Electrical and Electronic Engineering, Computer Engineering, Artificial Intelligence
10
Scopus Publications
64
Scholar Citations
5
Scholar h-index
3
Scholar i10-index
Scopus Publications
Multi-modal, multi-step human–robot interaction method for natural interaction of service robots Jargalbaatar Yura, Bat-Erdene Byambasuren, Donghan Kim Intelligent Service Robotics, 2025 Robots are becoming increasingly popular across various industries, including manufacturing, healthcare, agriculture, and personal services. This growing demand has spurred the development of robots with specialized abilities and applications, improving safety, productivity, and quality of life. This paper focuses on service robots with verbal and visual communication, which play a crucial role in human–robot interaction. We propose a novel method to enhance the accuracy of this communication by combining sound localization and image processing, enabling more natural interactions. Additionally, we aim to improve the service robot’s mobility and interaction by adjusting its rotational speed to match human movement, thus enhancing adaptability and usability in diverse environments. This paper provides an overview of similar research efforts in human–robot interactions, details our proposed methodology, describes human–robot test environments, and presents experimental results, conclusions and implications for future research.
Social Type-Aware Navigation Framework for Mobile Robots in Human-Shared Environments Sumin Kang, Sungwoo Yang, Daewon Kwak, Yura Jargalbaatar, Donghan Kim Sensors, 2024 As robots become increasingly common in human-populated environments, they must be perceived as social beings and behave socially. People try to preserve their own space during social interactions with others, and this space depends on a variety of factors, such as individual characteristics or their age. In real-world social spaces, there are many different types of people, and robots need to be more sensitive, especially when interacting with vulnerable subjects such as children. However, the current navigation methods do not consider these differences and apply the same avoidance strategies to everyone. Thus, we propose a new navigation framework that considers different social types and defines appropriate personal spaces for each, allowing robots to respect them. To this end, the robot needs to classify people in a real environment into social types and define the personal space for each type as a Gaussian asymmetric function to respect them. The proposed framework is validated through simulations and real-world experiments, demonstrating that the robot can improve the quality of interactions with people by providing each individual with an adaptive personal space. The proposed costmap layer is available on GitHub.
Transformable Gaussian Reward Function for Socially Aware Navigation Using Deep Reinforcement Learning Jinyeob Kim, Sumin Kang, Sungwoo Yang, Beomjoon Kim, Jargalbaatar Yura, Donghan Kim Sensors, 2024 Robot navigation has transitioned from avoiding static obstacles to adopting socially aware navigation strategies for coexisting with humans. Consequently, socially aware navigation in dynamic, human-centric environments has gained prominence in the field of robotics. One of the methods for socially aware navigation, the reinforcement learning technique, has fostered its advancement. However, defining appropriate reward functions, particularly in congested environments, holds a significant challenge. These reward functions, crucial for guiding robot actions, necessitate intricate human-crafted design due to their complex nature and inability to be set automatically. The multitude of manually designed reward functions contains issues such as hyperparameter redundancy, imbalance, and inadequate representation of unique object characteristics. To address these challenges, we introduce a transformable Gaussian reward function (TGRF). The TGRF possesses two main features. First, it reduces the burden of tuning by utilizing a small number of hyperparameters that function independently. Second, it enables the application of various reward functions through its transformability. Consequently, it exhibits high performance and accelerated learning rates within the deep reinforcement learning (DRL) framework. We also validated the performance of TGRF through simulations and experiments.
Clustered Lidar-Based Potential Field Path Planning Sungwoo Yang, Sumin Kang, Jargalbaatar Yura, Beomjoon Kim, Donghan Kim 2023 20th International Conference on Ubiquitous Robots Ur 2023, 2023 Mobile service robots operating in complex indoor environments require collision-free path planning to reach their goals. The artificial potential field (APF) path planning method is a commonly used technique that considers the robot as a point in the potential field and is easy to implement using lidar sensor data. Since the APF method has a significant disadvantage of local minima, many APF studies have focused on solving it. In this paper, however, we take a different approach to improve the performance of the APF. We propose a Clustered LiDAR-based Potential Field path planning method that introduces 2D lidar clustering to distinguish obstacles in partially observable environments. The proposed method is implemented in ROS and demonstrated using a simulated mobile manipulator in Gazebo.
