Hee-Seung Moon

I am a PhD student at Yonsei University advised by Prof. Jiwon Seo in the Intelligent Unmanned Systems Lab. I received my bachelor's degree in the School of Integrated Techonology from Yonsei University. My research interests lie at the intersection of human–robot interaction and machine learning.

Contact: hs.moon@yonsei.ac.kr
Curriculum Vitae


Deep Learning-based Haptic Guidance

2020 - present

Haptic guidance (HG) improves users’ task performance through physical interaction between robots and users. We presented two types of HG: optimal action-based HG (OAHG), which assists users with an optimal action, user prediction-based HG (UPHG), which assists users with their intended action, and proposed a combined HG approach which utilizes the both HG types. Deep learning-based approaches were applied, including self-play-based reinforcement learning for OAHG and meta-learning for UPHG. Through a user study, we validated each HG's assisting performance for users conducting a pHRI task and investigated how the user’s subjective evaluation differs for each HG.

Related Publication

Imaginary Rollout-based Robotic Guide Training

2017 - 2019

Training a robot that engages with humans is challenging, because it is expensive to involve humans in a robot training process requiring numerous data samples. We proposed a deep learning-based model that predict human path following a robot and an evolution strategy-baesd robot training method using "imaginary rollouts" generated by the human predictive model, which compensates for this sample inefficiency problem. We applied the proposed method to the training of a robotic guide for visually impaired people, which was designed to collect multimodal human response data and reflect such data when selecting the robot’s actions.

Related Publication

Study on the Effect of Haptic Information in Human Multitasking

2015 - 2016

We implemented a multimodal task interruption environment involving the simultaneous presentation of visual information and haptic stimuli in order to investigate how the combined stimuli affect the performance on the primary task (i.e., cost of interruption). A user test (n=21) indicated that, within a visuo-tactile task environment, redundant haptic information may not only increase accuracy on the primary task but also reduce the cost of interruption in terms of accuracy.

Related Publication



Sample-Efficient Training of Robotic Guide Using Human Path Prediction Network
H.-S. Moon and J. Seo
arXiv preprint arXiv:2008.05054, 2020 arXiv

International Journals & Conferences

Optimal Action-based or User Prediction-based Haptic Guidance: Can You Do Even Better?
H.-S. Moon and J. Seo
ACM CHI Conference on Human Factors in Computing Systems (CHI), 2021, accepted arXiv

Dynamic Difficulty Adjustment via Fast User Adaptation
H.-S. Moon and J. Seo
ACM Symposium on User Interface Software and Technology (UIST) Poster, 2020 arXiv Page

Prediction of Human Trajectory Following a Haptic Robotic Guide Using Recurrent Neural Networks
H.-S. Moon and J. Seo
IEEE World Haptics Conference (WHC), 2019 arXiv Page

Observation of Human Response to a Robotic Guide Using a Variational Autoencoder
H.-S. Moon and J. Seo
IEEE International Conference on Robotic Computing (IRC), 2019 Page

Observation of Human Trajectory in Response to Haptic Feedback from Mobile Robot
H.-S. Moon, W. Kim, S. Han, and J. Seo
International Conference on Control, Automation and Systems (ICCAS), 2018 Page

Monitoring and Mitigation of Ionospheric Anomalies for GNSS-Based Safety Critical Systems: A review of up-to-date signal processing techniques
J. Lee, Y. J. Morton, J. Lee, H.-S. Moon, and J. Seo
IEEE Signal Processing Magazine, vol. 34, no. 5, pp. 96–110, 2017 Page

Effect of Local-Adaptive Haptic Guidance on a Path-Following Task
H.-S. Moon and J. Seo
International Conference on Control, Automation and Systems (ICCAS), 2017 Page

Effect of Redundant Haptic Information on Task Performance During Visuo-Tactile Task Interruption and Recovery
H.-S. Moon, J. Baek, and J. Seo
Frontiers in Psychology, vol. 7, art. 1924, 2016 Page

Adaptive UI from Human Behavior Pattern on Small Screen Interface: Focused on Double-Swipe Interface
H.-S. Moon and D. Y. Ju
International Conference on Human-Computer Interaction (HCII) Poster, 2015 Page

Awards & Honors

Best Paper Award   •  2017
2017 Korea Navigation Institute (KONI) Conference

Graduate fellowship   •  2015 - 2018
ICT Consilience Creative Program supported by the Ministry of Science and ICT, Korea

Minister's Award from Ministry of Science and ICT, Korea   •  2014
Creative ICT Convergence Korea 2014

Academic Excellence Award   •  Spring 2014, Fall 2013, Spring 2013
Yonsei University, Korea

Undergraduate fellowship   •  2012 - 2014
ICT Consilience Creative Program supported by the Ministry of Science and ICT, Korea

Hee-Seung Moon  |  hs.moon@yonsei.ac.kr