Least Squares Support Vector Machine Kalman Filter for Physiological Tremor Suppression in Minimally Invasive Surgical Robot

Abstract

This paper studies the physiological tremor filtering in minimally invasive surgical robot. The surgeon's physiological tremor of the hand can cause the vibration of the tip of the surgical instrument, which may reduce operative accuracy and limit the application of surgical robots. Aiming at the vibration caused by physiological tremor of hand, we propose a Least Squares Support Vector Machine Kalman Filter (LSSVMKF), which can filter the tremor by estimating and modeling the tremor signal by Kalman filter and then superimposing it reversely in the control signal. When estimating and modeling the tremor signal, the filter uses the Least Squares Support Vector Machine (LS-SVM) to build the regression model of the constant parameters (Process Noise Covariance and Measurement Noise Covariance) of the traditional Kalman filter, which can dynamically adjust these parameters during the operation and improve the accuracy of Kalman filter. The simulation results show that the LSSVMKF can effectively filter out the tremor signal, thereby improving the accuracy of surgery.