jaehee Jung

IEEE Access (2026.9) Impact Factor: 3.6

We proposed the Hybrid A*-Diffusion Planner (HADP) to overcome the generalization and data-efficiency limitations of pure diffusion-based planning in dynamic environments. Our approach uses A* for global path planning in obstacle-free regions and a conditional diffusion model fed by a semantic map, robot pose, and local goal to generate responsive local avoidance trajectories upon detecting dynamic obstacles.

2. HiMSELF: A Hierarchical Misbehavior Classification with Sequence Embedding by Latent Features in Vehicular Ad-Hoc Networks

[Paper]

Mingyu Kim, Dae Hyun Yum, Jaehee Jung

IEEE Access (2026.8) Impact Factor: 3.6

We propose HiMSELF, a misbehavior classification system that learns latent sequence embeddings of multi-class BSM data, applies hierarchical clustering to construct a two-stage classification hierarchy, and achieves an average F1-score of 0.9918 on 19 misbehavior classes, outperforming existing models.