Since composite structures applied for supporting super conductivity coils are covered by structures forinsulation, health monitoring systems are very important for the structures. Composite structures will be used in strong variable magnetic fields. This causes significant difficulty in applying electrical methods for detection of defects. Although we can use several optical strain sensors, they are still expensive and have some problems of long term reliability. In the present study therefore, a novel optical strain sensor using power loss by macro scale bending was developed and investigated in detail. A curved optical fiber was attached to specimens. When the specimen is tensioned, the optical power loss changes due to the change of radius of the curved optical fiber. The optimum fiber path configuration (quarter circle) was found by computations. The applicability of the sensor was investigated experimentally under static and cyclic loading. As a result, the sensor was proved to be very effective and to be reliable at low cost.

抄録全体を表示

In the present study, real-time measurement of internal and residual strains in GFRP laminates with an optical fiber strain sensor was performed in an autoclave molding. An Extrinsic Fabry-Perot Interferometric (EFPI) type optical fiber strain sensor was embedded into prepreg sheets before the molding. The ion viscosity and the temperature of the epoxy resin matrix were also measured simultaneously with the internal strain measurement. The results of the experiments showed that the internal and residual strain of GFRP laminates during the molding could be monitored by the optical fiber sensor embedded in the parallel direction to the reinforcing fibers. It was also found that the circumstances around the embedded optical fiber strain sensor should be taken into account for the interpretation of the measured internal and residual strain.

抄録全体を表示

This study is concerned with the development of a new type of sensor for vibration control. This sensor belongs to the class of seismo-type. The main features of this sensor compactness, robust against shock and also to measure velocity and displacement simultaneously under the low frequency range. Until now, in order to measure the absolute displacement under low frequencies, the conventional sensor has to be made with its low natural frequency. To realize low natural frequency, sensor must have large mass with soft spring. That is to say, these sensors are big and weak against shock. To solve these problem, the proposed sensor uses positive displacement and negative acceleration feedback for reducing its natural frequency actively. Therefore, it is possible to construct a sensor with light mass and hard spring for realizing the requirement.

抄録全体を表示

Multimodal vibration control of a cantilever beam using bonded piezoelectric film sensors and actuators is studied both theoretically and experimentally. Equations of motion of the Bernoulli-Euler beam with collocated bonded piezoelectric sensors / actuators are derived. Then, direct velocity feedback control is employed as the vibration control law, and applying Lyapunov's direct method, it is shown that this control law is stable for arbitrary vibration mode. Transient responses of the beam, with control and without control, are studied for the cases in which an impact force is applied to the tip of the beam. Numerical results and experimental results show good agreement, which shows the validity of the proposed control scheme. It is also shown that first three bending modes are efficiently controlled, and that in order to achieve more broadband vibration control, it is necessary to consider the dynamics such as extensional vibrations of the beam which is not considered in the present study.

抄録全体を表示

This paper verifies the performance of the active microvibration control system using piezoelectric actuators with a table length of 3 m, a width of 1.8 m and a weight of 4000kg applied to large-scale semiconductor manufacturing equipment whose microlithographic accuracy was actually degraded due to the constant vibration. The mounted equipment was the integral-design semiconductor manufacturing equipment weighting 2600 kg. The controller was designed by a method dependent on the normalized coprime factor plant model in the preceding study used to prevent the elastic vibration of the equipment and table from causing the spillover phenomena of the control system. For the frequency-forming algorithm, the model matching method and the H^∞ method were employed. As a result of this system's application, the microlithographic accuracy of the manufacturing equipment could be improved from about 2μm (before application) to an intrinsic value of nearly 0.1 μm.

抄録全体を表示

To apply smart structures to buildings, a new vibration control strategy is proposed for flexural-shear-type frame structures with smart structure using piezoelectric actuators. Actuators are incorporated in columns, and the response of the entire structure is reduced by control of the bending moment and axial force of the columns. These two controls are applied in combination by appropriate arrangement of actuators. Two control strategies were tested. One is model matching method and the other is H^∞ control theory. Excitation tests were conducted with a 4-story frame structure with a total weight of 2000 kg and a height of 3.7 m having H-section steel beams incorporating actuators as columns. The effectiveness of combined use of bending moment control and axial force control of columns with smart structures was confirmed.

