Q‐Learning for Tracking Control of Autonomous Underwater Vehicle

ABSTRACT In this article, the tracking control problem of autonomous underwater vehicle (AUV) is considered. The main contribution is to give the explicitly optimal tracking controller by using a Q‐learning algorithm when the system model is unknown. The key is to discretize and linearize the kinematic equation of AUV, and then the Q‐factors are expressed as functions of input‐output vectors. In particular, a simulation study is carried out to demonstrate the effectiveness of the proposed method.