Papers
29
Total Citations
595
H-Index
12
About
Deok Jin Lee is a prominent researcher at the intersection of autonomous robotics, aerial systems, and artificial intelligence, with particular expertise in deep learning, reinforcement learning, and unmanned aerial vehicles (UAVs). His work has made substantial contributions to enabling robots and drones to perceive, navigate, and act intelligently in complex real-world environments. Lee's most influential contribution, a 2019 paper on deep learning-based real-time multiple-object detection and tracking from aerial imagery using GPU-embedded drone systems, has garnered 180 citations and exemplifies his ability to bridge cutting-edge AI with practical hardware deployment. His earlier work on optimal servicing of geosynchronous satellites (79 citations) demonstrates a breadth spanning aerospace mission planning as well. Throughout the 2018–2022 period, Lee produced a remarkable series of deep reinforcement learning studies addressing end-to-end autonomous navigation, collision avoidance, trajectory tracking, and map-less flight control for aerial robots, collectively amassing over 140 citations. His research consistently pushes toward real-world applicability, incorporating live flight experiments, multi-objective reward frameworks, and adaptive controllers for nonlinear systems. Additional contributions in voice-controlled UAVs and sensor-fusion-based motion planning further illustrate his commitment to human-robot interaction and practical autonomy, making his body of work highly valuable for students and researchers advancing intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Optimal Servicing of Geosynchronous Satellites79 citations · 2006
- 3
- 4
- 5
- 6
- 7Voice enabled smart drone control25 citations · 2017
- 8
- 9
- 10