Semi-globally/globally stable adaptive NN backstepping control for uncertain MIMO systems with tracking accuracy known a priori

摘要: Abstract This paper focuses on the problem of direct adaptive neural network (NN) tracking control for a class uncertain nonlinear multi-input/multi-output (MIMO) systems by employing backstepping technique. Compared with existing results, outstanding features two proposed schemes are presented as follows. Firstly, semi-globally stable scheme is developed to guarantee that ultimate errors satisfy accuracy given priori, which cannot be carried out using all NN schemes. Secondly, we propose novel approach such closed-loop system globally stable, and in meantime also achieve known different from methods where can just ensured determined priori designers before controllers implemented. Thirdly, main technical novelty construct three new nth-order continuously differentiable switching functions multiswitching-based designed successfully. Fourthly, contrast classic schemes, this adopts Barbalat׳s lemma analyze convergence rather than Lyapunov stability theory. Consequently, adjusted accurately according real-world requirements, signals uniformly ultimately bounded. Finally, simulation example provided illustrate effectiveness merits