Compensation function observer-based model-compensation backstepping control and application in anti-inference of quadrotor UAV

This study has considered the dynamical equations of a quadrotor unmanned aerial vehicle (UAV) with model uncertainties, time-varying loads and wind disturbances during actual flight. Then, a model-compensated control idea is innovatively proposed for quadrotor UAV attitude and position tracking control, and a compensation function observer-based model-compensated backstepping controller (CFO-based MC-BC) is designed on this basis. The benefits of this approach arc that the proposed controller not only retains the backstepping control strategy in terms of globally asymptotic stability based on Lyapunov criterion. As well as, a compensation function observer (CFO) with high estimation accuracy is used, which can make excellent use of the system state information to accurately estimate the modeling deviations, parameter variations with time, and external unknown disturbances of the system. And it is incorporated into the controller to dynamically offset this negative impact adaptively. This approach effectively addresses the issue of traditional backstepping control performance relying on modeling accuracy. In addition, this paper introduces a high-order differentiator (HOD) that extracts up to the derivatives of the signal with higher accuracy, effectively solving the issue of differential explosion. Finally, a large number of simulation results and experimental validation are presented. The results show that the proposed CFO-based MC-BC is superior than the other four schemes in accurate tracking performance, system transient performance, anti-interference ability, anti-time-varying-load, and unknown information estimation ability.

Automated summary

This study considers the dynamical equations of a quadrotor UAV with uncertainties and disturbances during flight. A model-compensated control idea is proposed for attitude and position tracking, and a controller is designed based on a compensation function observer. Simulation results and experimental validation demonstrate the superiority of this approach in terms of tracking performance, transient performance, interference ability, load variation resistance, and unknown information estimation.