非线性飞行混沌运动的修正RBF神经网络控制

如何求神经网络控制使非线性空间飞行混沌运动不会发生。该法是在输出层用回归权代替常数权再用EM算法来估计回归权的参数 ,这样修正的RBF的神经网络控制就可使非线性空间飞行不会出现混沌现象。这种算法R Langari,L Wang&J Yen (1997) 1] 在研究径向基函数网络时提出过。其突出的优点是把复杂的多参数的最优化问题分离为N个小型最优化问题 ,这里N是隐藏层单元数。

Modified RBF Neural Networks Control for the Nonlinear Flight of Chaotic Motion

In this paper we present how to find the neural control, such that the chaotic motion of nonlinear space flight will not develop. This approach is a modified radial basis function (RBF) network in which the regression weights are used to replace the constant weights in the output layer. Apply expectation-maximization (EM) algorithm to estimate the parameters of the regression weights. This method have been proposed by R.Langari, L. Wang and J.Yen. A salient feature of this algorithm is that it decomposes a complicated multiparameter optimization problem into N separate small-scale optimization problems, where N is the number of hidden units.