In the paper, a prescribed chattering reduction control using aperiodic signal updating is presented for quadrotors subject to parameter uncertainties and external disturbances. Using estimation errors instead of tracking errors to update adaptive laws, estimator-based minimum learning parameter (EMLP) observers capable of relaxing computational complexity are respectively explored in translational and rotational loops to reject fast time-varying disturbances, such that transient oscillations can be efficiently mitigated even with a large adaptive gain. Meanwhile, quantitative analysis for transient learning performance is characterized by means of L₂ norms of time differential of neural network weights. With the aid of disturbance estimates, a relative event-triggered robust control law is derived by inserting a compensation term to guarantee a favorable trajectory tracking with Zeno free behaviors and decreased sampling cost. Besides, an appointed-time prescribed performance control (APPC) is established, enforcing trajectory tracking errors to evolve within pre-given regions even in face of triggering errors, where a piecewise and continuous finite-time behavior function, rather than an exponential decaying function, is applied to enable a preassigned fast convergence time without retuning controller parameters. Finally, the stability of closed-loop system is proved via Lyapunov synthesis, while comparative studies are provided to validate the effectiveness of presented control method.

在本文中，针对具有参数不确定性和外部干扰的四旋翼飞机，提出了一种使用非周期性信号更新的规定的颤振减小控制方法。使用估计误差而不是跟踪误差来更新自适应律，分别在平移和旋转环路​​中探索了能够减轻计算复杂度的基于估计器的最小学习参数（EMLP）观察器，以拒绝快速时变扰动，从而可以有效地实现瞬态振荡即使具有较大的自适应增益也可以缓解。同时，利用L ₂表征了瞬时学习成绩的定量分析。神经网络权重的时差规范。借助干扰估计，通过插入补偿项来推导相对事件触发的鲁棒控制定律，以确保具有Zeno自由行为并降低采样成本的有利轨迹跟踪。此外，建立了指定时间的规定性能控制（APPC），即使在遇到触发错误的情况下，轨迹跟踪错误也会在给定区域内演变，在此情况下，分段和连续的有限时间行为函数而不是指数衰减函数，用于启用预分配的快速收敛时间，而无需重新调整控制器参数。最后，通过Lyapunov综合证明了闭环系统的稳定性，同时提供了比较研究以验证所提出的控制方法的有效性。