Research on the Recycle System for the Unmanned Surface Vehicle Based on the NAR Neural Network

Aiming at the problem that the recycle works all rely on manual operations for the unmanned surfaces vehicle, an automatic recycle system is developed for the unmanned surfaces vehicle. A cable-throwing mechanism is designed for establishing the physical connection, and a guiding/docking mechanism with a compatible hammer is developed for lifting and releasing the USV through the ship crane. The kinetic model is established under the multi-disturbances of the cable-throwing ball and the motion trajectory is calculated in real time by the LabVIEW. In addition, the nonlinear auto regressive neural network is used to predict the USV attitude in the recycle process, which provides the optimal recycle information for operators to launch the cable-throwing ball. The system guarantees a timesaving recycle time (≥200s) and a higher recycle success rate (≥95%). Sea experiment demonstrates the safety and efficiency of the proposed system.