Predictor-based practical fixed-time adaptive sliding mode formation control of a time-varying delayed uncertain fully-actuated surface vessel using RBFNN

This paper focuses on fixed-time formation control (FTFC) of a fully-actuated surface vessel (FASV) considering complex unknowns, including fully unknown dynamics and disturbances, input saturation and time-varying delays. First, using prediction idea to address time delay, a novel state predictor (SP) strategy combining with state transformation (ST) technique is devised for each FASV to predict the evolution of system states such that fixed-time stability can be ensured while solving the delay problem. Besides, the uncertainties in the transformed system are attentively considered. In addition, aiming to distinctly identify complex unknowns, predictor-based neural network is injected into the foregoing delay processing method. Finally, using time base generator (TBG), a new adaptive terminal sliding mode (ATSM) is incorporated into FTFC strategy which in turn contributes to decreasing control inputs and acquiring smooth convergence process. Simulation results and comparisons are thoroughly provided to testify the effectiveness and superiority of the designed FTFC scheme.