I am a fifth-year Ph.D. candidate in Mechanical Engineering at Beijing Institute of Technology (BIT), advised by Prof. Qiang Huang and affiliated with the BHR Lab. I received my B.E. in Mechatronic Engineering from BIT (2021). I am currently a research intern in embodied intelligence at The Hong Kong University of Science and Technology (Guangzhou), HKUST(GZ).
Over the past few years, I have contributed to the development and deployment of nine BHR-series humanoid robots across three generations and two commercial platforms, through seven research projects and one industrial project.
I expect to complete my Ph.D. in 2027 and am actively seeking opportunities in both academia and industry. Please feel free to contact me if you are interested in my work.
My research focuses on motion planning and control for humanoid and wheel-biped robots, involving model-based optimization, whole-body control, reinforcement learning, and sim-to-real deployment to achieve robust and adaptive locomotion.
Recently, I have been working on fusing model- and learning-based control, e.g. data-driven dynamics representation, to improve the efficiency and reliability of humanoid motions.
I am also interested in bridging high-level embodied intelligence, such as vision-language-action models, world/action models, and task reasoning, with low-level whole-body control, enabling humanoid robots to transform perception and intent into physically grounded motion skills.
First-author journal manuscript under second review on terrain-aware bipedal gait generation for mixed terrain.
First-author SCI journal paper on locomotion strategy for transformable wheel-biped humanoid robots.
Multimodal heterogeneous reconfiguration control for legged robots, covering complex-terrain walking, fall recovery, crawling, and vehicle riding/separation.
Mixed-terrain foot-ground interaction control for humanoid robots, including stairs, slopes, low obstacles, precise target reaching, and dynamic environment interaction.
System integration, real-time software architecture, wheel-foot mechanism design, and transformable locomotion control for a humanoid robot platform.
Low-posture crawling, slope climbing, omnidirectional biped walking, and dynamic obstacle avoidance for a primate-inspired mobile robot.
Motion-control algorithm work for World Robot Conference demonstrations and special humanoid robot project delivery.
Beginner guide for deploying an always-online OpenClaw AI assistant, covering platform deployment, API configuration, and platform integration.
Python visualizer and editor for AMP motion datasets, built for legged robot locomotion and trajectory optimization research.
C++ A* footstep planner for humanoid robots on complex terrain with kinematic constraints.
Lightweight C++ geometry and vector math library for robotics, including 2D/3D primitives, convex polygons, and spatial queries.
Bilingual educational tutorials for CoppeliaSim robotics simulation, including presentations, code examples, and simulation models.
Working on cerebrum-cerebellum coordinated humanoid control and data-driven dynamics representation.
The organization includes robot-motion-player, AStarFootstepPlanner, Heuclid, CoppeliaSim tutorials, and OpenClaw deployment notes.
The paper studies high-mobility locomotion and adaptive mode transitions for transformable wheel-biped humanoid robots.
The project focused on terrain-aware foot-ground interaction control for humanoid robots in mixed unstructured terrain.