Three Nested Kalman Filters-Based Algorithm for Real-Time Estimation of Optical Flow, UAV Motion and Obstacles Detection

We aim at developing a vision-based autopilot for autonomous small aerial vehicle applications. This paper presents a new approach for the estimation of optical flow, aircraft motion and scene structure (range map), using monocular vision and inertial data. The proposed algorithm is based on 3 nested Kalman filters (3NKF) and results in an efficient and robust estimation process. The 3NKF-based algorithm was tested extensively in simulation using synthetic images, and in real-time experiments.