I am a third-year undergraduate student at Xinjiang University, currently seeking a PhD position for Fall 2027 and research assistant opportunities.
I previously worked on robotic manipulation at Tsinghua AIR and Lightwheel Intelligence, and on medical image processing at the School of Computer Science, Chinese Academy of Sciences.
In addition, I participated in the RoboCon competition, which marked the starting point of my passion for computer vision and embodied intelligence. Computer vision and embodied intelligence is all my love.
💬 I am open to discussions and actively seeking collaborators! Feel free to reach out if you're interested in working together on exciting projects.
We propose RoboChemist, a framework for autonomous robotic chemical experimentation with long-horizon planning and safety compliance.
Short Bio
I am a researcher focusing on computer vision and embodied intelligence, with particular interests in low-light image enhancement and robotic manipulation. I have listed my detailed research interests below.
Research Interests
My primary research focuses are at the intersection of Computer Vision and Embodied Intelligence, aiming for generalizable and robust robotic systems:
1. Intuitive Physics: My work focuses on endowing robotic systems with intuitive physics knowledge, aiming to establish a priori understanding of object properties and dynamics. This is crucial for enhancing the robustness and long-horizon planning of robots, particularly in tasks involving deformable object manipulation and complex, contact-rich operations.
2. VLA-RL Integration: My goal is to design a VLA-RL integrated framework that enables a spatial-aware, dexterous manipulation model capable of long-horizon operations, primarily by distilling operational knowledge from large-scale human demonstration data.
3. World Modeling & Closed-Loop Autonomy: Driving Continuous Self-Improvement. My long-term research goal is to develop predictive world models that enable robots to perform efficient environment-grounded planning and reasoning. By constructing a closed-loop learning paradigm, I facilitate the iterative optimization of robots, transitioning smoothly from human-derived priors to self-generated robotic experience.
Beyond research, I enjoy playing the violin 🎻, running marathons 🏃, and capturing moments through photography 📷. These hobbies inspire my creativity and keep me balanced.
