Adptive Heading Control of Underactuated Unmanned Surface Vehicle Based on Improved Backpropagation Neural Network

Abstract Aiming at the challenges to accurate and stable heading control of underactuated unmanned surface vehicles arising from nonlinear interference caused by overlay interaction multi interference, also uncertainties model parameters, a algorithm for an vehicle based on improved backpropagation neural network is proposed. Based applying optimization theory realize that tracks desired yaw angle maintains it, momentum weight combined with tracking differentiator improve robustness system dynamic property control. A hyperbolic tangent function used establish mappings approximate method adopted summarize general mathematical expressions, gradient descent applied ensure convergence. The simulation results show proposed has advantages strong robustness, anti-interference high accuracy. Compared two commonly algorithms, accuracy in complex environment more than 50%.