论文部分内容阅读
针对多连杆柔性关节机械臂,设计了基于高斯径向基函数神经网络(GRBFNN)的滑模控制器,该控制器利用神经网络的逼近能力,将各关节的切换函数作为网络的输入,控制器完全由连续的RBF神经网络实现。利用该控制器与线性二次型跟踪器以及传统滑模控制器对三连杆柔性关节机械臂进行轨迹跟踪控制仿真。仿真结果表明:线性二次型跟踪器具有一定传输时延,滑模控制器跟踪轨迹有明显抖振,而神经滑模控制器取消了切换项,消减了抖振,具有良好的跟踪效果和稳定性。
For multi-link flexible joint manipulator, a sliding mode controller based on Gaussian radial basis function neural network (GRBFNN) is designed. By using the approximation ability of neural network, the controller takes the switching function of each joint as the input and control of the network Completely by the continuous RBF neural network to achieve. The controller and linear quadratic trackers and the traditional sliding mode controller are used to track and control the three-link flexible joint manipulator. The simulation results show that the linear quadratic tracker has a certain transmission delay, the tracking trajectory of the sliding mode controller has obvious buffeting, and the neural sliding mode controller cancels the switching term, reduces the buffeting, has a good tracking effect and stable Sex.