Optimal neuro control of robot manipulator

Recently, radial basis function network (RBFN) is used quite widely when using neural networks as controllers for subjects with multiple uncertain parameters such as the robot. The most important thing when using online learning neural network system is the choice of coefficient for networks with fast convergence speed. So far this coefficient has been chosen by experience and sometimes it takes quite a long time to find a coefficient that satisfies the requirement of the controlling task. Another problem is, when finding coefficients satisfying the required study of the problem and control, we can not conclude that the optimal coefficients. This article refers to the use of genetic algorithms (GA) to find optimal learning coefficient for RBF network is used as a controller for objects whose parameters are uncertain.