【摘要】 为了提高水面无人艇在外界干扰和自身运动状态变化等非线性及不确定因素下的航向控制精度，针对常规自抗扰控制（ADRC）在船舶航向控制中存在的参数整定难这一问题，提出将粒子群优化算法（PSO）与径向基函数神经网络（RBF）相结合的ADRC参数自适应整定和无人艇航向精确控制策略。设计一个RBF网络用于准实时无人艇运动模型建模。在此基础上，运用粒子群算法对无人艇航向ADRC控制器参数整定进行优化。仿真试验结果表明：设计的RBF-PSO-ADRC控制器在无人艇的航向控制中相比常规ADRC控制器和比例-积分-微分控制器（PID）控制器表现出更好的动态响应特性和稳定性，对未知扰动也具有较强的鲁棒性和抗干扰能力。RBF-PSO-ADRC控制器实现了复杂环境下ADRC控制器参数的自适应优化整定，这对于高精度的无人艇航向控制和航线跟踪具有一定实际意义，可为水面无人艇操纵设计提供有益参考。更多还原

【Abstract】 Aiming to enhance the precision of heading control for unmanned surface vessels(USV) under nonlinear and uncertain factors such as external disturbances and variations in the vessel’s own motion state,addressing the challenge of difficult parameter tuning in conventional active-disturbance-rejection-control(ADRC) for heading control in maritime vessels, a strategy is proposed that combines Particle swarm optimization(PSO) algorithm with Radial Basis Function(RBF) neural network for adaptive tuning of ADRC parameters and precise heading control of unmanned surface vessels. Firstly, an RBF network is designed for quasi-real-time modeling of the unmanned surface vessel motion model. Based on this, the PSO algorithm is applied to optimize the parameters of the heading ADRC controller for unmanned surface vessels. Simulation experiments demonstrate that the designed RBF-PSO-ADRC controller exhibits better dynamic response characteristics and stability in the heading control of unmanned surface vessels compared with that of conventional ADRC controllers and proportional-integral-derivative(PID) controllers. It shows robustness and disturbance rejection capabilities in the presence of unknown disturbances. The RBF-PSO-ADRC controller achieves adaptive optimization and tuning of ADRC controller parameters in complex environments,contributing to high-precision heading control and trajectory tracking of unmanned surface vessels. This provides practical significance and useful references for the maneuvering design of unmanned surface vessels.更多还原