Enhanced Attitude Control of Unmanned Aerial Vehicles Based on Virtual Angular Accelerometer

The small unmanned aerial vehicles are susceptible to wind disturbances due to the turbulent environment at their low operational altitude. Compared to conventional IMU output, the phase-advanced information provided by angular acceleration measurement is more beneficial for flight stability. This paper presents a general and systematic approach to enhance the attitude stability of small fixed-wing unmanned aerial vehicles. The methodology begins by designing a virtual angular accelerometer (VAA), which is based on 4 MEMS IMUs measurement fusion. Then, the angular acceleration is exploited to estimate and compensate disturbances. Finally, an enhanced attitude control scheme is developed to improve attitude tracking performance, where linear or nonlinear control approaches can be adopted. A VAA-based backstepping controller (BS) is designed and its stability is analyzed via Lyapunov function. Three simulation cases are carried out to verify the performance of tracking accuracy, robustness, and disturbance rejection. PID and backstepping methods are applied to the scheme. As a comparison, a widely used nonlinear disturbance observer (NDO)-based robust controller (RC) is studied. The simulation results show the excellent performance of the sensor-based VAA to observe the fast time-varying disturbance, where the NDO causes the time delay and phase lag. The tracking accuracy of the enhanced controllers (PID+VAA, BS+VAA, and RC+VAA), compared to the basic controllers, are improved by 14.3%-93.7% and the closed-loop control systems gain notable robustness.