Jian Cheng

University of Electronic Science and Technology of China, University of Shanghai for Science and Technology

Papers

Total Citations

107

H-Index

About

Jian Cheng is a versatile robotics and computer vision researcher whose work spans intelligent tracking systems, autonomous navigation, and miniature medical robotics. His most significant contribution to date is his 2018 paper on Deep Continuous Conditional Random Fields with Asymmetric Inter-Object Constraints for Online Multi-Object Tracking, which garnered 90 citations and addressed a critical limitation in existing tracking methods by elegantly unifying individual object motion modeling with inter-object relational constraints within a single deep learning framework — advancing applications in surveillance, autonomous driving, and robot navigation. Cheng has also made meaningful contributions to mobile robotics, proposing a hybrid obstacle avoidance algorithm that combines artificial potential field methods with deep reinforcement learning to enhance the safety and reliability of autonomous path planning. Earlier in his career, he pioneered work on flexible miniature pneumatic robots designed to navigate human body cavities, developing both electro-pneumatic pressure servo-control systems and high-speed electromagnetic valve controllers to enable precise inchworm-like locomotion. Together, his body of work reflects a sustained commitment to bridging intelligent perception, autonomous decision-making, and precision actuation across a compelling range of real-world robotic challenges.