孙乐源 <br> Leyuan Sun

孙乐源
Leyuan Sun

WUXI UNIVERSITY

☯️ About Me

Leyuan Sun was born in Zhenjiang, Jiangsu (江苏镇江) in August 1993. He received his B.Eng. degree from Jiangsu University, Zhenjiang, China, in 2015, and his M.Eng. and Ph.D. degrees from the University of Tsukuba (筑波大学, 智能机能系统专业), Tsukuba, Japan, in 2020 and 2023, respectively. From April 2018 to March 2023, under the supervision of Professor Fumio Kanehiro (金広 文男), who is the director of the CNRS-AIST Joint Robotics Laboratory (JRL), IRL, National Institute of Advanced Industrial Science and Technology (AIST) (国立研究開発法人産業技術総合研究所), Japan. From 2023.4 to 2025.3, he holds a postdoctoral position with the Computer Vision Research Team, Artificial Intelligence Research Center, hosted by Dr. Yusuke Yoshiyasu (吉安 祐介), who is the team leader of Computer Vision Research Team, Artificial Intelligence of Research Center, AIST. His research interests include robotics, embodied AI navigation, multimodal learning, and 3D vision. Since September 2025, he has been a lecturer at Wuxi University, beginning his career in Wuxi, Jiangsu, China.

🧘 Update

中文
最好的工作是终身学习者在自由市场中的创造性表达。
—— 出自《纳瓦尔宝典》

English
The best jobs are creative expressions of lifelong learners in free markets.
—— From The Almanack of Naval Ravikant

日本語
最高の仕事とは、生涯学び続ける者が、自由な市場の中で自らの創造性を表現することである。
—— 『The Almanack of Naval Ravikant』より

📝 Publications

My full paper list can be found at

Graph-MambaNav

IEEE RA-L, 中科院二区, 新锐二区TOP, JCR Q1, IF:5.3 Graph-MambaNav: Spatial-Temporal Graph Mamba Leveraging Object-Relation Knowledge for Object-Goal Navigation
Leyuan Sun*, Genxin Chen, Linwei Ye, Yan Zhang, Xi Kan, and Yanfei Sun
[arXiv] [IEEE] [Demo]

PSNet

ADVEI, CCF-B, 中科院一区TOP, JCR Q1, IF:11.5 Enhancing Multimodal-Input Object Goal Navigation by Leveraging Large Language Models for Inferring Room-Object Relationship Knowledge
Leyuan Sun*, Asako Kanezaki, Guillaume Caron, and Yusuke Yoshiyasu
[PDF] [Project page]

Memory-MambaNav

IMAVIS, CCF-C, 中科院小类2区, JCR Q1, IF:5.0 Memory-MambaNav: Enhancing Object-Goal Navigation through Integration of Spatial-Temporal Scanning with State Space Models
Leyuan Sun*, Yusuke Yoshiyasu.
[PDF] [Project page]

TransFusionOdom

IEEE Sensors Journal, 中科院二区TOP, JCR Q1, IF:4.3 TransFusionOdom: Transformer-Based LiDAR-Inertial Fusion Odometry Estimation
Leyuan Sun*, Guanqun Ding, Yue Qiu, Y. Yoshiyasu and F. Kanehiro.
[PDF] [Dataset sample]

CertainOdom

IEEE ROBIO 2022 CertainOdom: Uncertainty Weighted Multi-task Learning Model for LiDAR Odometry Estimation
Leyuan Sun*, Guanqun Ding, Y. Yoshiyasu and F. Kanehiro.
[PDF] [Best Conference Paper Award]

PSNet

IEEE Humanoids 2022 Enhanced Visual Feedback with Decoupled Viewpoint Control in Immersive Humanoid Robot Teleoperation using SLAM
Yang Chen*, Leyuan Sun* (equal contribution), Mehdi Benallegue, Rafael Cisneros Limon, Rohan Pratap Singh, Kenji Kaneko, Arnaud TANGUY, Guillaume Caron, Kenji Suzuki, Abderrahmane Kheddar and Fumio Kanehiro
[PDF] [Youtube full demo] [ANA Avatar XPRIZE team web] [Project page]

PSNet

RSJ 2019 Robust SLAM in Dynamic Environment based on Object’s Mask
Leyuan Sun*, and Fumio Kanehiro
[PDF] [International Session Best Presentation Award Finalist]

🏆 Selected Honors and Grants

  • 2019.09: International Session Best Presentation Award Finalist of the 37th Annual Conference of the Robotics Society of Japan (RSJ 2019)

  • 2022.12: Best Conference Paper of 18th IEEE International Conference on Robotics and Biomimetics (ROBIO 2022).

  • 2021.10 - 2023.3: JST Support for Pioneering Research Initiated by the Next Generation: SPRING,
    日本文部科学省（日本教育部），Japan Science and Technology Agency（国立研究開発法人 科学技術振興機構），Grant Number JPMJSP2124, 主持，已结题。
    Research topic: Study on deep learning-based multi-sensor fusion with uncertainty and attention awareness for odometry estimation in challenging environment
    Grant: 1,000,000 Japanese yen
    Scholarship: 200,000 Japanese yen per month

  • 2023.4 - 2025.3：NEDO3.0実世界の観測やデジタルルインに基づく空間・人・もの・行為等の記号的表現会得の研究プロジェクト, NEDO 国立研究開発法人 新エネルギー・産業技術総合開発機構, 参与。
    Research topic: Object-goal navigation

  • 2023.4 - 2025.3：Japan Society for the Promotion of Science (JSPS 日本学術振興会) KAKENHI, Grant Number 23H03426, 参与。
    Research topic: 大規模学習モデルと３Ｄ仮想空間をつなぐ実世界AIのゼロショット学習. (Zeroshot learning of real-world AI by fusing large deep learning models and 3D virtual world)

💼 Employments

📖 Educations

💻 Academic Service

Reviewer:

  • IEEE Transactions on Circuits and Systems for Video Technology
  • Advanced Engineering Informatics
  • IEEE Robotics and Automation Letters (RA-L)
  • Robotics and Autonomous Systems
  • IEEE Sensors Journal
  • Measurement
  • Expert Systems with Applications
  • Scientific Reports
  • Multimedia Systems
  • Advanced Robotics
  • Winter Conference on Applications of Computer Vision (WACV)
  • The International Conference on Robotics and Automation (ICRA)
  • The International Conference on Intelligent Robots and Systems (IROS)
  • The International Conference on Humanoid Robots (Humanoids)
  • …

🏠 Family Members

alt text




Flag Counter