ADAPTIVE NEURAL NETWORK UNKNOWN TRAJECTORY TRACKING CONTROL FOR MARINE SURFACE VESSEL WITH CONSTRAINTS

Most marine surface vessel (MSV) trajectory tracking problems ignore the output, state and input saturation constraints in the operation of the actual system. In addition, the desired/target trajectory for tracking is based on known conditions. To address these problems, this research introduced an adaptive tracking control scheme under the MSV system with constraints. The overall controller design is based on the backstepping control technology. Firstly, the parametric approach is adopted to estimate the desired trajectory. Secondly, an integral Barrier Lyapunov Function (iBLF) is used to directly handle system state constraints. Finally, using the meanvalue theorem to deal with input saturation. In addition, the adaptive ability of fully tuned radial basis function neural network (FTRBFNN) is used to make system better compensate for uncertainties, and further improve the adaptive ability of the backstepping approach to uncertainties. With the proposed approach, the constraints will never be violated during the entire system, and the MSV system state is bounded. At the end of the research simulations proves the effectiveness of the proposed approach.