Dengqing Tang

National University of Defense Technology

Papers

Total Citations

H-Index

About

Dengqing Tang is a robotics researcher whose work bridges the gap between autonomous systems and practical manipulation, with a focus on bio-inspired control, visual perception, and calibration. His key research areas include deep reinforcement learning for robotic locomotion, stereo vision for unmanned aerial vehicles, and robust extrinsic calibration for industrial pick-and-place tasks. Tang’s most notable contribution is his pioneering approach to end-to-end formation control for robotic fish, where he combined deep reinforcement learning with non-expert imitation learning—a method that achieved 21 citations and offers a scalable pathway for training multi-agent underwater systems without expert demonstrations. In the domain of visual robotics, his work on robust task-oriented markerless extrinsic calibration (9 citations) addresses the critical challenge of sensor noise in camera-robot systems, directly enhancing reliability in automated pick-and-place scenarios. Earlier, Tang demonstrated his systems engineering skills by developing an open-source ROS-based ground stereo vision detection system (5 citations) for cluttered environments, supporting autonomous landing of UAVs. His research consistently emphasizes practical, open-source implementations that lower barriers for other researchers, making him a valuable contributor to the fields of field robotics and intelligent control.