Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sejoon Lim has a total of 34 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    V2I-Based LiDAR and RSSI-Based Ranging-Assisted Visual-Inertial Odometry for Robust Ego Vehicle Localization in GPS-Denied Environments

    by Donggyu Kim, Byeongjin Jung, Jihoon Nam, Sejoon Lim

    This paper proposes a two-dimensional localization framework that integrates Visual-Inertial Odometry (VIO) and roadside units (RSUs) equipped with LiDAR sensors and vehicle-to-infrastructure (V2I) communication, while opportunistically incorporating GPS measurements when available. The proposed method first estimates the local pose of the vehicle using VIO based on onboard IMU and camera sensors. Then, vehicles broadcast Basic Safety Messages (BSM). While BSMs are not explicitly designed for localization, their periodic beaconing is opportunistically reused as a signal carrier. RSUs share this signal data via backhaul communication to estimate coarse vehicle positions using RSSI-based trilateration. These estimates are refined by correlating with object data from RSU-mounted LiDAR sensors, enabling accurate ego vehicle identification and localization even in multi-vehicle scenarios. The corrected ego vehicle position is transmitted back to the vehicle using Sensor Data Sharing Messages (SDSM), compliant with the SAE J2735 standard. Subsequently, this RSU-based position is adaptively fused with VIO and available GPS measurements using dynamically assigned weights based on sensor reliability. This approach effectively mitigates VIO drift, ensuring seamless and robust positioning performance across both GPS-available and GPS-denied environments. The framework was implemented and validated in the CARLA simulator under varying RSU spacing, GPS shadowing, and inter-vehicle distances. Experimental results show that the weighted-average fusion method (VIO+GPS+RSU) reduces localization errors by up to 77.5% compared to standalone VIO and 96.4% compared to standalone GPS, maintaining accuracy within 3 m. Ego vehicle identification performance was strongly influenced by RSU density and inter-vehicle distance, exceeding 95% when RSUs were spaced 50 m apart, but dropping to 20–35% at 150 m. Wider inter-vehicle distances generally yielded more stable identification.

  • Open Access

    One-Stage Brake Light Status Detection Based on YOLOv8

    by Geesung Oh, Sejoon Lim

    Despite the advancement of advanced driver assistance systems (ADAS) and autonomous driving systems, surpassing the threshold of level 3 of driving automation remains a challenging task. Level 3 of driving automation requires assuming full responsibility for the vehicle’s actions, necessitating the acquisition of safer and more interpretable cues. To approach level 3, we propose a novel method for detecting driving vehicles and their brake light status, which is a crucial visual cue relied upon by human drivers. Our proposal consists of two main components. First, we introduce a fast and accurate one-stage brake light status detection network based on YOLOv8. Through transfer learning using a custom dataset, we enable YOLOv8 not only to detect the driving vehicle, but also to determine its brake light status. Furthermore, we present the publicly available custom dataset, which includes over 11,000 forward images along with manual annotations. We evaluate the performance of our proposed method in terms of detection accuracy and inference time on an edge device. The experimental results demonstrate high detection performance with an mAP50 (mean average precision at IoU threshold of 0.50) ranging from 0.766 to 0.793 on the test dataset, along with a short inference time of 133.30 ms on the Jetson Nano device. In conclusion, our proposed method achieves high accuracy and fast inference time in detecting brake light status. This contribution effectively improves safety, interpretability, and comfortability by providing valuable input information for ADAS and autonomous driving technologies.


  • BYEL: Bootstrap Your Emotion Latent

    by Hyungjun Lee, Hwangyu Lim, Sejoon Lim


  • Open Access

    Deep Learning-Based Driver’s Hands on/off Prediction System Using In-Vehicle Data

    by Hyeongoo Pyeon, Hanwul Kim, Rak Chul Kim, Geesung Oh, Sejoon Lim

    Driver’s hands on/off detection is very important in current autonomous vehicles for safety. Several studies have been conducted to create a precise algorithm. Although many studies have proposed various approaches, they have some limitations, such as robustness and reliability. Therefore, we propose a deep learning model that utilizes in-vehicle data. We also established a data collection system, which collects in-vehicle data that are auto-labeled for efficient and reliable data acquisition. For a robust system, we devised a confidence logic that prevents outliers’ sway. To evaluate our model in more detail, we suggested a new metric to explain the events, considering state transitions. In addition, we conducted an extensive experiment on the new drivers to demonstrate our model’s generalization ability. We verified that the proposed system achieved a better performance than in previous studies, by resolving their drawbacks. Our model detected hands on/off transitions in 0.37 s on average, with an accuracy of 95.7%.


