A diagonal recurrent neural network-based hybrid direct adaptive SPSA control system

A direct adaptive simultaneous perturbation stochastic approximation (DA SP SA) control system with a diagonal recurrent neural network (DRNN) controll er is proposed. The DA SPSA control system with DRNN has simpler architectu re and parameter vector size that is smaller than a feedforward neural netw ork (FNN) controller. The simulation results show that it has a faster conv ergence rate than FNN controller. It results in a steady-state error and is sensitive to SPSA coefficients and termination condition. For trajectory c ontrol purpose, a hybrid control system scheme with a conventional PID cont roller is proposed.