PhD, Department of Mechanical Engineering
2024-10-01
The University of Tokyo
MEng, Department of Mechanical Systems Engineering
2022-10-01
2024-09-30
Nagoya University
BEng, Mechanical Design, Manufacturing and Automation
2018-09-01
2022-06-30
Hunan University
My research focuses on vision-language-action (VLA) models for robots — enabling aerial and ground robots to understand natural-language instructions and act autonomously in the real world. I am also interested in vision-and-language navigation and learning-based robot control.
Please feel free to reach out if you are interested in collaboration 😃
A spatially queryable sector flow-field representation for local language-conditioned navigation, trained with frame-level supervision distilled from VLN-CE episodes.
Reformulates language-conditioned navigation as learning a continuous flow field over the workspace: trajectories from any start point by field integration, trained on 500k+ …
A hierarchical LLM + VLM framework for an aerial-ground heterogeneous robot team: the drone builds a global semantic map and guides the ground robot's local semantic navigation and …
Improves LLM-based autonomous manipulation by combining GPT-4 task decomposition and YOLO-based perception with human-robot collaboration via teleoperation and dynamic movement …
Our paper "CoFL-S: Spatially Queryable Sector Flow Fields for Local Language-Conditioned Navigation" has been accepted to the Conference on Robot Learning (CoRL) 2026.
