The design of Tracking Trajectory System for an Underaccuated Unmanned Surface Vehicle by Using an Adaptive Neuro-Fuzzy Sliding Mode Controller

This work introduces an adaptive neuro-fuzzy sliding mode controller strategy for the trajectory tracking of an underactuated unmanned surface vehicle (USV). The proposed scheme takes into account uncertainties and unknown disturbances. This research elaborates on a cooperative control strategy that combines NN, FLC, and SMC, or sliding mode control. To determine the equivalent control component, a feed-forward neural network is employed. To approximate the corrective control, a continuous fuzzy logic controller is utilized. Initially, the PID sliding surface is formulated, and afterwards, the coefficients associated with it are determined through the utilization of an adaptive control rule. To mitigate the occurrence of excessive chatter, a fuzzy logic controller (FLC) is employed to approximate the corrective control component. The same control is also computed using a feed-forward neural network (NN). Using the adaptive PID sliding surface and the gradient descent method, the weights of the neural network are computed. This particular technique has the capability to attain asymptotic stability and exhibit accelerated convergence.