A physically motivated control algorithm for an autonomous underwater vehicle

An Autonomous Underwater Vehicle (AUV) is an underactuated mobile robotic system. This paper is focused on the control of an AUV, which is an interesting problem in robotics. In engineering applications, achieving a precise mathematical model for a given system is not realistic due to the existence of many unknown and unpredictable phenomena in real systems. Therefore, less reliant algorithms on the system mathematical equations are certainly preferable. To this end, first, the mathematical model of the AUV is presented. Then, the desired signals for the control algorithm are produced. After that, a nonlinear PID-based kinematic control is proposed to solve the trajectory tracking problem for the AUV. Subsequently, a Lyapunov-based dynamic control is designed for system actuator torques. In contrast to many existing solutions, control formulation does not require any model transformation or approximation and it is formulated in the original configuration space of the system. The proposed control law is simple, with an explainable mechanism and straightforward tuning, and it leads to the non-oscillatory robot motions in the Cartesian space.