Humanoid Manipulation
I am interested in manipulation in cluttered and dynamic environments, especially when interaction with movable objects is part of the task rather than a nuisance to avoid.
Incoming PhD Student, NUS Computer Science
Building safer and sharper robotic intelligence.
I am Yijie Liao, an incoming PhD student in Computer Science at the National University of Singapore, where I will join Professor Shao Lin's group. My research sits at the intersection of control, learning, and robotics, with a focus on humanoid manipulation, autonomous driving, and safety-critical systems that demand both precision and reliability.
- Shanghai Jiao Tong University Mechanical Engineering, 2022-2026
- University of Michigan Computer Engineering, 2024-2026
- Interested in world models for safety-critical robotics
About
Background and motivation
I am currently completing a dual-degree undergraduate education across the UM-SJTU Joint Institute at Shanghai Jiao Tong University and the University of Michigan. At SJTU, I study Mechanical Engineering from 2022 to 2026, and at Michigan, I pursue Computer Engineering from 2024 to 2026.
My long-term goal is to develop robotic systems that can act effectively in complex, interactive environments without sacrificing safety. I am especially interested in critical robotic systems where robustness, precision, and interpretable decision-making matter as much as raw capability.
During my undergraduate years, I have worked with Professor Zhongqiang Ren at Shanghai Jiao Tong University, Professor Tulga Ersal at the University of Michigan, and Professor Changliu Liu at the Carnegie Mellon University Robotics Institute during a summer research internship. These experiences shaped my interest in planning, control, embodied intelligence, and reliable autonomy.
Research
Current interests
Autonomous Driving
I study decision-making and control for autonomous systems operating around uncertainty, human interaction, and real-world constraints where failure is expensive.
Safety-Critical World Models
I want to explore world models for robotic systems that must be accurate, safety-aware, and actionable enough to support planning and control in critical settings.
Publications
Selected work
2025
Search-Based Path Planning in Interactive Environments among Movable Obstacles
Zhongqiang Ren, Bunyod Suvonov, Guofei Chen, Botao He, Yijie Liao, Cornelia Fermuller, Ji Zhang
IEEE International Conference on Robotics and Automation (ICRA), 2025
Research on planning in interactive environments with movable obstacles.
Beyond Research
Personal interests
- K-pop fan, especially TWICE
- Football enthusiast and Cristiano Ronaldo supporter