带有输入饱和及输出受限非仿射系统固定时间跟踪控制  

Fixed-time tracking control of pure-feedback system with input saturation and output constraints

作　　者：黄燕华 代冀阳[1] 应进[1] 蒋沅[1] 李叶鼎 鲁亮亮 HUANG Yan-hua;DAI Ji-yang;YING Jin;JIANG Yuan;LI Ye-ding;LU Liang-liang(School of Information Engineering,Nanchang Hangkong University,Nanchang 330100,China)

机构地区：[1]南昌航空大学信息工程学院,南昌330100

出　　处：《控制与决策》2023年第2期429-434,共6页Control and Decision

基　　金：国家自然科学基金项目(61663032);南昌航空大学研究生创新专项基金项目(YC2021-S681)。

摘　　要：针对具有量化输入饱和及输出受限的非线性非仿射系统,提出固定时间自适应神经网络跟踪控制方法.引入中值定理解决系统具有非仿射结构的问题;基于反步法,使用Barrier Lyapunov函数约束系统输出,并利用RBF神经网络逼近未知函数;根据固定时间控制理论设计输入信号,该输入信号由滞后量化器量化,以降低控制信号的通信速率,并保证该系统在满足量化输入饱和及输出受限的条件下,系统可以在固定时间内跟踪上期望信号,且该系统收敛时间与初始状态无关.最后通过Matlab仿真软件验证所设计控制器的有效性.A fixed-time adaptive neural tracking control method is firstly proposed for nonlinear pure-feedback systems with quantized input saturation and output constraints. To resolve the problem of the system with non-simulation structure,the mean value theorem is introduced firstly. The Barrier Lyapunov function is used to constrain system output, and the RBF neural network is used to approximate the unknown function based on the back stepping method. According to the fixed time control theory, we design an input signal to guarantee this system to trace desired signal at fixed time under the situation of saturation input and limited output. Besides, the input signal is quantified by a lagging quantizier, which contributes to decrease the communications rate of the control signal. and the convergence speed is unrelated to the initial state in this system. Finally, Matlab simulation software is conducted to verify the effectiveness the proposed controller.

关 键 词：固定时间控制 量化输入饱和 输出受限 Barrier Lyapunov函数 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]