Novel design of artificial eye using EOG (electrooculography) Eunha Moon, Hyeonjun Park, Jargalbaatar Yura, Donghan Kim Proceedings 2017 1st IEEE International Conference on Robotic Computing Irc 2017, 2017 In this paper, we present the system that controls the artificial eye based on EOG. The makes use of artificial eye model, consists of the mask and the eye mechanism for natural movement like the movement of the real eye. To move up, down, left, and right of the artificial eye, one servomotor is mounted above the eye of the artificial eye model and one below the eye. The link connected to the motor movement and moves up, down, left, and right. The movement of the artificial eye must move in synchronize with the real human eye. Therefore, in order to control such movement, a signal of real human eye motion is measured by EOG (Electrooculography). Signals of up, down, left, and right movement required for eyeball movement are processed using the signal of EOG. Therefore, measure the vertical and horizontal signals separately and check the maximum and minimum signals when moving up, down, left, and right.
Inspection robot based mobile sensing and power line tracking for smart grid Bat-erdene Byambasuren, Donghan Kim, Mandakh Oyun-Erdene, Chinguun Bold, Jargalbaatar Yura Sensors Switzerland, 2016 Smart sensing and power line tracking is very important in a smart grid system. Illegal electricity usage can be detected by remote current measurement on overhead power lines using an inspection robot. There is a need for accurate detection methods of illegal electricity usage. Stable and correct power line tracking is a very prominent issue. In order to correctly track and make accurate measurements, the swing path of a power line should be previously fitted and predicted by a mathematical function using an inspection robot. After this, the remote inspection robot can follow the power line and measure the current. This paper presents a new power line tracking method using parabolic and circle fitting algorithms for illegal electricity detection. We demonstrate the effectiveness of the proposed tracking method by simulation and experimental results.
A study about sound quality for violin playing robot Hyeonjun Park, Wonse Jo, Kyeongmin Choi, Hwonjae Jung, Yura jargalbaatar, Bum-Joo Lee, Donghan Kim Procedia Computer Science, 2015 This paper introduces a violin playing robot that imitates the playing technique of human. A violinist learns how to play through an endless practice. A bowing velocity, bowing force, and sound point are important factors in determining the sound quality. Thus, in this paper, the sound quality has been analysed in the variable speed using the violin playing robot, where an industrial vertical multi-joint robot arm is used. Fast Fourier transform is used to convert the played sound using a 32-bit microcontroller, and then the result is compared to the natural frequency of the G string. This paper also studies the robotic hand for violin fingering. The robotic finger by constructing a 3-axis load cell at the end of finger. Based on the amplified strain gauge value, it was able to apply an appropriate force on the string when playing the violin. The robot hand that is similar to the size of a hand of adult male made a form of Anthropomorphic-dexterous. So this has a total of 12 degrees of freedom(DOF). A control method works based on Wire driven. And then, the end of the hand equipped with a 3-axis load cell to measure the force of fingertip at real-time. Lastly, this paper describes the mechanism and experimental results of anthropomorphic robotic finger, which is developed to facilitate the performance of violin playing robot. In order to present the feasibility of accurate control, a 3-axis load cell is developed and mounted at the end of finger.
