Two-stage Kalman filter for estimating the actuator control effectiveness factors of UAV

摘要

The purpose of this study is to build a two-stage Kalman filter (TSKF) for estimating the actuator control effectiveness factors of an unmanned aerial vehicle (UAV). The actuator faults can be determined and isolated using controller effectiveness factor estimates. A linear quadratic regulator (LQR) controller is designed for the modelled UAV. The control matrix is identified by the TSKF. In the case of actuator faults, LQR controller can be reconfigured for the identified control distribution matrix. For faulty actuator scenarios (change in elevator, throttle, aileron control effectiveness factors are taken into account), the designed system is tested, and the success of TSKF is shown. The closed-loop systems which include LQR-type controllers in the longitudinal and lateral dynamics are considered and simulations are done in this system to estimate the true states and the control effectiveness factors. True values of the effectiveness factors are estimated, and it is shown that the estimation of the states can be done effectively using TSKF for lateral and longitudinal models.