An extremely sensitive, accurate and simple microwave level meter has been developed. This level meter consists of a continuous wave radar in which carrier is modulated by the M-sequence (Maximal-length-sequence). Employing two M-seq. which are the same in code but generated by the slightly different clock frequencies, the implementation of the crosscorrelation between the received and the reference signals can be simple. This radar successfully gains the sensitivity and suppresses the noise. Moreover, the quantization error of this radar can be designed smaller than the conventional FMCW (Frequency-Modulated-Continuous-Wave) microwave radar. The principle of this microwave level meter is studied and its design criteria are described.
This microwave level meter is useful in the case that exclusive performance in terms of sensitivity and accuracy is required for the crucial measurement conditions in the environment such as the iron and steel making process where dust and flare may make it difficult to employ the laser beam or supersonic as sensing waves. The application of the M-seq. modulated microwave level meter is discussed in the experiments measuring the metallurgical slag and the molten metal levels.
In the prototype microwave level meter, the carrier frequency is 10GHz, the output power 40mW, the clock frequencies driving the two M-sequence generators about 220MHz, the difference of the two clock frequencies 5.4kHz or 10kHz, and the lengths of the M-seq. 127bit or 1, 023 bit. The water-cooled horn antennas with the gain of 17dB are used. The slag level in a mini-converter was continuously measured with the accuracy of ±88mm. The sensitivity was high enough to detect the weak reflected wave from the slag surface whose reflection coefficient was -40dB. The molten metal level in a tundish of a continuous casting process was measured with the good accuracy of ±20mm.

View full abstract

This paper describes about a shape data acquisition system in which a vision and a new type of proximity sensor are integrated for the detection of various 3-D data required in handling tasks by robot.
The vision sensor is used for acquiring the entire image of objects over the working area and extracting the edge image which is used to positioning the proximity sensor to the interesting area on the 3-D object. The proximity sensor detects the range data and also extracts the jump edges of the surface in the interesting area.
In the developed proximity sensor, a ring beam pattern which is generated by passing a laser beam through a cone lens, is projected onto an object; range data are detected from the position of the objective lens where the ring beam pattern is focussed to a point. The position and the direction of a jump edge are estimated by analyzing the image of a ring beam pattern projected on the jump edge.
To confirm the basic principle of the system, experiment is carried out on the surface of the regular icosahedron.

View full abstract

Measurement of instantaneous velocities of athletes or a ball are often helpful for them to understand clearly the relationships between their body motions and the resultant performances, if essential information extracted from the measurement data is available in real time. For this purpose an optical remote measurement system was developed and its measurement capabilities were examined.
A hexagonal rotatory mirror scanned horizontally a vertical slit beam at 360Hz across a measurement field of 20m in diameter. A special reflector was stuck on a moving object, which was made of numberless micro corner cubes.
A photo detector aligned on the optical axis of the scanner captured reflected light just when the beam passed over a moving object. By determining the exact time of reflection via a correlation filter and an interpolation method, current object poistion along a predetermined trajectory was calculated in realtime on every 2.78ms.
The system has following capabilities. Field of view: 56°, resolution: 0.06mm (at a measuring distance of 10m), 0.18mm (at 15m), accuracy: 0.7mm across the whole range. Applicability to measurement of multi objects was also investigated. By sticking a different bar coded pattern of reflector on each object, discrimination ability of the correlation method among 3 objects were examined. Discrimination among far more than 3 objects was seen possible.

View full abstract

A fiber-optic particle size monitor using dual-beam laser interferometer which forms two sets of intersecting laser beams with different intersection angles is proposed. The output of this monitor is a 2-bit binary code which indicates whether the measuring particle size is within the preset size range, or larger or smaller than the range. The preset size range is changed by changing the intersection angles of illuminating laser beams. The dual-beam interferometer can easily change the intersection angles without changing the intersecting point. Therefore the position of measurement volume is kept unchanged when the preset size range is changed. The scattered lights by a particle in the measurement volume are detected by an optical fiber and transmitted to photodetectors. The visibilities of photodetector outputs corresponding to respective intersection beams are measured and compared with the threshold values. Then, the 2-bit binary code is obtained.
The setting procedure of preset size range is discussed in terms of configuration parameters of this monitor, such as the intersecting angle of illuminating laser beams, the spatial frequency selectivity of detection optical fiber and the threshold level of visibility. The experimental results using glass spheres confirm the operation of this monitor.

View full abstract

This study is concerned with an optimal servo system with reduced sensitivity in the low frequency range.
The design is based on the same principle that we used for the continuous system. It is a model matching method with pole-zero cancellations whose model is an optimal control system without servo compensator designed using the optimal regulator technique.
It is pointed out that, in the digital control system, a configuration of the cancelled polezero pairs becomes a parameter for the trade-off between the sensitivity and the stability of the control system. A numerical example shows that sufficient trajectory sensitivity characteristics are gained at the expense of a little loss in the stability.
Moreover, a new caluculation method of the initial value of the servo compensator is also proposed.

