No abstract

No abstract

At present, the track control of unmanned surface vessel (USV) is mainly divided into direct track control and indirect track control, both of which have their advantages in the aspect of USV track control. Though having high control precision, the former still fails to meet the related requirements. To tackle this problem, the nonlinear model predictive control (NMPC) algorithm was optimized based on the convolutional neural network (CNN) in order to design a direct USV track controller. In consideration of the nonlinearity and time lag problems that are likely to occur in USV track control, the convolutional neural network was used to solve the optimal control sequence of the predictive control algorithm. The simulation test indicates that this algorithm has effectively improved the accuracy and real-time performance of track control.

No abstract

No abstract