Real-time pose estimation and motion control for a quadrotor UAV

Real-time pose estimation and reliable motion control are two crucial tasks for a quadrotor UAV. In order to obtain accurate and robust pose estimation, an estimation algorithm based on Cubature Kalman Filter (CKF) is adopted in this paper. Considering the limitation of a quadrotor working in GPS-denied indoor environments, a sparse optical flow algorithm is utilized in our work to estimate the quadrotor's velocity and position. After acquiring the quadrotor's real-time pose information, the pose and position control of a micro quadrotor are accomplished by using PID and PI-D controller. A series of experiments are performed on a home-made quadrotor UAV named Smart-QR. The experimental results show the validity and practicability of the proposed approach.