抄録全体を表示

A multiaxial constitutive model to describe the pseudoelastic and shape-memory behavior for TiNi shape memory alloys has been developed from a phenomenological point of view. The rhombohedral and martensitic phase transformations of TiNi shape memory alloys have been taken into account. A generalized expression for the transformation strain range which depends on applied stress and current phase volume fraction has been proposed to obtain closed hysteresis loops for closed cyclic loading paths. The predictive capability of the present model for the hysteretic behavior of TiNi shape memory alloys has been demonstrated through simulations of the stress-strain curves under isothermal uniaxial and multiaxial loading-unloading cycles. A comparison between existing constitutive models for shape memory alloys has been also presented.

抄録全体を表示

This paper describes experimental and numerical results of the transient response of a Shape Memory Alloy (SMA) plate considering thermomechanical coupling and an application of the SMA plate to an adaptive fludi-flow controller. SMA wire has been used as a component of actuators to date. But SMA plates have some advantages over SMA wire, such as larger force generation, shape change and heat generation due to induced current. Therefore, since plates might be used in actuators, especially in microactuators, it is important to evaluate their dynamic response. In this study, the dynalmic response caused by the transformation between the R-phase and the parent was evaluated by both numerical analysis and experiment. An SMA plate remembers an arc shape at high temperature and a flat shape at low temperature. Thus, this plate shows an all-round shape memory effect. In this analysis, SMA temperature was calculated considering latent heat due to the transformation, then the transformation kinetics was obtained and finally dynamic response was evaluated. The numerical results of these responses almost agreed with the measured ones. As an application of this SMA plate, we fabricated an adaptive fluid-flow controller. This controller can separate hot fluid and cold fluid, and direct them to their respective outlets by changing the shape of the SMA plate. It continued to operate even after one thousand heat cycles. This SMA actuator can move without sensors and bias springs.

抄録全体を表示

Crack closure in an epoxy resin plate with embedded shape memory (SMA) as an actuator in intelligent structure members is investigated as a function of the length of crack, duration of the supply of electric current to SMA and the volume fraction of SMA under tensile loads using photoelastic experiments. The effect of heat generation due to the SMA in the plate is also investigated. The crack closure is evaluated by the stress intensity factor obtained from photoelastic fringes generated near the crack tip. The results show that the closure strongly depends on the crack length and duration of the current supply. The closure is not found in the plate with a crack which doesn't pass through SMA. A complete cessation of the supply causes the disappearance of the closure with time. Continuous current supply results in change in quality of the epoxy matrix and separation between SMA and the matrix. The intermittent current supply prevents the separation and maintains the closure.

抄録全体を表示

This paper discusses change in stiffnesses with temperature of composite laminates in which TiNi shape memory wires are embedded. Young's modulus in Shape Memory Alloys (SMA) markedly increases with temperature passing through its phase transition temperature. GF laminate embedded with SMA wires can therefore increase its stiffnesses with temperature. The inplane stiffnesses of UD laminates with SMA wires are measured, and it is shown that the SMA does have influences on all inplane elastic coefficients. The stiffnesses of UD material can be used as the basic ingredients for the classical lamination theory in which stiffness change with temperature can be accounted for. Specific examples of symmetric laminates having layers with and without SMA are calculated. This paper can serve as a base for stiffness design of composite laminates possessing such novel function that the laminate can increase its rigidity by e. g., electric current and therefore can avoid vibration resonance.

抄録全体を表示

This paper deals with a controllable small vibration-proof mount designed for precision devices equipped on cars and other transportation vehicles. The properties of this mount can be controlled by the applied electric field strength, and this is one of application studies of electrorheological (ER) fluid. In this study, liquid crystal was selected as ER fluid. Liquid crystal is a homogeneous organic compound characterized by molecular orientation responsding to external electric or magnetic field, and because of molecular orientation, its viscosity can be varied by the applied electric field. In the present paper, a mount of small size designed for mass-produced vehicles is constructed and its performance was investigated. It is shown experimentally that the vibration amplitude of a mass supported by the present mount can be decreased by changing the damping properties.