RECENT SCHOLAR PUBLICATIONS
Multi-modal, multi-step human–robot interaction method for natural interaction of service robots J Yura, BE Byambasuren, D Kim Intelligent Service Robotics 18 (3), 389-401 , 2025 2025 Citations: 3
Transformable gaussian reward function for socially aware navigation using deep reinforcement learning J Kim, S Kang, S Yang, B Kim, J Yura, D Kim Sensors 24 (14), 4540 , 2024 2024 Citations: 12
Transformable Gaussian Reward Function for Socially-Aware Navigation with Deep Reinforcement Learning J Kim, S Kang, S Yang, B Kim, J Yura, D Kim arXiv preprint arXiv:2402.14569 , 2024 2024 Citations: 3
Determining Grasp Positions with 4-Finger Gripper Manipulator Using Reinforcement Learning M Kim, S Kang, S Yang, J Yura, D Kim International Conference on Intelligent Autonomous Systems, 179-186 , 2023 2023
Clustered Lidar-Based Potential Field Path Planning S Yang, S Kang, J Yura, B Kim, D Kim 2023 20th International Conference on Ubiquitous Robots (UR), 576-580 , 2023 2023
Sound improvement of violin playing robot applying auditory feedback W Jo, J Yura, D Kim Journal of Electrical Engineering & Technology 12 (6), 2378-2387 , 2017 2017 Citations: 4
Novel design of artificial eye using EOG (electrooculography) E Moon, H Park, J Yura, D Kim 2017 First IEEE International Conference on Robotic Computing (IRC), 404-407 , 2017 2017 Citations: 10
Modeling of violin playing robot arm with MATLAB/SIMULINK J Yura, M Oyun-Erdene, BE Byambasuren, D Kim Robot Intelligence Technology and Applications 4: Results from the 4th … , 2016 2016 Citations: 8
Inspection robot based mobile sensing and power line tracking for smart grid B Byambasuren, D Kim, M Oyun-Erdene, C Bold, J Yura Sensors 16 (2), 250 , 2016 2016 Citations: 19
Development of robotic finger using 3-axis load cell for violin playing robot H Park, W Jo, K Choi, H Jung, J Yura, S Lee, BJ Lee, DH Kim Advanced Science and Technology Letters 90, 22Y26 , 2015 2015 Citations: 5
MOST CITED SCHOLAR PUBLICATIONS
Inspection robot based mobile sensing and power line tracking for smart grid B Byambasuren, D Kim, M Oyun-Erdene, C Bold, J Yura Sensors 16 (2), 250 , 2016 2016 Citations: 19
Transformable gaussian reward function for socially aware navigation using deep reinforcement learning J Kim, S Kang, S Yang, B Kim, J Yura, D Kim Sensors 24 (14), 4540 , 2024 2024 Citations: 12
Novel design of artificial eye using EOG (electrooculography) E Moon, H Park, J Yura, D Kim 2017 First IEEE International Conference on Robotic Computing (IRC), 404-407 , 2017 2017 Citations: 10
Modeling of violin playing robot arm with MATLAB/SIMULINK J Yura, M Oyun-Erdene, BE Byambasuren, D Kim Robot Intelligence Technology and Applications 4: Results from the 4th … , 2016 2016 Citations: 8
Development of robotic finger using 3-axis load cell for violin playing robot H Park, W Jo, K Choi, H Jung, J Yura, S Lee, BJ Lee, DH Kim Advanced Science and Technology Letters 90, 22Y26 , 2015 2015 Citations: 5
Sound improvement of violin playing robot applying auditory feedback W Jo, J Yura, D Kim Journal of Electrical Engineering & Technology 12 (6), 2378-2387 , 2017 2017 Citations: 4
Multi-modal, multi-step human–robot interaction method for natural interaction of service robots J Yura, BE Byambasuren, D Kim Intelligent Service Robotics 18 (3), 389-401 , 2025 2025 Citations: 3
Transformable Gaussian Reward Function for Socially-Aware Navigation with Deep Reinforcement Learning J Kim, S Kang, S Yang, B Kim, J Yura, D Kim arXiv preprint arXiv:2402.14569 , 2024 2024 Citations: 3
Determining Grasp Positions with 4-Finger Gripper Manipulator Using Reinforcement Learning M Kim, S Kang, S Yang, J Yura, D Kim International Conference on Intelligent Autonomous Systems, 179-186 , 2023 2023
Clustered Lidar-Based Potential Field Path Planning S Yang, S Kang, J Yura, B Kim, D Kim 2023 20th International Conference on Ubiquitous Robots (UR), 576-580 , 2023 2023
