Modeling Based On Adaptive Frequency Observer And Tracking Control For Heart Signals

In the recent years, the morbidity of coronary heart disease increases because of human’s irregular life style and pressure. Coronary heart disease becomes a first enemy which threatens human health. Now, the most effective treatment of coronary heart disease is Coronary Artery Bypass Grafting(CABG). CABG can be operated by either “on-pump” or “off-pump”. “On-pump” CABG means that the blood’s extracorporeal circulation is realized using cardio-pulmonary monitor. In the “off-pump” CABG, stabilizer is used to avoid heart beating, blood flows in the body instead of going through the cardio-pulmonary monitor. No matter in “on-pump” or “off-pump” CABG, heart does not beat in the operation, which does a great damage to heart muscle and longers the patient’s recovery time. With the help of robot operation systems, which eliminate the relative motion between heart and robot tool, doctors who perform remote operation can see a still scene from the camera and perform operations just like the heart is not beating, which overcomes the drawbacks of traditional operation.From the beginning of this dissertation, by analyzing of heart motion signals, it can be concluded that heart motion is the coupling between heartbeat motion and respiration motion. According to the feature of heart motion signals: long bandwidth and nonlinearity, a dynamic heart model is constructed. By using adaptive frequency observer identification method, the phase angle, frequency, amplitude and future position of heart motion model can be estimated in time. Furthermore, proofs of convergence and effectiveness of this new identification method are given. Then we analyze the kinematics and get the dynamics equations by lagrange formula for the surgical robot PHANTo M. We choose output regulation as control algorithm. At last, Through the simulation on Simulink/Matlab, we compare and analyze the results of different identification methods, for example, one-step prediction algorithm, LM, combination RLS and EKF. Simulation results demonstrate that tracking error is minimal by adaptive frequency observer identification which is satisfied with the precision. It proves that heart operation robotic system based on this new identification method is useful and available.