Composite Controller Design for Quadrotor UAVs With Uncertainties and Noises Based on Combined Kalman Filter and GPIO
Ting Li, Zhenhua Zhao, Shihong Ding, Jinya Su
Abstract
This article investigates the attitude control problem of quadrotor unmanned aerial vehicles (UAVs) with disturbances, actuator faults, and measurement noises. First, to avoid the noise amplification, the attitude system is transformed into a linear decoupled system, and the influences of uncertainties (including disturbances, actuator faults, and nonlinearities) are regarded as lumped disturbances. Second, to handle the uncertainties and noises simultaneously, the combined Kalman filter generalized proportional integral observer (KFGPIO) is introduced to estimate the filtered states and lumped disturbances. And then, a composite controller is proposed based on the estimation of KFGPIO. Simulation results validate that the proposed method achieves good disturbance rejection and fault tolerance performance and guarantees the continuity of control action even under serious measurement noises.