基于RBFNN的桥式起重机AHSMC控制策略

Adaptive Hierarchical Sliding Mode Control Strategy for Overhead Cranes Based on Radial Basis Function Neural Network

针对欠驱动三维桥式起重机控制策略难以在微控制器资源受限的条件下进行验证,以及传统分层滑模控制策略难以在不确定性条件下进行准确控制的问题,提出了基于径向基神经网络的三维桥式起重机自适应分层滑模控制策略。首先,基于分层滑模控制方法构建起重机全驱动和欠驱动子系统的一阶滑模面;然后,将一阶滑模面进行线性组合,形成第二阶滑模面;进一步,利用径向基函数神经网络对控制参数进行自适应估计并更新滑模面,提高不确定性条件下控制策略的鲁棒性。最后,通过仿真分析和物理实验验证了所提自适应分层滑模控制策略的有效性。

The control strategy of underactuated three-dimensional overhead crane is difficult to verify under the condition of limited microcontroller resources, and the traditional hierarchical sliding mode control strategy is difficult to implement accurately under the condition of uncertainty. To tackle the above problems, an adaptive hierarchical sliding mode control strategy for three-dimensional overhead crane based on radial basis function neural network is proposed. Firstly, the first-order sliding surface of the actuated and underactuated subsystems of the three-dimensional crane is constructed based on the hierarchical sliding mode control method.Then, the first-order sliding surface is linearly combined to form the second-order sliding surface. Furthermore,the radial basis function neural network is used to adaptively estimate the control parameters and update the sliding surface to improve the robustness of the control strategy under the condition of uncertainty. Finally, the effectiveness of the proposed adaptive hierarchical sliding mode control strategy is verified by simulation analysis and physical experiments.