State estimation using inertial optical flows for a fixed-wing UAS

Autonomous navigation and obstacle avoidance in complex terrains such as canyons or urban/indoor environments is one of the biggest challenges for the next generation of unmanned aircraft systems. Insects such as honeybees use optical flow for various navigation tasks. Similar strategies can be used by unmanned aircraft as well. More importantly, optical flow can be combined with inertial measurements to achieve state estimation such as ground velocity and terrain shape. In this paper, a new state estimation algorithm is proposed for the estimation of ground speed and simple terrain shape using inertial optical flows. Epipolar constraints are combined with optical flow motion of equation for robust state estimation. Simulation results show the effectiveness of the proposed algorithm. I.