针对一类含有完全未知关联项的多输入/多输出非线性系统,提出了输出反馈动态面自适应控制方案,克服了反推控制中的微分爆炸问题;利用神经网络逼近系统中的未知关联项,对于每个子系统只需对一个参数设计自适应律;引入性能函数和输出误差变换,跟踪误差信号的收敛速率、最大超调量和稳态误差等控制性能指标均可得到保证.理论证明了闭环系统的所有信号半全局一致有界,仿真结果验证了所提方案的有效性.
An output feedback adaptive dynamic surface control(DSC) scheme is proposed for a class of MIMO nonlinear systems with completely unknown interconnections. In this scheme, the explosion of complexity problem inherentin traditional backstepping design is eliminated. The radial-basis-function(RBF) neural network(NN) is employed to approximate the uncertain interconnected items. The advantage is that there is only one parameter needed to be updatedonline for each subsystem. Moreover, performance function and output error transformation are introduced to guaranteethe convergence rate of the tracking errors, the allowable maximum overshoot, and the steady-state error, etc. It is provedthat all signals in the closed-loop system are semi-globally uniformly ultimately bounded. Simulation results show theeffectiveness of the proposed scheme.