抄録全体を表示

A liquid crystal is usually applied to displays of televisions and computers. The objective of this paper is to clarify the electrorheological effect of a liquid crystal under an external electric field and to make a fundamental study on a rotary-type damper using a liquid crystal. A controllabie rotarytype damper using liquid crystal with nematic phase is proposed and applied to the vibration suppression of a rotary shaft and a pendulum. Two disks are mounted inside a housing which is attached to the rotary shaft. These disks are free to rotate inside and relative to the housing. Damping is introduced by filling the space between housing and disk with liquid crystal. Damping torque is produced by the controllable viscosity of the liquid crystal. The results show that the torsional vibration of the rotary shaft, and the free vibration and parametric resonance of the pendulum are suppressed effectively.

抄録全体を表示

This Paper describes the dynamic behavior of a fluidic oscillator. This fluidic element, which can be used to actuate or control micro devices, was investigated by a monotone stremline upwind finite-element simulation as well as an experiment. The element, if its size is several millimeters, might be used as a kind of a micro-intelligent actuator without any moving parts and electricity. A fluidic oscillator treated here has two feedback channels. A periodic switching between two output ports is achieved by feeding back a part of flow. The dynamic behavior of the oscillator, whose nozzle width is 1 mm, was described. A fluidic oscillator experiment was carried out using water. The oscillation frequency is found to be dependent on supply volume flow rate and has a lower limit of frequency. Amplitude of pressure oscillation is also dependent on supply volume flow rate. To investigate the dynamic behavior of the fluidic oscillator, numerical simulation was carried out and the resuits were compared with those by measurement. A fluid jet keeps winding shape while the jet is switching. The relation between oscillation frequency and supply volume flow rate shows the same tendency as that of experiment.

抄録全体を表示

The vibrations of a structure having a vibration absorber without damper vanish at the anti-resonance frequency theoretically. Hence if the rigidity of the absorber is controlled to produce an anti-resonance state according to the forced frequency, the vibrations of the structure can be deleted for sinusoidal forces. However, external vibration force involves higher modes, in general, so the structure is excited by various frequencies. Hence it is difficult to delete the vibrations using the anti-resonance procedure. This paper presents a method to control vibration of structures by combining the auto-tuning control of anti-resonance and optimal dynamic absorbers. Theoretical results under the control have been presented. To validate the present method and the theoretical results, experimental tests have been performed.

抄録全体を表示

A method of vibration control for a column carrying a mass at its tip subjtected to earthquakes is described. A vibration control mechanism consisting of a gear train for the column is also described. Experimental tests have been performed for a model of the present mechanism. It is shown that the vibration displacements and the moments in the column are reduced significantly in comparison with those in a conventional column without the mechanism. In the present column, energies for controlling vibrations are not required because of the use of earthquake force as a control force. Therefore, the column has advantages over the columns having an active vibration control system.

抄録全体を表示

Many researchers have studied the corrugation of railway track rails. It is thought that the corrugation is caused by the vibration and slippage of the rail or the wheelsets, but to date a detailed analysis of entire system has not been carried out. In this research, we divided the rails and sleepers into small elements and simulated a three dimensional vibration system. We compared results with those obtained from an actual curved track, and found that both sets of results were in agreement. Therefore we concluded that the vibration mode which cause corrugation in this case was the twisting and lateral bending vibration of the rail.

抄録全体を表示

The flow of granular materials on the vibratory conveyor is studied both experimentally and analytically. The vibratory conveyor consists essentially of a near-horizontal conveying vessel which is undergoing translational harmonic vibrations in a direction to the horizontal. The motion of individual particles contacting each other is solved by the discrete element method proposed by Cundall. The Normal component of the contact force acting on particle is firstly given by the sum of the forces due to the spring and the dash-pot. Then, the tangential component of the contact force applies the Coulomb law of friction. It is shown that the mean velocity of transport of granular materials depends on the frequency and amplitude of the vibratory input as well as various physical parameters. Also, the numerically observed flow properties agree quantitatively with experimental results.

抄録全体を表示

The estimation theory of sound power which has been established based on the analytical modal analysis is extended to application based on FEM modal analysis. We tried to estimate the sound power radiated from a semisphere, which is a simple model of the lid of an automobile transmission case and is difficult to analyze the natural modes by analytical modal analysis. Next, the mesh size of the FEM model, with which the estimation of sound power will converge within the accuracy of ±0.3 dB, was investigated using some FEM models with various mesh sizes. As a result, the estimated sound power coincided with the measured values with an accuracy sufficient for practical usage, and converged within the accuracy stated above when the maximum mesh size was smaller than one-sixth the wavelength of the highest frequency component. These results were confirmed by applying them to similar structures, such as derby hat and silkhat structures.

