Heart failure which is reduced contraction of the ventricular myocardio is one result of the serious heart disease. The condition of heart failure patients depend chiefly on the level of impaired functioning in the right and left ventricle of the heart, the cause of the disease, and the presence of arrhythmia. Only a doctor with considerable experience can accurately diagnose heart failure. This paper describes a diagnostic system of classification employing a circulatory model. The system employs a simple exercise and a circulatory model that is based on information about partial pressure of carbon dioxide in the arteries during exercise. When a patient suffers from heart failure contraction of the ventricular myocardio declines, stroke volume depends chiefly on the heart rate. Exercise remarkably changes stroke volume in a patient with heart failure. Stroke volume is simulated using the circulatory model, which is composed of parameters such as heart rate, mean blood pressure, and partial pressure of carbon dioxide in the artery during exercise. The measured results of stroke volume during exercise using the ultrasonic quantitative blood flow measurement system coincides with the simulated results of stroke volume with the use of the circulatory model. While the ultrasonic quantitative blood flow measurement system is expensive, the diagnostic system proposed in this paper can be realized economically with conventional 16-bit microcomputers.
A solid-state pulse generator has been developed to be applicable to drive a high power pulse laser. The new technology provids a promising mean to replace the thyratrons. The pulse generator is constructed with series and parallel connection of MOS FET (Maximum Rating; 1, 000V, 8A) without any saturable reactors for sharpening the current pulse waveform. With 60-parallel and 20-series connected of FETs, the characteristics of the pulse generator are that the switching time is 38ns, the rate of rise of the current pulse is 20A/ns, and that the jitter of the turn-on time to the input signal is ±0.42ns. Appling the pulse generator with 120-parallel and 30-series connected of FETs to a 80mm bore copper vapor laser, the laser power obtained is 112W when switched pulse energy is 2J with a repetition rate of 5kHz.
After some training, human operators can control manually very unstable objects when some proper informations are given. But they can hardly explain how they do it, because they operate intuitively and not logically. In this paper, we study the human behavior during the control of a double inverted pendulum and identify their control rules experimentally. The motion of a double inverted pendulum is simulated by a micro-computer and some of the state variables are indicated on CRT which are observed by a subject and controlled with key. In order to find which informations are used during the operation by a subject, his visual points are examined by an eye-camera. As the result, we see that there are three phases of operation, that is the decrease of initial deviation, the prevention of over-shoot, the keeping of stability. Next the motion of pendulum is analyzed qualitatively in each phase so as to identify the control rules of human operator. By this analysis, we see that the intuitive manipulation of human operator is quite reasonable from the physical viewpoint, and represent it by some linguistic rules. From these results, we suggest a hierarchical structure of fuzzy rules as a model of human operator which is verified through the experiments of fuzzy control. It is concluded that this fuzzy controller acts as a skilled operator but its performance is much superior to human.
The technology used in the navigation system to locate vehicle's position is dead reckoning. This navigation system, however, has a problem that errors may be cumulative. Suppose that sensor errors are reduced, cumulative errors are reduced but inevitable over a long drive. Map matching allows total accuracy of positioning relative to the street network. In this paper, a method of map matching with a simple algorithm has been proposed and examined by computer simulation. The method consists of two phases: the road selection and the correction of the vehicle's position. In a typical map matching simulation based on this method, the matching accuracy is improved from 5% to a value around 3%. The processing speed is close to five points per second. This performance is effective to improve the accuracy of dead reckoning.