Model simplification for AUV pitch-axis control design

Although the use of low-order equivalent models is common and extensively studied for control of aircraft systems, similar analysis has not been performed for submersible systems. Toward an improved understanding of the utility of low-order equivalent models for submersible systems, we examine control design for pitch-axis motion of an autonomous underwater vehicle (AUV). Derived from first principles, the pitch-axis motion of a streamlined AUV is described by third-order dynamics. However, second-order approximate models are common for system identification and control design. In this work, we provide theoretical justification for both the use of and limitations of a second-order model, and we verify our results in practice via a series of case studies. We conclude that a second-order pitch-axis model should often be sufficient for system identification and control design.