A novel neural network-based nonlinear controller for shipboard rotary cranes against random wave interference

The three-dimensional (3D) shipboard rotary crane is a complex, underactuated mechanical and electrical piece of equipment with highly nonlinear and strongly coupled characteristics. However, it suffers from random, unknown interferences induced by ocean waves and currents. Additionally, unidentified system parameters can influence the control process of transporting loads. Therefore, this paper proposes a neural network adaptive control method that compensates for the system under complex interference. Additionally, the developed scheme eliminates the influence of unknown parameters through an adaptive law to achieve the control purpose. Next, the mathematical proof of local asymptotic stability for the shipboard crane system in closed-loop configuration is demonstrated by employing the Lyapunov theorem and LaSalle's invariance principle. Finally, the simulation results show that the peak of the load swing angle is kept below 2.6^∘, and the positioning accuracy error of the cantilever is within 0.16^∘, verifying the effectiveness of the method.