基于模糊辨识方法的机械臂振动控制研究

Research of Manipulator Vibration Controlling Based on Fuzzy Identification

为了抑制机器人等复杂结构的振动,提高复杂结构的振动控制精度,提出一种用于辨识机械臂连接结构的非线性模型在线辨识的模糊算法,并以此为基础研究了机械臂振动控制方法。该算法将非线性输入输出系统用时变线性系统模型来拟和,并把此模型表示成模糊模型的形式,用在线调节模糊模型的方法来辨识时变线性模型的相关参数。将递推模糊聚类方法与卡尔曼滤波法用于在线调整模糊模型参数,将此算法应用在两自由度柔性杆件的扭转振动的控制上,并设计相应的硬件控制系统,实验结果表明了此算法的有效性。将该算法应用于工程实践中,实际使用效果表明,此算法具有重要的工程应用价值。

For controlling complex structure vibration and advancing precision of controlling complex structure, an online fuzzy identification method for joint structure of manipulator is presented. By the method, nonlinear system is substituted by time-varying linear system, and the multi-input and single output model is expressed by fuzzy model. An online identification algorithm based on recursive fuzzy clustering method is presented. The result of experiment demonstrate that the method is effective. The method is used to practice engineer and the effect is good.