Li Yongming, Li Kewen, Tong Shaocheng
IEEE Trans Neural Netw Learn Syst. 2020 Jul;31(7):2532-2543. doi: 10.1109/TNNLS.2019.2933409. Epub 2019 Aug 30.
This article investigates the adaptive neural network (NN) finite-time output tracking control problem for a class of multi-input and multi-output (MIMO) uncertain nonlinear systems whose powers are positive odd rational numbers. Such designs adopt NNs to approximate unknown continuous system functions, and a controller is constructed by combining backstepping design and adding a power integrator technique. By constructing new iterative Lyapunov functions and using finite-time stability theory, the closed-loop stability has been achieved, which further verifies that the entire system possesses semiglobal practical finite-time stability (SGPFS), and the tracking errors converge to a small neighborhood of the origin within finite time. Finally, a simulation example is given to elaborate the effectiveness and superiority of the developed.
本文研究了一类幂次为正奇有理数的多输入多输出（MIMO）不确定非线性系统的自适应神经网络（NN）有限时间输出跟踪控制问题。此类设计采用神经网络来逼近未知的连续系统函数，并通过结合反步设计和添加幂积分器技术构造了一个控制器。通过构造新的迭代李雅普诺夫函数并利用有限时间稳定性理论，实现了闭环稳定性，进一步验证了整个系统具有半全局实际有限时间稳定性（SGPFS），并且跟踪误差在有限时间内收敛到原点的一个小邻域内。最后，给出了一个仿真例子来说明所提出方法的有效性和优越性。
