Comparison of a PID controller versus a LQG controller for an autonomous underwater vehicle

Currently autonomous underwater vehicles (AUVs) are facing an increase in their applications in the offshore industry, allowing maximum utilization of the maritime resources. Therefore, the development of autonomous control systems is imperious, for fast decision making in unknown situations or environments. In order to achieve this, is necessary to solve the control and navigation problem efficiently. In this article two control techniques, previously used in AUVs, are compared, in order to apply one of them in the submarine ICTIOBOT, a development platform designed by the research group INTELYMEC of the Faculty of Engineering of the UNICEN. Firstly two controllers are designed, a PID with a Kalman filter for state estimation, and a LQG controller. An evaluation of the performance is done in simulation, with regards to different operation conditions. The results show that the LQG strategy has a better performance that the PID although the latter is computationally more tractable.