Building autonomy for machines that must see, and fly, when the world is trying to deceive them.

about

I work on the hard end of drone autonomy - passive electro-optical detection at range, synthetic data, and reinforcement learning for flight under adversarial conditions. Formerly Cranfield University, Centre for Autonomous & Cyber-Physical Systems.

selected work
journal · J. Intell. Robot. Syst., 2025
Benchmarking Deep Reinforcement Learning for Navigation in Denied Sensor Environments
M. G, P. Chatzithanos, W. Guo, A. Tsourdos
journal · Drones 8(6):235, 2024
Towards Fully Autonomous Drone Tracking by a Reinforcement Learning Agent Controlling a Pan–Tilt–Zoom Camera
M. G, Z. A. Rana, I. Petrunin, A. Holt, S. Harman
preprint · arXiv:2411.09077, 2024
Drone Detection using Deep Neural Networks Trained on Pure Synthetic Data
M. G, Z. A. Rana, I. Petrunin, A. Holt, S. Harman
publications
2026 Deep & Hierarchical RL for UAV Autonomy under Adversarial Attacks: A Survey in review 2025 Benchmarking Deep Reinforcement Learning for Navigation in Denied Sensor Environments JIRS 2024 Towards Fully Autonomous Drone Tracking by a RL Agent Controlling a Pan–Tilt–Zoom Camera Drones 2024 Drone Detection using Deep Neural Networks Trained on Pure Synthetic Data arXiv 2022 Drone Model Classification using a Convolutional Neural Network Trained on Synthetic Data J. Imaging