A three-domain fuzzy wavelet width learning filter system and method

Abstract

The invention discloses a three-domain fuzzy wavelet width learning filter system and method. The method comprises: using the serial port software of windows to obtain the data collected by the sensor group in the master-slave mechanical arm, and extracting the time series as a training sample and a test sample through a window function ;Establish a three-domain fuzzy wavelet width learning model, and train it through training samples; send test samples to the trained three-domain fuzzy wavelet width learning model for prediction, and predict tremor signals; eliminate tremors according to the basic filtering mathematical model Signal. The present invention improves the existing breadth learning system (BLS), and adds fuzzy domain into its design, so that it can not only extract latent features in time domain and frequency domain, but also extract tremor interference in teleoperation signal in fuzzy domain. Finally, the tremor signal is eliminated through the basic filtering mathematical model, so as to achieve the purpose of eliminating the tremor interference caused by the human arm on the basis of ensuring the existing performance.

Description

technical field [0001] The invention relates to the technical field of machine learning and filtering algorithms, in particular to a three-domain fuzzy wavelet width learning filtering system and method. Background technique [0002] With the development and maturity of technology, teleoperating systems can replace humans to complete complex tasks in harsh environments. Due to its advantages of strong robustness, high precision and good reliability, teleoperating systems play an increasingly important role. However, there is arm tremor interference from the operator on the master side in the master-slave teleoperating system. This will affect the precision of the operation. [0003] In recent years, many filtering algorithms have been proposed and applied to predict and compensate for the disturbance caused by the operator's hand tremor. The relevant representative research works include: C.N.Riviere et al. proposed a weighted-frequency Fourier linear combiner (WFLC) for ...

AI Technical Summary

Problems solved by technology

[0004] However, these existing filtering algorithms only design filtering algorithms and models in the time domain and frequency domain, and cannot eliminate the tremor interference caused by the human arm.

Images