Motion estimation of UAV based on video streaming

摘要

This paper describes a way of the motion estimation of four-rotor UAVrnbased on video streaming, mainly researches the real-time vision algorithms,rncomprehensively completes video stream computing and understanding, andrncompletes the reconstruction of UAV three-dimensional motion andrnstructure. The study gives a solution of motion estimation of UAV that it isrnfull embedded, light weight, low-cost autonomous positioning based onrnvideo streaming in arbitrary and unknown environment. In this paper, therncalculation is based on the triple nested Kalman filter (3NKF) and 3NKFrncan synthesis three distinct areas (video stream computation, data fusion,rnrestore the structure from motion) in a proper way.