A decoupling method for multi-dimensional force sensor in noisy environment

Abstract

A decoupling method for multi-dimensional force sensors in noisy environments. The method includes the following steps: step 1, obtaining the calibration data corresponding to the sensor output voltage U and the standard weight (force) F through the signal conditioning circuit and the acquisition system; step 2, adding noise to the calibration data obtained in step 1, Among them, the signal-to-noise ratio is controlled between 20~30dB; step 3, the multiple groups of noise samples generated in step 2 are spliced ​​sequentially, and normalized at the same time, forming a new data sample; step 4, the obtained in step 3 Input the sample into the deep neural network (DNN) for training, and judge whether the condition of model convergence is met. If the condition is met, go to step 5, otherwise continue to step 4; step 5, the model training is over, and the training is obtained The model embedded in the sensor acquisition system is finally applied to the actual industrial site. The invention improves the robustness of the sensor in the noise environment and has good practical application value.

Description

technical field [0001] The invention relates to the field of multi-dimensional force sensor data acquisition, in particular to a decoupling method of a multi-dimensional force sensor in a noise environment. Background technique [0002] Multi-dimensional force sensors are widely used in the research of robotic fingers and claws; robotic surgery research; finger force research; dental research; force feedback; brake detection; precision assembly, cutting; restoration research; plastic surgery research; product testing; Teach to learn. The industry covers robotics, automobile manufacturing, automated assembly line assembly, biomechanics, aerospace, textile industry and other fields. The coupling error of the multi-dimensional force sensor affects its detection accuracy, and also limits its application in the field of high-precision measurement and control. [0003] At present, optimizing the sensor structure and enhancing the decoupling algorithm are two effective ways to im...

AI Technical Summary

Problems solved by technology

[0004] Domestic patents related to multi-dimensional force sensor decoupling methods include "Genetic algorithm-based multi-dimensional force sensor calibration experiment data fitting method" (201610232792.4), by deriving the coefficient solution formula of the sensor coupling error theoretical model, and then using the genetic algorithm on MATLAB software Determine the global optimal solution to solve the data fitting problem, but the optimal solution obtained by the genetic algorithm in this patent may be a local optimal rather than a global optimal, resulting in an increase in the fitting error

National Invention Patent "A Dynamic Decoupling Method for Multi-dimensional Force Sensors" (201910160583.7), this method first conducts a dynamic test on the sensor, then dynamically compensates the output signal, and finally brings the compensated signal into the decoupling model to realize decoupling, but this method does not take into account the impact of noisy noise in the actual industrial environment on the data collected by the sensor, and there may be certain limitations in practical applications

Images