Nonparametric Hammerstein Model Based Model Predictive Control for Heart Rate Regulation

Medicine and Biology Society
Publication Type:
Conference Proceeding
Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, 2007, pp. 2984 - 2987
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2006015232.pdf527.62 kB
Adobe PDF
This paper proposed a novel nonparametric model based model predictive control approach for the regulation of heart rate during treadmill exercise. As the model structure of human cardiovascular system is often hard to determine, nonparametric modelling is a more realistic manner to describe complex behaviours of cardiovascular system. This paper presents a new nonparametric Hammerstein model identification approach for heart rate response modelling. Based on the pseudo-random binary sequence experiment data, we decouple the identification of linear dynamic part and input nonlinearity of the Hammerstein system. Correlation analysis is applied to acquire step response of linear dynamic component. Support Vector Regression is adopted to obtain a nonparametric description of the inverse of input static nonlinearity that is utilized to form an approximate linear model of the Hammerstein system. Based on the established model, a model predictive controller under predefined speed and acceleration constraints is designed to achieve safer treadmill exercise. Simulation results show that the proposed control algorithm can achieve optimal heart rate tracking performance under predefined constraints.
Please use this identifier to cite or link to this item: