Kunmiao Tian
Papers
1
Total Citations
2
H-Index
1
About
Kunmiao Tian is a rising researcher in robotics and artificial intelligence, with a primary focus on deformable object manipulation and graph-based learning for dynamic systems. Their most notable contribution is the development of a Global-Local Graph Attention Network for modeling deformable linear objects (DLOs) interacting with their environment—a critical challenge for robotic manipulation tasks like cable routing or surgical thread handling. This work, published in 2024 and already garnering 2 citations, addresses a key limitation in existing Graph Neural Network (GNN) approaches: the inability to effectively capture both local particle interactions and global structural dynamics during deformation. By introducing an attention mechanism that balances these scales, Tian’s framework enables more accurate and stable predictions of DLO behavior under environmental constraints, paving the way for advanced robot control strategies. Their research sits at the intersection of geometric deep learning, physics simulation, and robot manipulation, offering practical solutions for automating tasks involving flexible materials. As a young investigator, Tian’s work demonstrates significant potential to influence future developments in soft robotics and automated manufacturing.
Research Focus
Key Achievements
Top Papers
- 1