Nga Teng Chan
I am Nga Teng Chan (Louisa), a first-year MPhil student at HKUST, advised by Prof. Qifeng Chen. I received my Bachelor’s degree from Tsinghua University in 2025, where I completed my thesis under the supervision of Prof. Tao Zhang. Currently, I am also bridging academia and industry as a Research Intern at Shanghai Jushen Technology Co., Ltd..
From Control to Cognition: My Research Journey
My long-term goal is to build intelligent robots capable of autonomously exploring unstructured and dangerous environments, such as deep space.
My journey began with a focus on the “body” of the robot. During my undergraduate studies, I built a foundation in traditional robotics, working on the mechanical design and control of drones and lunar rovers.
However, I realized that a robust body needs an intelligent mind. A core question began to drive my research: “How can we enable robots to think and act like humans?” To explore this, I pivoted my focus to Robot Learning and Embodied AI. I studied different embodiments of intelligence — from dexterous robotic hands, to bimanual manipulation systems, and ultimately to humanoid robots. Through this progression, I became deeply interested in how perception, decision-making, and control can be unified into a coherent embodied system.
Current Focus
I am currently working to bridge the gap between high-level reasoning (the “cerebrum”) and low-level motor control (the “cerebellum”). My research interests include:
- Humanoid Loco-manipulation — unifying locomotion and manipulation into coordinated whole-body intelligence.
- Continual Learning — enabling embodied agents to continuously adapt, acquire new skills, and retain prior knowledge over time.
Open for Opportunities
I am currently seeking PhD opportunities in robotics and embodied intelligence. If you are looking for prospective PhD students or see a potential research fit, please feel free to reach out.
news
| Jan 15, 2026 | 🎉 RoTri-Diff was accepted to ICRA 2026! |
|---|---|
| Oct 01, 2025 | Preprint “RoboHiMan” (Compositional Generalization in Long-horizon Manipulation) was released on arXiv. |
| Jun 01, 2025 | 🎓 Graduated from Tsinghua University with a Bachelor’s degree! |
selected publications
- arXivRobo-Cortex: A Self-Evolving Embodied Agent via Dual-Grain Cognitive Memory and Autonomous Knowledge InductionarXiv preprint arXiv:2605.18729, 2026
- ICRARoTri-Diff: A Spatial Robot–Object Triadic Interaction-Guided Diffusion Model for Bimanual ManipulationIn Proceedings of the 2026 IEEE International Conference on Robotics and Automation (ICRA), 2026