模型参考的广义最小方差鲁棒自校正控制器及其应用

本文提出一种具有参考模型的广义最小方差鲁棒自校正控制算法。采用鲁棒的估计器和改进加权因子的选择方法改善自校正控制的鲁棒性，通过选择参考模型确定系统的希望响应特性。对具有大间隙、摩擦、饱和、弹性、死区、滞后等固有本质非线性因素的大功率伺服系统进行实验研究，结果表明，由该算法构成的控制系统具有良好的动态响应、良好的鲁棒性和高的控制精度，可以广泛应用地炮、火箭炮等要求高精度快速定位的系统中。

Generalized Least Square Error Robust Self-tuning Controller with Model Reference and Its Application

A kind of generalized least square error robust self tuning controller algorithm with model reference is proposed.By using robust estimator and correcting select method of weighting factors the robustness of self tuning control is improved.By selecting reference model,desired response property of system is established.Experiments show that control systems formed by the algorithm have better dynamic response,robustness and accuracy,for large power DC servo mechanism which has inherent essential nonlinear factors,such as large backlash,elasticity,saturation,friction,dead,delay,and so on.It may be extensively applied to systems that require better accuracy and fast location for land gun and rocket gun,and so on .