News
-
Feb 2026
-
Feb 2026
Advanced Robotics & Embodied AI
My research focuses on vision-driven closed-loop robotic systems for real-world manipulation and assembly, with a specific emphasis on construction-scale tasks. By integrating 6D pose estimation, pose-based visual servoing (PBVS), and reinforcement learning across both task planning and motion planning layers, the developed systems achieve high-precision execution under real-world uncertainty — including occlusion, geometric variability, and unstructured environments. My work emphasizes seamless sim-to-real transfer of intelligent policies from high-fidelity digital twins to large-scale industrial workstations.
Looking ahead, I am exploring how Vision-Language-Action (VLA) models can serve as high-level semantic planners to further extend these systems toward open-vocabulary instruction following and broader generalization — with the goal of enabling autonomous, high-precision assembly of large-scale architectural structures.
Featured Projects
High-Precision Real-Time Computer Vision 6-DoF Pose Estimation Algorithm Demo
Education
-
University of Florida · Gainesville, FL Expected May 2027
-
University of Florida · Gainesville, FL Dec 2023
-
Southwest Jiaotong University · Chengdu, China Jun 2013
Awards & Honors
Research Interests
- Embodied AI & VLA Models: Developing Vision-Language-Action architectures that bridge high-level semantic reasoning with low-level motor control for generalizable task planning.
- 3D Scene Understanding for Action: Engineering robust 6DoF object pose estimation and tracking algorithms using RGB-D data and point cloud geometry to enable precise manipulation.
- Active Perception & Object-level Reasoning: Leveraging geometric constraints and semantic priors to resolve state estimation under heavy occlusion and environmental uncertainty.
- Closed-loop Visual Servoing: Implementing adaptive control laws and real-time feedback loops for precision-critical assembly and construction-scale tasks.
- Robotic Assembly under Uncertainty: Addressing complex challenges in symmetry, tight tolerances, and imperfect components through perception-action coupling.
- Perception-Planning-Execution Co-design: Architecting integrated systems that ensure reliable deployment on large-scale industrial workstations.
- Digital Twins & Sim-to-Real Pipelines: Building high-fidelity simulations to facilitate the scalable development and transfer of intelligent robotic policies.
Selected Publications
Advancing Robotic Automation in Wood-Framed Construction Using Vision-Driven Adaptive Control
A Framework for Automated Quality Control of Wood-Framed Panels in Robotic-Based Manufacturing Using Computer Vision and Deep Learning
Quantitative Analysis of the Role of Robotics in Promoting Industrialized Construction for Resilient Post-Disaster Recovery: Exploratory Research
Digital Twin-Enabled Adaptive Control for Perception-Driven Robotic Assembly in Industrialized Construction
Bridging the Gap: A Framework for Robotics Education in Industrialized Construction
Teaching
-
BCN4612C — Construction Estimating 2
-
BCN5905 — Advanced Construction Technology (Autodesk-Funded Course)
Contact
I welcome collaboration and invited talk opportunities.