Controller Tuning Using Artificial Immune Systems Applied to Inverted Pendulum

In this paper, we describe the application of a bioinspired algorithm based on Artificial Immune Systems (AIS) for an angular position control system tuning of an inverted pendulum. Two controllers are detailed and used in two different cases; for Case 1, a PID controller has been designed using a simulated model of the inverted rotary pendulum in MATLAB and an AIS program for tuning the controller gains. Comparisons of the performance of controllers during their transitive process are made to measure the extent to which they improve the maximum overshoot, settling time, and steady-state error of the base controllers and the controller tuned with the AIS algorithm. Finally, for Case 2, the tuning performance of the AIS is verified on an Inverted Rotary Pendulum (IRP) physical plant with a State Variable Feedback (SVF) controller. The results demonstrate that using bioinspired techniques such as AIS for tuning a previously designed controller improves the performance of the non-linear system, compared to the results before the tuning process.