基于多模型结构的非线性鲁棒自适应跟踪控制

针对复杂的非线性系统,提出一种基于多模型结构的鲁棒自适应控制方法,使得系统可以在不同的运行环境下跟踪给定的信号.由多个线性模型和一个模糊模型及其相应的控制器构成基本的多模型控制系统,再引入动态结构自适应神经网络以保证系统的稳定性及抑制由频繁切换引起的噪声.最后,对某小型飞机进行全包络飞行跟踪控制的仿真,验证所提控制方法是有效的.

Robust adaptive control of complex nonlinear systems using multiple models

For complex nonlinear systems,a kind of robust adaptive control method using multiple models is presented to track the given signal under different working conditions.The basic multiple-model control system is formed by several linear models and one fuzzy model with their corresponding controllers.And the dynamic structure adaptive neural network is introduced to stable the whole system and restrains the disturbance influence caused by frequent switching.The simulation results show the presented control method is effective by demonstrating the full envelope tracking control of a puddle-jumper.