针对存在参数不确定性和外界随机扰动的欠驱动桥门式起重机防摇摆跟踪控制问题，提出一种基于径向基函数(RBF)神经网络的直接鲁棒自适应控制器。首先利用第二类拉格朗日方程建立双摆桥门式起重机系统动力学微分方程，并针对无扰动工况设计理想控制器；随后引入RBF神经网络拟合理想控制输出，在考虑扰动项影响的情况下设计神经网络权值更新律，并证明系统的Lyapunov稳定性；最后进行仿真验证。文中所提控制器仅需要小车实时位置与速度信息作为控制输入，无需系统参数信息，控制精度取决于RBF神经网络的拟合精度。仿真结果表明，文中提出的控制器具有良好的轨迹跟踪性能与防摆效果，且鲁棒性较强。

A directly robust adaptive controller based on the radial basis function (RBF) neural network was proposed to solve the problem of the trajectory tracking for anti-sway control of an underacuated overhead and gantry crane with the parameter uncertainties and external disturbances. Firstly, the dynamic model of a double-pendulum crane was established by Lagrange equation of the second kind, and an ideal controller was designed for undisturbed conditions. Secondly, the RBF neural network was introduced to fit the ideal control output and its weight update rate was designed considering the effects of disturbances, after that the Lyapunov stability of the system was proved. Finally, a simulation test was carried out to verify the controller performance. The controller only takes the real-time position and speed information of the trolley as the control input, and no system parameter information is needed. The control accuracy depends on the fitting accuracy of the RBF neural network. The simulation results show that the controller proposed in this paper has excellent trajectory tracking and anti-sway performance and strong robustness.