Papers
5
Total Citations
32
H-Index
2
About
Zaolin Pan is a rising researcher at the forefront of intelligent construction robotics, focusing on human-robot collaboration (HRC) and autonomous task execution. His work bridges computer vision, deep reinforcement learning (RL), and multi-agent systems to enable robots to understand and assist in complex, unstructured construction environments. Pan’s most influential paper, “Learning multi-granular worker intentions from incomplete visual observations for worker-robot collaboration” (2023, 18 citations), tackles the critical challenge of enabling robots to infer human intent from partial data—a key step toward seamless teamwork. He further advances the field by optimizing heterogeneous multi-robot teams for long-horizon tasks (2024, 8 citations), using time- and utilization-guided simulations to improve efficiency. His recent work on teaching robot end effectors to grasp irregular construction tools via deep RL (2025) addresses the delicate manipulation of objects like hammers and drills, while his training-free few-shot detection method (2025) leverages pre-trained vision-language models for rapid tool and material recognition. By learning multi-granularity task primitives from construction videos (2024), Pan is pioneering methods for robots to decode implicit task flows, directly tackling the industry’s productivity and safety challenges. With over 30 citations to date, his research is shaping the future of intelligent, collaborative construction sites.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5