An online adaptive logic-oriented neural approach for tracking control

This study presents an adaptive logic-oriented neural control system for Unmanned Surface Vehicle (USV). An autonomous fuzzy neural controller is constructed by using rudder to regulate both the tracking error and heading error. The proposed approach based on logic-oriented neural network with domain knowledge easily incorporated in advance can be easily interpreted and it can be directly translated into a series of logic expressions formed over a collection of information granules. The proposed neural controller can adapt itself on-line from a direct evaluation of the control accuracy, so there is no need for a supervised training process. The control system is implemented in a model of USV. Simulations for track-keeping control under environmental wind and waves have been investigated. Simulation results presented in the paper demonstrate the effectiveness of the proposed scheme.

► A modeling method for unmanned surface vehicle (USV) is introduced. ► An adaptive logic-oriented neural control system for USV is presented. ► Simulation experiments were conducted to demonstrate that the proposed scheme is effective.