Shipboard UAV launch & recovery
Monocular relative navigation (TNN-MO) for approach and recovery on a moving ship — validated in Chesapeake Bay flight tests.
TNN-MO on GitHub →Robotics · Drones · Rovers
Maneesha Wickramasuriya
Passionate about robots that fly and roll — drones, UAVs, and ground rovers — and the perception that keeps them autonomous in the real world.
Ph.D. work at GWU on vision, LiDAR, and RADAR for shipboard UAV launch and recovery, field-tested with NAVAIR and the U.S. Naval Academy.
About
Robotics is the through-line: building and flying drones, developing perception for GPS-denied maritime UAVs, and integrating ground rovers into human–AI teams. I care about systems that leave the lab — Jetson onboard compute, ROS 2, and ocean flight tests — not just simulation.
Ph.D., Mechanical and Aerospace Engineering, The George Washington University (Flight Dynamics and Control Lab). B.Sc. Engineering (Hons.), Mechanical Engineering, University of Peradeniya.
Projects
Vision-in-the-loop validation
Indoor closed-loop testing of the full perception–estimation–control stack against a photorealistic 3D Gaussian Splatting ship.
LiDAR pose estimation at sea
Deep Point Transformer for real-time 6-DoF UAV-to-ship pose from a single LiDAR in ocean conditions.
Project page →RADAR-based relative pose
Learning-based 6-DoF estimation from onboard RADAR for GPS-denied shipboard operations.
UGV / rover for human–AI teams
Ground rover (UGV) integration for collaborative autonomy under the ONR-funded CHASE program — robots that work with people in contested settings.
Program →Vision-based mobile robot obstacle avoidance
Early ground-robot work: CSI camera, Arduino + Raspberry Pi, PID control for onboard obstacle avoidance.
Publications
Google Scholar →- LiDAR-Based UAV Pose Estimation in Ocean Environment with Deep Point Transformer Network. Journal of Guidance, Control, and Dynamics, 2026. DOI
- Deep Transformer Network for Monocular Pose Estimation of Shipborne UAV. Journal of Guidance, Control, and Dynamics, 48(8), 2025. DOI
- Drone Operator Workload Analysis for Integration onto a Naval Vessel. Journal of Intelligent & Robotic Systems, 111(2), 2025. DOI
- Invariant Kalman Filter for Relative Dynamics. arXiv:2412.10519, 2024. arXiv
- Hardware- and Vision-in-the-Loop Validation of Deep Monocular Pose Estimation for Autonomous Maritime UAV Flight. ICUAS, 2026. IEEE Xplore
- Vision-in-the-Loop Simulation for Deep Monocular Pose Estimation of UAV in Ocean Environment. ICUAS, 2025. IEEE Xplore
- Deep Monocular Relative 6D Pose Estimation for Ship-Based Autonomous UAV. AIAA SciTech Forum, 2024. DOI
- Invariant Kalman Filter for Relative Dynamics. IFAC-PapersOnLine, 59(19), 2025. ScienceDirect
- Drone Operator Workload Analysis for Dynamic Interface Flight Test. AIAA SciTech Forum, 2024. DOI
- Optimizing Drone Operator Workload in the Terminal Area for Shipboard Operations. ICUAS, 2024. IEEE Xplore
- Verification of YP689 Flow Field Models for Dynamic Interface Flight Test. AIAA SciTech Forum, 2023. DOI