Yoh Yonezawa

An Indirect Method of Computing Time Optimal Control of Linear Systems Considering the Terminal Error Norm

Eaton (1962) has solved theoretical problems in computing time-optimal control of linear systems. But in the practical point of view, his method has two main drawbacks with respect to both the convergent rate and the accuracy of the solution. In this paper, we discuss a new indirect method for the terminal control problems, taking into account of the fact that such an approximate control is practically required as making the terminal state error norm smaller than some acceptable value. And combining the method with Eatons method, a new effective algorithm of finding time optimal control is proposed. Also the integrating method to prevent the accumulation of round off error at the evaluation of the cost function is proposed. The accuracy and the efficiency of the method are shown by using several numerical examples.

Automatic measurement system for crater wear depth of a throwaway chip by the optical cut method

This paper describes a software design of an inexpensive automatic image processing system. In order to avoid the use of a threshold operation, center ordinates of strip images of objects are assessed by means of the weighted mean. Further, to decrease an influence of the chip surface roughness (2-3μm) on accuracies of measured values, coefficients of a standard line, utilized to calculate the wear depths, are determined by the moving average method. Thus, the system detects the maximum crater wear depth with ±1μm accuracy and represents a morphology of the chip wear. The processing time of the system is about 2 minutes per wear.

A New Disturbance Compensating Method in Model Reference Adaptive Control System

This paper proposes a design method of MRACS (Model Reference Adaptive Control System) for unknown plants in the presence of deterministic disturbance or periodic disturbance, so that the output error will converge to zero as time tends to infinity.In the traditional adaptive control systems dealing with disturbance, the disturbance was usually restricted to be deterministic. To remove the effect of disterbance, the general way was to use extended system including a disturbance model, or to use integral compensators. Here, a new disturbance compensating method is proposed, in which disturbance compensating signals are introduced into MRACS, and are combined adaptively to cancel the effect of disturbance. In this proposed method, because the filter signals of input and output are composed in the same way as the disturbance free cases, the MRACS has a simple constructure. Even for the periodic disturbance with unknown shape, if the period is known, by using Fourier series, the proposed method is also possible to remove the effect of this sort of disturbance. Furthermore, being used, disturbance compensating signals in MRACS, it is possible to take the initial responses of filters into account, and the characteristics of parameter convergence can be improved. The effectiveness of this proposed method is illustrated by some computer simulations.

Robustness of Double-Adaptive Observers

The usual process identification methods for parameter adjustment, along with dead zone, are effective in providing a robust adaptive observer of an unknown plant under uncertain input disturbance and output measurement noise. But it is difficult to decide the size of the dead zone due to the lack of knowledge concerning the disturbance. Furthermore, an error may remain in the dead zone despite the existence or nonexistence of the disturbance and the noise. In this paper, we propose a double-adaptive observer and discuss its robustness. Unlike previous work, this scheme does not introduce parameters that depend on the characteristics of the unknown disturbance. So, when the disturbance or the noise does not exist, the parameters converge over time. The stability of the system is discussed and computer simulation results are presented.

Position Inspection System for Welded Nuts by Image Processing.

開発したシステムは現場での実験結果より, 周囲の明るさの変動に関係なく, 十分な処理速度で溶接ナットの位置を確実に検査でき, 実用に供することができると判断した.また, 独自に考案した動的しきい値決定方法も良好であることが確認できた.この検査原理は明るい画面内の暗い部分の位置, 大きさを検査するものである.従って, ナットだけでなくその他の位置検査にも応用可能である.

An Experimental Examination of Various Methods of Designing Constrained Controllers for Linear Systems

In order to investigate the practical aspect of a series of constrained control laws proposed by the present authors, an experimental examination using a servomechanism as a controlled object has been done. The control laws are as follows : (i) linear control, using an admissible gain, (ii) variable gain feedback control, (iii) relay control via the gain satisfying the Popov condition, and (iv) relay control for a quadratic performance index.The results of the experiment have shown high performance of Controls (ii), (iii) and (iv). Control (i) is not effective, because of the dead zone in the servo system. These control laws are so simple that the sampling time for Direct Digital Control can be 30-70 ms when using the interpreter-type BASIC program on a 16-bit personal computer (the clock frequency of CPU is 10 MHz).