FDCL congratulates Beomyeol (“Ben”) Yu and Maneesha (“Maneesh”) Wickramasuriya on successfully defending their doctoral dissertations at The George Washington University. Both conducted their research under the direction of Professor Taeyoung Lee, advancing complementary approaches to autonomous flight.
Ben defended his dissertation, Geometric Deep Reinforcement Learning for Quadrotor UAV, on April 1, 2026. His research incorporates the geometry and physical symmetries of quadrotor dynamics into reinforcement learning, helping flight controllers learn more efficiently. He developed modular learning architectures, neural networks that respect these symmetries, and methods for transferring controllers from simulation to physical aircraft. Validated through simulations and flight experiments, his work connects mathematical structure with practical learning-based control, including the challenging task of balancing an inverted pendulum on a flying quadrotor.
Maneesh defended his dissertation, Deep Transformer Network for Autonomous UAV Launch and Recovery in Ocean Environments, on July 15, 2026. His research enables UAVs to estimate their position and orientation relative to a ship using a single onboard camera. It combines transformer-based visual perception, geometric state estimation, and photorealistic mixed reality testing. The resulting system was validated through laboratory experiments and autonomous flights above a ship’s flight deck, demonstrating a path from synthetic training images to maritime operations.
Together, their dissertations reflect FDCL’s approach to autonomous systems: connecting mathematical foundations, artificial intelligence, computational methods, and experimental validation.
Congratulations, Ben and Maneesh, on these milestones, and thank you for your contributions to the lab’s research and community!