船舶航向非线性系统的模糊神经网络智能控制器设计

船舶航向控制系统具有典型的非线性和不确定性特性，并受自动舵执行能力的约束，这使得作为船舶智能化基础的航向控制极具挑战性。首先分析了船舶航向运动特性，给出带有舵约束的航向运动非线性数学模型；然后以模糊神经网络为控制器结构，在噪声加入和参考轨迹设置算法的支持下使用遗传算法对控制器参数进行自动搜索和优化，设计一种船舶航向智能控制器；最后对航向控制进行仿真。结果表明，所设计的航向智能控制器对船舶参数摄动和扰动具有良好的鲁棒性能。

The Design of Ship Course Intelligent Controller Based on FNN of Non - Linear System

Becanse of the strong non - linearity and uncertainty, and as well as the dynamics restraints of autopilots, designing a high performance ship course controller is yet a challenge. Firstly, the ship course motion characteristics are analyzed and the nonlinear model with autopilot dynamic restrictions being given. Then a kind of intelligent course controller based on fuzzy - neural network is designed. Under the sustainments of the noise - adding and the reference trajectory setting algorithms, the controllers parameters are automatically searched and optimized by genetic algorithm. At last, the simulations of ship course control are done. The simulation Jesuits show that the intelligent controller has good robustness to the ship parameter perturbations and the outside disturbances.