Adaptive Visual-Inertial Navigation for Fixed-Wing Aircraft Landing in GPS-Denied Condition

Abstract

AbstractThe landing of fixed-wing aircraft is still one of the greatest challenges during the flight. Precise kinetic parameters are significant to control aircraft land accurately in GPS-denied environment. The work presented in the paper focuses on a visual-inertial optimization-based motion estimation method that integrates inertial data and vision cues extracted from FLIR video. The proposed method runs on two significant flight sub-stages: (1) final approach, sparse runway features, inertia data and runway geo-information are fused to estimate ego-motion; (2) final landing, direct sparse odometry and inertial data are used for motion estimation. A Y-12F aircraft is used to collect flight data and verify the proposed approach. Finally, the experiments on real flight data demonstrate that accurate pose estimation for aircraft landing can be achieved in GPS-denied condition.KeywordsMotion estimationLandingWindowed optimizationGPS-Denied condition