View full abstract

In this paper we propose design methods of adaptive deadbeat controller for non-minimum phase deterministic linear systems, based on weighted control criteria.
First, we will construct a basic controller on the basis of the weighted control criterion of the reference input, control input and system output. For the basic controller we use a stable compensator to guarantee the stability of the closed loop system. Adaptive deadbeat control is thus possible for unstable and/or non-minimum phase systems.
We introduce special control criteria to determine the control polynomial parameters. Several adaptive deadbeat control laws are proposed:
1 Minimizing the tracking error
2 Minimizing the effect of the disturbance
3 Minimizing the sensitivity function to reduce the effect of estimation error of parameters.
The controllers used have two degrees of freedom. The pre-compensator is not rational, but polynomials. Therefor the calculation of the controller is relatively simple.
Finally, the effectiveness of proposed methods is shown through simulation for non-minimum phase plants.

View full abstract

In most of the studies of Model Reference Adaptive Control (MRAC), the controlled systems are confined to minimum phase systems, since the MRAC techniques utilize the control laws involving cancellations of zeros of the systems. It makes the scope of application of MRAC too restrictive, for non-minimum phase discrete-time systems can often appear. For example, when continuous-time systems with relative degree greater than 2, are sampled at a fast rate, those give rise to non-minimum phase discrete-time systems. Hence, the study of MRAC for non-minimum phase systems is of great importance.
In the present paper, we propose a design method of MRAC for non-minimum phase systems. The poles and zeros of the controlled systems are relocated by the periodic time-varying feedback control with multirate sampling, and no cancellation of zeros occurs in our method. It is shown that even if unstable zeros exist, the output error converges to zero asymptotically, while the control input remains uniformly bounded. Finally, some simulation results for the non-minimum phase system also show the effectiveness of the proposed method.

View full abstract

This paper derives a mathematical model and investigates its properties for vibration control of a flexible beam attached to a moving base. When the external forces exerted on the base plus beam system are zero, the mass center of the whole system remains unchanged with respect to the inertial coordinate. So that if there is vibration of the flexible beam, the base has to move accordingly. This motion in turn affects the motion of the flexible beam. It is shown that the equations of motion of the flexible beam can be written as a perturbed second order evolution equation, with the perturbation term reflecting the effects due to the motion of the base. The existence of a weak solution is first proved by constructing an approximate solution. Furthermore, it is made clear that the vibration frequency of every mode is increased by at least a certain amount. These new findings are believed to be useful in modeling and control of flexible structures.

View full abstract

Recently, the emission regulation tends to be intensified and it becomes important to control the air-fuel ratio accurately in the car engine.
To realize this, it is necessary to detect the air mass flow rate at the inlet port precisely.
In the present control system, an air flow meter or a pressure sensor is used to detect the air mass flow rate. But the precise air mass flow rate at the inlet port is not detected in the transition because of the measurement delay, the sensor position, the pulsation suppressing filter and the control structure that the fuel injection pulse is calculated before the induction stroke.
In this paper, first, the new inlet air detection scheme is proposed. The scheme estimates the intake air flow based on the physical model using throttle angle and engine speed, which have small measurement delay and no effect of pulsation.
In the present system, the future air mass flow rate is necessary for determination of the proper fuel injection pulse because of the above mentioned control structure. In this paper, two schemes for predicting the future air mass flow rate are proposed. One is the indirect prediction scheme which is based on the throttle angle prediction, another is the direct prediction scheme which is based on the air mass flow rate prediction.
Lastly, the fuel control system based on the predicted air flow rate is evaluated by experiments and it is shown that the proposed method is superior to the conventional method in the air-fuel ratio control performance and the driving performance.

View full abstract

Having closed loop structure, parallel manipulators are able to generate internal forces in the joints between the members. This can be achieved by redundantly actuating the manipulators. Although the intetnal forces are the vectors of force elements that balance each other and do not create any total force, they change the structural stiffness of the manipulators.
In this study, the effect of internal forces on the stiffness of the parallel manipulators was clarified based on the analysis of a six degrees of freedom (d.o.f.) spatial parallel manipulator which is composed of at least three sub-arms each having 3 d.o.f.. In the analysis, the static equilibrium state and corresponding internal forces were determined from joint servo potential. Then the stiffness of the manipulator was determined around this equilibrium state. It was shown that, internal forces cause an additional stiffness which depends on the geometry of the manipulator and the magnitude of the internal force. The results were verified both numerically and experimentally for a 3 d.o.f. planar parallel manipulator which is composed of two sub-arms each having 2 d.o.f..

View full abstract

PID control has been widely used in process control. The main feature of PID control is a simple algorithm constructed by P-action, I-action and D-action.
It is, however, difficult to satisfy control performance commonly for different conditions such as start-up, shut-down and feed switching. In many such process control systems, expert operators play an auxiliary role for compensating PID control.
The aim of the paper is to design fuzzy-PID combination control system which replaces expert operation with fuzzy logic controller. The fuzzy logic controller consists of 15 linguistic control rules and is designed for improving the performance of PID control.
The fuzzy-PID combination control system is applied to real charge heater outlet temperature control of hydrogen unit. Finally, it is shown that the fuzzy-PID combination control system can control heater outlet temperature perfectly not only in stationary state condition but also in nonstationary conditions such as start-up, shut-down, and feed switching.