  • Open Access

    PU-MFA: Point Cloud Up-Sampling via Multi-Scale Features Attention

    by Hyungjun Lee, Sejoon Lim

    Recently, research using point clouds has been increasing with the development of 3D scanner technology. According to this trend, the demand for high-quality point clouds is increasing, but there is still a problem with the high cost of obtaining high-quality point clouds. Therefore, with the recent remarkable development of deep learning, point cloud up-sampling research, which uses deep learning to generate high-quality point clouds from low-quality point clouds, is one of the fields attracting considerable attention. This paper proposes a new point cloud up-sampling method called Point cloud Up-sampling via Multi-scale Features Attention (PU-MFA). Inspired by prior studies that reported good performance at generating high-quality dense point set using the multi-scale features or attention mechanisms, PU-MFA merges the two through a U-Net structure. In addition, PU-MFA adaptively uses multi-scale features to refine the global features effectively. The PU-MFA was compared with other state-of-the-art methods in various evaluation metrics through various experiments using the PU-GAN dataset, which is a synthetic point cloud dataset, and the KITTI dataset, which is the real-scanned point cloud dataset. In various experimental results, PU-MFA showed superior performance of generating high-quality dense point set in quantitative and qualitative evaluation compared to other state-of-the-art methods, proving the effectiveness of the proposed method. The attention map of PU-MFA was also visualized to show the effect of multi-scale features.


  • Open Access

    Multimodal Data Collection System for Driver Emotion Recognition Based on Self-Reporting in Real-World Driving

    by Geesung Oh, Euiseok Jeong, Rak Chul Kim, Ji Hyun Yang, Sungwook Hwang, Sangho Lee, Sejoon Lim

    As vehicles provide various services to drivers, research on driver emotion recognition has been expanding. However, current driver emotion datasets are limited by inconsistencies in collected data and inferred emotional state annotations by others. To overcome this limitation, we propose a data collection system that collects multimodal datasets during real-world driving. The proposed system includes a self-reportable HMI application into which a driver directly inputs their current emotion state. Data collection was completed without any accidents for over 122 h of real-world driving using the system, which also considers the minimization of behavioral and cognitive disturbances. To demonstrate the validity of our collected dataset, we also provide case studies for statistical analysis, driver face detection, and personalized driver emotion recognition. The proposed data collection system enables the construction of reliable large-scale datasets on real-world driving and facilitates research on driver emotion recognition. The proposed system is avaliable on GitHub.


  • Open Access

    Deep Reinforcement Learning Based Dynamic Proportional-Integral (PI) Gain Auto-Tuning Method for a Robot Driver System

    by Joonghoo Park, Heejung Kim, Kyunghun Hwang, Sejoon Lim

    To meet the growing trend of stringent fuel economy regulations, automakers around the world are designing modules such as engines, motors, transmissions and batteries to be as efficient as possible. In order to verify the effect of these designs on the overall fuel efficiency of the vehicle, the vehicle equipped with each module is placed on the chassis dynamometer, driven to follow the target vehicle speed, and actual fuel efficiency is measured. These tests are traditionally performed by human operators, but are now being replaced by robots (physical or software) to ensure the accuracy and reliability of test results. Although the conventionally proposed proportional integral (PI)-based controller has a simple structure and is easy to implement, it requires the process of finding the optimal gain whenever the test conditions such as vehicle or drive cycle change, which is difficult and time consuming. In this study, we propose a proportional integral controller gain adjustment algorithm using deep reinforcement learning. The reinforcement learning agent learns to dynamically modify the PI gain value of the acceleration/deceleration pedal to better follow the target vehicle in a simulation environment. The perturbation is used in each training episode to reduce the difference between the simulation and real testing environment. Upon completion of the training process, the trained agent performs an adjustment process that generates a reference gain table. We then use this reference gain table to perform a real test. The performance of the proposed system was evaluated using Hyundai Tucson HEV (NX4) on an AVL chassis dynamometer. We also compared the performance of our proposed algorithm to traditional fuzzy logic-based PI controllers. The obtained experimental results show that the proposed control system achieved a performance improvement of aounrd 46.8% compared to the conventional PI control system in terms of root mean square error.


  • Open Access

    Development of a Simple Robotic Driver System (SimRoDS) to Test Fuel Economy of Hybrid Electric and Plug-In Hybrid Electric Vehicles Using Fuzzy-PI Control

    by Kyunghun Hwang, Joonghoo Park, Heejung Kim, Tea-Yong Kuc, Sejoon Lim

    Over the past decade, new models of hybrid electric vehicles have been released worldwide, and the fuel efficiency of said vehicles has increased by more than 5%. To further improve fuel efficiency, vehicle manufacturers have made efforts to design modules (e.g., engines, motors, transmissions, and batteries) with the highest efficiency possible. To do so, the fuel economy test process, which is conducted primarily using a chassis dynamometer, must produce reliable and accurate results. To accurately analyze the fuel efficiency improvement rate of each module, it is necessary to reduce the test deviation. When the test conducted by human drivers, the test deviation is somewhat large. When the test is conducted by a physical robot driver, the test deviation is improved; however, these robots are expensive and time-consuming to install and take up considerable amount of space in the driver’s seat. To compensate for these shortcomings, we propose a simple, structured robot system that manipulates electrical signals without using mechanical link structures. The controller of this robot driver uses the widely used PI controller. Although PI controllers are simple and perform well, since the dynamics of each test vehicle is different (e.g., acceleration response), the PI controller has a disadvantage in that it cannot determine the optimal PI gain value for each vehicles. In this work, the fuzzy control theorem is applied to overcome this disadvantage. By using fuzzy control to deduce the optimal value of the PI gain, we confirmed that our proposed system is available to conduct tests on vehicles with different dynamics.