Disturbance estimation for RUAV using UKF with acceleration measurement

Rotorcraft unmanned aerial vehicles (RUAV) have been widely used in many fields. However, one of the bottleneck problems of RUAV control is how to ensure its motion steady under huge external disturbances such as wind. In this paper, a new algorithm on disturbance estimation is proposed using UKF algorithm and acceleration measurements. Firstly, the disturbance is regarded as an expanded state of the attitude dynamics of RUAV which is actuated by white noises; subsequently, the expanded dynamic system is rewritten into discrete form thus the UKF strategy can be utilized; then, UKFs with different information in the measurement equations are conducted to estimate the disturbance; finally, simulations are carried out and the results show that UKF with measurements of both Euler angles and angular accelerations has the fastest response to disturbance tracking, the least mean square of estimation errors and certain extent of robustness since acceleration is supposed to be a direct reflection of disturbance.