四旋翼飞行器的自校正PID控制器设计

Design of Self-Tuning PID Controller for Quadrotor UAV

研究四旋翼飞行器姿态控制器的设计优化问题。四旋翼飞行器在飞行中由于气流扰动、电源电压变化等因素使模型参数发生变化时,常规的控制算法将导致姿态控制性能变差甚至失去平衡。模型参数变化时控制器参数在线调整算法的设计是难点。提出一种基于自校正PID控制的姿态控制算法,使用带遗忘因子的递推最小二乘估计算法在线辨识四旋翼飞行器各通道的CARMA模型参数,并结合极点配置自校正PID控制的递推算法实时调整各通道的控制器参数,达到飞行姿态平衡的有效控制。通过不同条件下的仿真,结果表明改进控制方法能够实现四旋翼飞行器在不确定性干扰和模型参数变化的情况下对飞行姿态的稳定控制。

The optimization problem of the attitude controller of the quadrotor UAV is studied. Because of the dis- turbance of air flow and the change of power supply voltage, the parameters of the model are changed. The conven- tional control algorithm can not adjust the parameters automatically, which will lead to the deterioration of the attitude control performance and even imbalance. It is difficult to design the on-line adjustment algorithm of controller param- eters when the model parameters change. An attitude control algorithm based on self-tuning PID control is proposed. The recursive least squares estimation algorithm with forgetting factor is used to identify the CARMA model parameters of the quadrotor UAV on-line, and the recursive algorithm of the self-tuning PID controller is used to adjust the con- troller parameters of each channel in real time to achieve effective control of flight attitude. Simulation results under different conditions show that the control method can achieve stable control of the flight attitude of the quadrotor UAV under the condition of uncertainty disturbance and model parameter variation.