Unmanned system airspeed estimation and prediction with fusion filtering algorithm

For unmanned aerial vehicle (UAV), airspeed sensor faults are commonly happening because of incidents. When the Pitot tube is clogging or icing, the airspeed sensor will return data with faults. This will cause a misoperation or system fault of the flight control system which can finally crash the UAV. Estimating airspeed of this moment and next moment with other sensor readings provides a basis to judge if the airspeed sensor is working properly. This paper designs a method of airspeed estimation and prediction with Kalman Filter (KF) and Least Squares Support Vector Machine (LSSVM). It is tested on a software in the loop (SITL) platform. The result of the estimation is tallied with the airspeed which is gathered by the airspeed sensor.