Robust Control Design of a Human Heart Rate System for Cardiac Rehabilitation Exercise

Abstract

Automatic, precise, and accurate heart rate control during treadmill exercise is an interesting topic among researchers. The human heart is a highly nonlinear system. Conventional control techniques are not sufficient and it is difficult to accurately model the human heart. Two different robust controllers were designed for this nonlinear system. Firstly, sliding mode control (SMC) was implemented; SMC is robust against parametric uncertainties and external disturbance but its robustness is not guaranteed during the reaching phase, especially in heart rate control, and implementation of SMC requires the linear parameters of the system (human heart rate model). In this research, the signal compression method (SCM) was used for approximately linearized modeling of the human heart rate. The extraction of the human heart rate model using SCM requires experiment and computation. Furthermore, it was observed in this research that SCM is not a precise method. Therefore, integral sliding mode control (ISMC) was designed and implemented to overcome these difficulties. By introducing an auxiliary sliding surface, the reaching phase and effect of the perturbation on an actual sliding surface were eliminated; furthermore, implementation of ISMC does not require the linear parameters of the system. Simulations were performed in MATLAB/Simulink and experiments were conducted in a hospital. Six clinical subjects participated in this experiment. Both forms of control logic were implemented during the desired heart rate tracking test. Results showed that the desired heart rate tracking of ISMC is better than that of SMC. The tracking error of ISMC is smaller than that of SMC. However, ISMC control output has chattering, which needs to be reduced.