On line tuning of a neural PID controller based on variable structure RBF network

This paper presents the use of a variable structure radial basis function (RBF) network for identification in PID control scheme. The parameters of PID control are online tuned by a sequential learning RBF network, whose hidden units and connecting parameters are adapted online. The RBFnetworkbased PID controller simplifies modeling procedure by learning inputoutput samples while keep the advantages of traditional PID controller simultaneously. Simulation results of ship course control simulation demonstrate the applicability and effectiveness of the intelligent PID control strategy