基于扩张观测器的四旋翼无人机轨迹鲁棒滑模控制

Robust sliding mode control for quadrotor UAV trajectory based on extended state observer

针对四旋翼无人机轨迹追踪问题,提出了一种基于扩张状态观测器的鲁棒滑模控制方法。考虑无人机系统受到内外部扰动、线速度未知等不确定性影响,通过引入扩张状态观测器,对系统不确定因素进行实时估计并给予补偿,实现了系统对扰动的鲁棒性和对环境的高度适应性。同时,滑模控制通过引入切换函数来消除干扰及不确定项,但较大的切换增益会引起系统颤振,因此,干扰和不确定项是颤振的主要来源,利用扩张状态观测器来估计干扰及不确定项并加以补偿,消除了颤振。利用Lyapunov理论,证明了控制系统的稳定性。系统仿真实验结果表明,所提出的控制方法能够保证四旋翼无人机轨迹追踪的鲁棒性,旋翼转速最大跳变幅值降低86.4%~94.5%,提高了系统稳定性。

For the robust trajectory tracking of the quadrotor vehicle, an extended state observer（ESO） based sliding mode control（SMC） is proposed. In the presence of disturbances and unavailable linear velocity, an extended state observer is introduced in the SMC to estimate the uncertain instabilities, with their estimations being compensated in the control, which ensures the quadrotor system robustness and survivability to environment. Besides, the SMC can counteract uncertainties by introducing switching functions. However, the large switching gain may cause system chattering. The incorporation of ESO into SMC alleviates the chattering, due to the superior estimation of uncertainties and their timely compensation in the control. The stability of the controlled system is proved by Lyapunov theorem. Simulation experiment results verify the robustness of the proposed control strategy, and show that the maximum jump-change of the rotor velocity is decreased by 86.4%-94.5%, which improves the system stability.