Zhengshen Zhang
Papers
2
Total Citations
8
H-Index
2
About
Zhengsheng Zhang is a robotics researcher advancing the frontier of autonomous manipulation through novel perception and learning frameworks. His primary research areas include 6-DoF robotic grasping, dynamic scene reconstruction, and learning from human demonstration. Zhang’s most impactful work, “You Only Scan Once” (2024, 6 citations), introduces a groundbreaking pipeline that reconstructs dynamic scenes from a single scan, enabling robust grasping of novel objects despite severe occlusion—a critical improvement over traditional partial-point-cloud methods. In complementary work on “Learning Grasping for Robot with Parallel Gripper from Human Demonstration via Contact Analysis” (2024, 2 citations), he pioneers a contact-aware approach that translates human grasping strategies into robot policies, moving beyond purely geometric feature-based grasp generation. By integrating real-time scene understanding with human-inspired learning, Zhang’s research directly addresses two fundamental bottlenecks in robotics: occlusion-robust perception and intuitive skill transfer. His work holds promise for advancing autonomous systems in cluttered, unstructured environments, from warehouse automation to assistive robotics. With a focus on practical, deployable solutions, Zhang is establishing himself as an emerging voice in intelligent robotic interaction.
Research Focus
Key Achievements
Top Papers
- 1
- 2