Hu Cheng
Papers
12
Total Citations
272
H-Index
8
About
Hu Cheng is a robotics researcher whose work sits at the intersection of computer vision, robot manipulation, and autonomous navigation. Best known for his pioneering contributions to robot grasping systems, Cheng has developed a series of increasingly sophisticated deep learning frameworks for grasp pose detection, progressing from anchor-based multi-scale detectors to single-stage anchor-free architectures capable of operating on single RGB or depth images. His most cited works — a vision-based grasping platform and an anchor-free grasp detector (59 and 56 citations respectively) — demonstrate both the breadth and practical relevance of his research. Complementing this focus on manipulation, his widely referenced 2018 survey on autonomous navigation in human environments (58 citations) has become a valuable resource for researchers tackling dynamic obstacle avoidance and service robotics. Cheng's work extends to real-world applications, including an airport trolley deployment robot and robust visual SLAM in dynamic scenes. Across his portfolio, he consistently bridges theoretical deep learning advances with deployable robotic systems, making meaningful contributions to the challenge of enabling robots to perceive, navigate, and interact reliably in unstructured, human-centered environments.
Research Focus
Key Achievements
Top Papers
- 1A Vision-Based Robot Grasping System59 citations · 2022
- 2Autonomous Navigation by Mobile Robots in Human Environments: A Survey58 citations · 2018
- 3A Robot Grasping System With Single-Stage Anchor-Free Deep Grasp Detector56 citations · 2022
- 4
- 5
- 6
- 7
- 8Grasp Pose Detection from a Single RGB Image12 citations · 2021
- 9Real-Time Robot End-Effector Pose Estimation with Deep Network7 citations · 2020
- 10