Shunshoku Kanae

An Adaptive Robust Nonlinear Motion Controller Combined With Disturbance Observer

Parameter adaptation and disturbance observer (DOB) have been considered as two contrastively different approaches to handle uncertainties in motion control problems. The purpose of this brief is to merge both techniques into one control design with theoretically guaranteed performance. It is shown that the DOB compensates low-passed components of the lumped uncertainties without the necessity of parameterization, whereas the adaptive mechanism is only automatically activated in the cases where the fast-changing components of the uncertainties beyond the pass-band of the DOB can be parameterized by unknown parameters. It is thus shown theoretically how the DOB and adaptive mechanism play in a cooperative way so that the controller is more effective than the individual ones. Simulation results are provided to support the theoretical results.

A Unified Framework for Bias Compensation Based Methods in Correlated Noise Case

This technical note presents a unified framework for bias compensation principle (BCP)-based methods applied for identification of linear systems subject to correlated noise. By introducing a non-singular matrix and an auxiliary vector uncorrelated with the noise, the unified framework is established. Since there are rich possibilities of the choices of the introduced matrix and vector, the proposed unified framework is very flexible. It can be verified that the existing BCP-based methods are special cases of the achieved result. It also shows that the unified framework can be used for deriving new or simplified versions of the BCP type methods.

Estimation of pulmonary elastance fuzzy model by data combination of two respiration phases

Pulmonary characteristics differ in patients, and the suitable setting of ventilation condition is needed for every patient in the artificial respiration. The pulmonary elastance is one of the important features of lung, and it is a basis for deciding the airway pressure limit value. To get the pulmonary elastance of the of the patient from measurement data of the artificial respiration, the fuzzy logic technique has been proposed for estimating the pulmonary elastance and the static P - V curve in our previous works. In this paper, a new technique of fuzzy modeling based on data combination of two respiration phases is proposed to improve the estimation precision, and some estimation examples using real patient data are given to illustrate the superiority of the proposed method over the previous algorithm in the precision.

Decentralised robust control of interconnected uncertain non-linear mechanical systems

This paper proposes a new method of decentralised robust control for large-scale interconnected uncertain non-linear mechanical systems, by using disturbance observers. Rigorous stability analysis is given for the overall non-linear system. Simulation results on a coupled double pendulum system are presented to confirm the established theoretical results.

L1/2 Norm Regularized Echo State Network for Chaotic Time Series Prediction

Echo state network contains a randomly connected hidden layer and an adaptable output layer. It can overcome the problems associated with the complex computation and local optima. But there may be ill-posed problem when large reservoir state matrix is used to calculate the output weights by least square estimation. In this study, we use L1/2 regularization to calculate the output weights to get a sparse solution in order to solve the ill-posed problem and improve the generalized performance. In addition, an operation of iterated prediction is conducted to test the effectiveness of the proposed L1/2ESN for capturing the dynamics of the chaotic time series. Experimental results illustrate that the predictor has been designed properly. It outperforms other modified ESN models in both sparsity and accuracy.

Subspace identification of Hammerstein systems with frequency domain data

This paper addresses identification of Hammerstein systems using the sampled input-output data in frequency domain. When the static memoryless nonlinear part of Hammerstein model can be considered as a polynomial with a known order, the following linear part can be estimated as a multi-input single-output (MISO) systems with the numerical algorithm for subspace state space system identification method. Simulation results illustrate this approach is effective.

Pulmonary elastance estimation considering respiratory periodicity

To perform artificial respiration safely and comfortably, it is necessary to get the information on characteristics of patient respiratory system timely, and to set the ventilation conditions fitted to each patient under the information. Based on our earlier works on modeling and estimation of respiratory system, in this paper, data averaging method, parameter averaging method and elastance averaging method are introduced to reduce the effects of measurement noise, drift and perturbation of breath, and to improve the estimation accuracy of pulmonary elastance.

A New Respiratory Model considering Hysteresis for Artificial Respiration

Abstract Building a precise respiratory model is very helpful for setting appropriate ventilation conditions to fit each patient when artificial respiration is performed on the patient. The authors have proposed two types of second order nonlinear differential equation model of respiratory system. However, these model cannot cover the hysteresis characteristic of pulmonary elastance. In this paper, A new respiratory model with two sets of different parameters for inspiration phase and expiration phase is proposed. In addition, a parameter estimation algorithm dealing with static constraints is addressed. Simulation results are presented to show the effectiveness of the proposed method.

Robust output feedback control of a magnetic levitation system via high-gain observer

This paper proposes a novel robust output feedback controller for an electromechanical system in the presence of external disturbance and uncertainties of physical parameters. By exploiting the cascade features of backstepping design, a simple disturbance observer is proposed to suppress the effects of the uncertainties, and a high-gain observer is applied to estimate the unmeasureable states of the system. Strict analysis of the nonlinear control system is given. Experimental results are provided to support the theoretical results.