The control of higher order linear system is one of the main fields of research area that has been studied for decades because of the difficulty in designing a controller for such systems. One of the best approaches to solve this problem is by reducing the order of the system into a second orders, based on this reduction many approaches can be proposed for controlling the higher order system, therefore many reduction methods are suggested and developed for this purpose, one of these methods is the Mixed Reduction Method (MRM). The first contribution of this paper is to improve the efficiency of MRM by using a flower optimization algorithm.

The second contribution of this paper lies in proposing a hybrid Neuro-Robust deadbeat controller using Matlab facilities to control higher order linear systems based on the optimized MRM. Where the robust deadbeat control algorithm is combined with a modified adaptive radial basis neural network to improve the robustness and efficacy of the deadbeat controller, which is partially lost when designing this controller for the higher order based on model reduction. The suggested radial basis function neural network has a simple design. The proposed control scheme assures the stability of the overall closed loop-controlled system; therefore, it can be applied to control any linear higher order systems. Results of different simulation examples show the efficiency of the proposed hybrid controller (Neuro-robust deadbeat) in tracking different reference signals compared to the robust deadbeat controller.