Research on Nonlinear Decoupling Method of Piezoelectric Six-Dimensional Force Sensor Based on BP Neural Network

3rd International Conference on Automation, Control and Robotics Engineering

工业技术;计算机科学;无线电电子学

Li, Yingjun^1 ; Wang, Guicong^1 ; Han, Binbin^1 ; Yang, Xue^1 ; Feng, Zhiquan^2

School of Mechanical Engineering, University of Jinan, Jinan

250022, China^1

School of Information Science and Engineering, University of Jinan, Jinan

250022, China^2

关键词: BP neural networks;    Calibration experiments;    Decoupling methods;    Input and outputs;    Measurement accuracy;    Nonlinear characteristics;    Nonlinear decoupling;    Six-dimensional force;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/428/1/012041/pdf
DOI  :  10.1088/1757-899X/428/1/012041

来源: IOP

【 摘 要 】

The six-dimensional force sensor has become one of the major bottlenecks restricting the development of robots in China. In this paper, the problem of the decoupling of the piezoelectric six-dimensional force sensor with four-point support structure is studied, and the static decoupling method is studied. Firstly, the principle of nonlinear decoupling algorithm for six-dimensional force sensor is analyzed, and experimental data obtained by decoupling are acquired through calibration experiments, and sample selection and normalization processing are performed. After that, the BP forward feedback neural network was used to optimize the multi-dimensional nonlinear characteristics of the sensor output system, and the input and output mapping relationship of the sensor was determined, and the decoupled sensor output data was obtained. The determinant sensor's measurement accuracy evaluation index is compared with linearity error and coupling rate error.