Papers
9
Total Citations
39
H-Index
4
About
Kaixin Bai is a robotics researcher whose work centers on bridging the sim-to-real gap for dexterous manipulation, with a particular focus on precise grasping in cluttered and delicate environments. Their major contributions include developing an enveloping gripper inspired by Asian elephant trunks for damage-less fruit grasping (11 citations), and creating a physically-based structured light synthetic data simulation to overcome data acquisition challenges in industrial robotics (8 citations). Bai’s research on model-free robotic grasping with sim-to-real transfer learning (6 citations) addresses the critical issue of sparse datasets and sensor errors, while their Collision-Aware Cable Grasping method (CG-CNN) enables robust manipulation of cables in cluttered spaces (4 citations). Further notable work includes ToolEENet for 6D tool pose estimation under occlusion (4 citations) and ContactDexNet, which leverages hand-object contact semantic mapping for multi-fingered grasping in cluttered settings (2 citations). Bai’s innovative combination of physics-based simulation, deep learning, and bio-inspired design has yielded practical solutions for agriculture, manufacturing, and logistics, earning recognition for advancing reliable robotic grasping in real-world conditions.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4A Collision-Aware Cable Grasping Method in Cluttered Environment4 citations · 2024
- 5ToolEENet: Tool Affordance 6D Pose Estimation4 citations · 2024
- 6
- 7
- 8
- 9