Abstract
This paper describes a new method of an LQI design using an unsupervised neural network. The LQI controllers are being widely used for industrial applications because of their excellent performance and robustness. However, we cannot expect the desired response for nonlinear system such as pneumatic cylinders. In this paper, we have come up with an approach to auto-tune the control gains using an unsupervised neural network which has flexible learning ability for nonlinear systems. The simulation and experimental results conducted show that the proposed method has better performance than conventional LQI controllers.
