I build intelligent aerial robots that see, think, and act—pushing the boundaries of AI-driven autonomy.
Research projects completed as part of graduate thesis work in aerial robotics and learning-based autonomy.
Designed a fault-tolerant quadrotor controller with deep reinforcement learning, exploring both the potential and practical limits of learned recovery strategies.
Developed a single-shot UAV path planner that combines supervised learning and reinforcement learning in an image-to-image planning framework.
Side explorations used to prototype ideas in perception, mapping, and autonomous navigation.
Built an autonomous driving prototype that maps its surroundings, follows a predefined track, avoids collisions, stays on the road, and reacts to traffic signals.
Combined computer vision and geometric modeling to reconstruct point clouds from disparity images and generate 3D surface meshes.