基于过程神经网络的六维力传感器动态解耦研究  

Research on dynamic decoupling of six-axis force sensor base on the procedure neural networks

作　　者：王毓[1] 许德章[1] 张家敏[1] 许曙[1] 

机构地区：[1]安徽工程大学机械与汽车工程学院,安徽芜湖241000

出　　处：《重庆文理学院学报（社会科学版）》2016年第5期34-40,共7页Journal of Chongqing University of Arts and Sciences（Social Sciences Edition)

基　　金：国家自然科学基金资助项目(51175001)

摘　　要：六维力传感器动态解耦方法主要集中于不变性动态解耦方法和迭代解耦方法,其解耦效果取决于建模精度,强耦合情况下解耦误差大,工程实现复杂.文章针对这一问题提出一种基于过程神经网络动态解耦算法,将六维力传感器输入输出及网络权函数进行相同正交基展开,简化计算过程,求解六维力传感器输出输入耦合关系.实验结果表明,过程神经网络在六维力传感器动态解耦中应用效果良好,为动态解耦提供了一种新方法.The main dynamic decoupling method of six - axis force sensor focused on Constance dynamic de-coupling method and the iterative dynamic decoupling method. In the case of the strong coupling, decou-pling will produce large errors and the project is complex. Aiming at the problem, this paper offers a dynam-ic decoupling algorithm for this problem based on process neural network, which expand the input and output functions of the six - axis force sensor and the connection weight functions of the network based on orthogonal function basis expanses and simplify calculation process. Also it is to solve the relationship of the six - axis force sensor and the input, output of coupling. Experimental results show that the process of application of neural network has good effect on the six - axis force sensor dynamic decoupling and provides a new method for the dynamic decoupling.

关 键 词：六维力传感器 动态解耦 过程神经网络 正交基展开 

分 类 号：TP212.12[自动化与计算机技术—检测技术与自动化装置]