论文部分内容阅读
针对无人机防滑刹车系统工作过程中同时出现系统输出滑移率稳定区域受限、控制输入饱和与刹车执行机构故障的多重约束问题,提出了一种基于障碍Lyapunov形式的自适应神经网络反演容错控制器的设计方法。当刹车执行机构发生故障时,通过自适应神经网络补偿刹车系统中的非线性及不确定项。根据反演设计原理,应用神经网络输出设计相应的容错控制律,同时,在控制器的设计中引入鲁棒切换控制项,优化系统快速容错的暂态性能。首先本文设计的容错控制器无需精确获取执行机构在线故障的重构信息,也能使刹车闭环系统能够快速稳定,然后基于Lyapunov方法分析了系统的稳定性,最后通过数值仿真结果表明,所提出的容错控制算法能够有效地保证刹车执行机构故障时控制系统的稳定性和有效性。
Aimed at the multi-constraint problem that the stability of the output slip ratio is limited, the saturation of control input and the failure of the brake actuator simultaneously, the adaptive neural network inversion based on the barrier Lyapunov form is proposed. Fault-tolerant controller design method. When the brake actuator failure, the adaptive neural network to compensate for the nonlinear and uncertainties in the braking system. According to the principle of inversion design, the corresponding fault-tolerant control law is designed based on neural network output. At the same time, the robust switching control is introduced into the controller design to optimize the transient performance of the system. First of all, the fault-tolerant controller designed in this paper does not need to obtain the reconstruction information of the actuator online fault accurately, and also can make the closed-loop braking system quickly stabilize. Then the stability of the system is analyzed based on Lyapunov method. Finally, the numerical simulation results show that the proposed The fault-tolerant control algorithm can effectively ensure the stability and effectiveness of the control system when the brake actuator fails.