Motion estimation of UAV based on video streaming

Abstract

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