View full abstract

We developed an EMG controlled prosthetic forearm with three degrees of freedom actuated by small size ultrasonic motors. Its weight is less than 700g and the size is the same as the adult's forearm. The conventional proosthetic arm using DC motors produces a motion noise. So the reduction of the motion noise was one of the big problems to be soleved. Since our prosthetic forearm uses ultrasonic motors, it produces no motion noise. In addtion, the amputee can control six kinds of motions i.e. pronation and supnation of the forearm, flection and extension of the wrist, and grasping and hand-opening. The interface between the amputee and the prosthetic forearm was designed on the basis of the fact that the amputee still preserves the phantom limb motor map after amputation. It can discriminate the amputee's intended motion among six kinds of motions using the EMG signals from the remained muscles and also produce the command signals to control the prosthetic arm. Consequently, the amputee is able to manipulate the prosthetic arm as he intends and his control load is relieved sharply.

View full abstract

As actuators for an prosthetic arm, we use traveling wave ultrasonic motors because of their light weight, compact size and silent motion.
The prosthetic arm has to better move compliantly around every joint of it according to environmental force the same as the human arm. In order to realize the compliant motional prosthetic arm we think that the actuator have a compliant characteristic, but the traveling wave ultrasonic motor has no compliant characteristic in itself. Incidentally, the traveling wave ultrasonic motor is driven by two different phase sine-wave electric signals which propagate elastic traveling waves in its stator, then the traveling waves generate torque to its rotor. Though, usually the phase difference is adjusted to just +90deg or -90deg to get highest output torque and highest rotation speed, we adjust it in the range of -90deg to +90deg to regulate output torque and rotation speed continuously by the phase difference change. Then we control the phase difference with deflection angle of the output shaft from desired angle and with desired elastic coefficient.
We realized adjustable compliant motion on the traveling wave ultrasonic motor using the phase difference control of the two driving sine-wave electric signals. The motor acts as a spring with damping. The damping characteristic is observed as a constant speed response of the ultrasonic motor to the step change of the phase difference of the driving signals.
This system is useful for actuators of the prosthetic arm.

View full abstract

The estimation of Markov jump processes from noisy observations is a nonlinear estimation problem with strong nonlinearity. A new approach to solve the estimation problem is presented, using a neural network model. The neural network is designed to minimize an energy function, which consists of two terms: one represents reliability of observation data and the other imposes penalties on estimates on the basis of a priori information about the processes to be estimated. It is shown that nearly optimal estimates are obtained using the neural network as a sliding window filter. The quality of the estimates depends on the ratio between two terms in the energy function. It is also shown that an adequate value of the ratio is learnable from samples of true processes and observation data using a stochastic approximation method.

View full abstract

There have been many new developments in neural network (NN) research, and many new applications have been studied. The classification of remotely sensed multispectral data using classical statistical methods has been worked on for several decades. Among the multispectral data, we concentrate on the LAND-SAT-5 Thematic Mapper (TM) image data which has been available since 1984. Using the classical maximum likelihood approach, a category is modeled as a multivariate normal distribution; however, the distribution for LANDSAT images is unknown. It is well known that NN approaches have the ability to classify without assuming a distribution. We apply the NN approach to the classification of LANDSAT TM images in order to take the spatial and spectral information into consideration, and we investigate the utility of this approach for classification. We confirmed that the NN approach behaves similar to human perceived classification.

View full abstract

Rhythmic movements of walking, swimming, etc. are controlled by mutually coupled endogenous neural oscillators. These rhythms coordinate with one another to generate temporal and spatial moving patterns suitable for their environments and purposes. The walking motion of quadruped is a representative example; here, the periodic motions of each limb coordinate to generate stable gait patterns suitable for the organism's moving speeds and environments. These patterns are mainly decided by the locomotive speed; i.e. as a cat moves faster, the gait patterns change from “walk” to “trot”, and lastly to “gallop”. This moving pattern generator system can be ragarded as one of the autonomous distributed systems which generate global patterns suitable for their environments and purposes.
This paper shows an approach for constructing such a moving pattern generator system by use of the bifurcation concepts in nonlinear system. Let the total dynamic system that represents a global pattern be a gradient system with an auxiliary potential function, then the desirable patterns can be generated by specifying the potential functions. Also, the patterns are changeable by altering the potential functions. Further, using the bifurcation concepts, it is possible to construct a system that suitably changes patterns discontinuously according to continuous change of a parameter. This synthesis approach is applied to construct a gait pattern generator of a quadruped artificially. A gait pattern generator is realized to couple four oscillators in which each oscillating state is regarded as each limb's rhythmic motion. Finally, it is shown using computer simulations that the proposed system generates and changes patterns suitable for some moving speeds; i.e. “walk”, “trot” and “gallop”.

View full abstract