Pengwei Xie
Papers
10
Total Citations
89
H-Index
5
About
Pengwei Xie is a robotics researcher whose work focuses on advancing robotic manipulation in complex, real-world environments. His primary research areas include 6-DoF grasp detection, articulated object manipulation, and human-robot interaction. Xie's major contributions center on developing efficient, generalizable frameworks for robotic grasping and handover tasks. His most cited work, "Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes" (2023, 42 citations), introduces a fast and robust method for generating high-quality grasps by leveraging global semantic guidance from point clouds. He further pushes the boundaries of human-robot collaboration with "GenH2R" (2024, 15 citations), a scalable framework for learning generalizable handover skills that handle unseen object geometries and complex trajectories. Xie also addresses the challenge of manipulating novel articulated objects through his "Part-Guided 3D RL for Sim2Real" work (2023, 9 citations), which combines reinforcement learning with visual affordance guidance. His research consistently emphasizes real-time performance, sim-to-real transfer, and adaptability to cluttered, dynamic scenes, making significant strides toward practical, deployable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes42 citations · 2023
- 2
- 3Part-Guided 3D RL for Sim2Real Articulated Object Manipulation9 citations · 2023
- 4GAP-RL: Grasps as Points for RL Towards Dynamic Object Grasping5 citations · 2024
- 5
- 6
- 7
- 8
- 9
- 10Target-Oriented Object Grasping via Multimodal Human Guidance2 citations · 2025