Tianyu Sun

Papers

Total Citations

H-Index

About

Tianyu Sun’s research lies at the intersection of robotic manipulation, 3D computer vision, and intelligent manufacturing, with a focus on enabling robots to interact with complex, unstructured environments. His major contributions include pioneering the use of part-guided 3D reinforcement learning for Sim2Real transfer in articulated object manipulation—a breakthrough that allows robots to handle unseen objects through visual feedback alone, as demonstrated in his 2023 work (9 citations). Sun also advanced few-shot learning for 3D affordance segmentation, enabling robots to generalize manipulation skills to novel object categories with minimal training data (2025, 3 citations). In the domain of industrial robotics, he developed uncertainty-aware laser stripe segmentation with nonlocal mechanisms for welding robots, addressing the challenge of extracting clean signals from noisy, real-world welding environments (2025, 2 citations). Collectively, his work bridges the gap between simulation and reality, reduces data dependency, and enhances robustness in both service and industrial robotics. Sun’s research is particularly notable for its practical impact—tackling fundamental perception and control problems that directly enable more adaptable, autonomous robots in manufacturing and home settings.