Research On The Mechanism Of Tremor And Its Inhibition Method In Minimally Invasive Surgery Robots

Minimally invasive surgical robot is the perfect combination of high technology and advanced medicine,which has the advantages of flexible degree of freedom,scaling motion and 3D vision etc.However,the tremor phenomenon,occurring at the tip of the surgical instrument during the operation,seriously affects the quality and safety of the surgery as well as increases the difficulty of-performing the fine motions such as needles and suture.In view of the above-mentioned problems,the tremor mechanisms were studied and the corresponding tremor suppression methods have been put foward,which have been validated by using the master-slave minimally invasive surgical robot.The main contents of this thesis are as follows:Firstly,the tremor mechanisms were researched from the perspectives of non-smooth operational instruction of surgeons and robot dynamic behaviors etc.The key factors leading to the tremor phenomenon were concluded by simulations and experiments.Secondly,a new zero phase adaptive fuzzy Kalman filter(ZPAFKF)was proposed to suppress physiological tremor of surgeons.The desired instructions no containing physiological tremor signal were obtained by adding the tremor signal with the same amplitude but opposite phase to the original command signals in real time.This method can effectively avoid the disadvantage that traditional low pass filter may cause the phase delay and information loss.Then,a fuzzy neural network sliding mode controller(FNNSMC)was proposed to alleviate vibration caused by the dynamic behaviors of minimally invasive surgical robot especially the "stick-slip" phenomenon.The smooth trajectory points were obtained by the sliding mode control as well as compensating the off-line identified frictional torque,irrelevant disturbances estimated by neural network and the error of estimation.This algorithm overcomes the shortcomings that conventional PD controller may contribute to the "crawling" and "flatting" phenomenon in a low speed state and the traditional sliding mode controller may cause the "chattering"phenomenon when control law sharply switches.Meanwhile,the stability of the FNNSMC has been proved by Lyapunov function.Finally,the validities of the proposed suppression methods were verified by the simulations and experiments.The results showed that they could effectively suppress the tremors of minimally invasive surgical robot and improve the motion accuracy of the end effector.