Improved Kalman Algorithm For Vertical Height And Velocity Information Fusion Of Multi-Rotor Unmanned Aerial Vehicle

The paper presents an improved Kalman filter algorithm. It can ensure the precision and stability of the vertical height and velocity information of the multi-rotor unmanned aerial vehicle (UAV), when the statistical characteristics of the system noise and the measurement noise are unknown. The algorithm uses the accelerometer to provide prediction information and adopts the barometric altimeter to provide measurement information. It can estimates the system noise and the measurement noise based on the characteristics of these two sensors. The flight experiment and analysis shows that the obtained vertical height information error is less than 1m, the information update frequency can reach 100Hz, and the good fusion effect of vertical velocity information is obtained. The improved Kalman algorithm can meet the needs of the multi-rotor unmanned aerial vehicle flight control.