Recurrent neural network modeling and control of flexible plate material

Proposes a new trajectory control method for flexible plates with unknown parameters. At first the authors make a recurrent neural network (RNN) learn the dynamics model of the flexible plate handled by a robotic manipulator. When obtaining the dynamics model, the number of input units of the NN must be considered. Flexible plates have infinite modes so a NN having finite units can not obtain the dynamics perfectly. The authors make clear a relation between the number of input units and the number of modeled modes with simulation results. Next, the authors obtain the feedfoward control input based on the RNN model using the proposed learning control method. The authors applied this method to the trajectory control of a flexible plate material. Simulation examples are given to show the method's effectiveness