Papers
10
Total Citations
245
H-Index
6
About
Hengli Wang is a robotics and computer vision researcher whose work sits at the intersection of autonomous navigation, deep learning, and intelligent transportation systems. His research spans several interconnected domains, including drivable area detection, optical flow estimation, autonomous racing, robotic path planning, and drone-based surveillance — all unified by a drive to make mobile robots more perceptive and self-sufficient in complex real-world environments. Wang's most influential contribution, "Dynamic Fusion Module Evolves Drivable Area and Road Anomaly Detection" (2021, 106 citations), introduced a benchmark and algorithmic framework that significantly advanced how autonomous vehicles jointly identify safe driving zones and unexpected road hazards using convolutional neural networks. His work on deep imitative reinforcement learning for autonomous car racing (82 citations) demonstrated how end-to-end systems can outpace traditional modular pipelines in dynamic, high-speed scenarios. He has also pioneered unsupervised optical flow methods, notably CoT-AMFlow, and applied these techniques to drone-based parking violation detection through his ATG-PVD framework. Earlier work on reconfigurable ankle rehabilitation robots reflects a broader engineering versatility. Across his career, Wang has consistently translated sophisticated deep-learning theory into practical robotic applications, earning recognition from the broader autonomous systems research community.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8ATG-PVD: Ticketing Parking Violations on A Drone5 citations · 2020
- 9
- 10