Ant colony fuzzy neural network controller for cruising vessel on river

• We realize the ACFNN controller for the cruising vessel on an uncertain river.
• To achieve the stability, the vessel velocity is online updated by an FNN controller.
• The parameters of FNN controller are updated online via adaptive and estimator law.
• The time-varying path of AC algorithm is designed for the uncertain river.
• A searching pattern of AC algorithm can be rebuilt under different conditions.

This paper proposes an ant colony fuzzy neural network (ACFNN) controller for a cruising vessel on a river. The proposed controller comprises an ant colony (AC) algorithm, a fuzzy neural network (FNN) controller, and a switching law. The approximately optimal sailing line and short sailing time are obtained using the AC algorithm. First, the searching pattern of the AC algorithm is constructed using the data of the tidal current, river current, vessel velocity, and position of the coordinate. From a tracking error viewpoint, the switching law determines that the approximately optimal sailing line and the shorter sailing time are obtained using the AC algorithm, and that uncertain nonlinear factors are compensated by the FNN controller. The controller consists of an FNN identifier and a robust controller. The identifier is used to estimate the vessel velocity, and its parameters are tuned online by the adaptive law derived from the Lyapunov function. The robust controller is used to compensate for uncertainties of the tidal current and river current through the estimated law. The output of the ACFNN controller is the sum of the FNN identifier, the robust controller, and an auxiliary function. Finally, a simulation and a practical cruising vessel on a river are performed to verify the effectiveness of the presented controller.