抄録全体を表示

Hybrid experimental method, in which a vibration experiment using an actuator and a vibration analysis using a computer are simultaneously conducted by exchanging information with each other, is being developed as a new seismic experimental method. We developed a real-time hybrid experimental system using a hydraulic actuator with large excitation capacity and used it in a seismic experiment of an energy absorber. This absorber is a new type of pipe-support absorbing vibration energy by elastoplastic deformation By comparing the results of the hybrid experiment with those of the shaking-table experiment, we showed that the results obtained by the two experimental methods were close to each other. Therefore reliability of the developed hybrid experimental system was verified.

抄録全体を表示

A new tupe tuned-mass-damper, which is designed to reduce the response of main structures against not only weak earthquakes / wind but also destructive earthquakes, has been investigated. The basic construction of this system is obtained by dividing a TMD mass into two parts. The impact forces which are generated by the upper mass and stopper, reduce the displacement of the lower mass. Until the previous report, it was confirmed that the performance of the vibration control system was almost the same as that of an optimal TMD although the maximum amplitude of lower mass is markedly reduced to about a half of the TMD mass. In this report, we applied an active control into this system (for weak earthquakes) using an existing building model and confirmed the change rule, and investigated the performance of IHMD in comparison with an optimal TMD.

抄録全体を表示

This paper deals with vibrations of a preloaded compliance system as an example of vibrations of nonlinear systems with discontinuities. The resonance curves of harmonic, higher-harmonic and subharmonic vibrations are obtained by using the direct numerical integral method presented previously, which is a highly accurate type of shooting method. Chaos is generated in the system treated here. Influences of amplitude and frequency of an external force and damping ratio on the resonance curves and the stability of solutions are discussed. It is found that if a trajectory enters a discontinuous point, a bifurcation is realized. A stable periodic solution becomes unstable by this bifurcation and chaos may suddenly occur at the halfway point of the period doubling route. In this way, another chaos route was also found in a preloaded compliance system in the same manner as in the forced self-excited vibration system with dry friction.

抄録全体を表示

The bending strength of cantilever beam, which is formed by map function, has been investigated. The map function contains a parameter m and p which are able to change a shape of a curve, and its shape is symmetrical in the case of m=3. From the calculated results of stress, it has been found that the stress of cantilever beam formed by the map function of m=3 and m=1.5 shows the uniform strength, and the stress of beam of m=2 shows the minimum value. Though the A₁ bifurcation splits firstly in the case of m=2, the bifurcation of A₁ and the stress of cantilever beam are influenced by the shapes of a map function.

抄録全体を表示

This paper deals with a theoretical instability analysis of a rotating flexible disk subjected to swirling fluid flow in a confined fluid, and a influence of dissipative effect (structual damping, fluid viscosity and fluid friction) on the dynamic instability of the rotating flexible disk. The aim of this paper is to demonstrate the important effect of dissipation on the dynamic instability. In the instability analysis, the basic equations of swirling fluid flow around the rotating disk are derived by integrating Navier-Storkes equations over the gap width between the rotating disk wall and the shroud wall. The structural vibration equation of the rotating flexible disk is based on the Kirchhoff-Love plate model. The equations of coupled fluid-structure motion take into account the moving boundary conditions with respect to both the rotating disk and the fluid flow. The calculated results show that the dissipative effects play a key role in governing the threshold of dynamic (flutter) instability, and a influence of dissipative effect on the dynamic instability.

抄録全体を表示

The well-known methods for improving servo tracking performance are (1) increasing controller gain higher and (2) applying a feed-forward term. However, the former is limited by the stability of the system and the latter is also limited because in implementing a feed-forward controller, the command input needs to be differentiated, which may cause ripple or machine vibration. The controller proposed in this paper can achieve equivalent high gain performance by using comparatively lower position gain. Instead of implementing the conventional first order loop controller, the third order one is applied. By tuning the servo parameters properly, the circular tracking performance is improved by 400% compared to the first order controller with the same gain. Moreover, this controller can reduce the jerk of the commanded control input. Results show that by applying this controller, machine vibration is reduced and the response becomes more stable.

抄録全体を表示

The method of measuring static lateral deflection of a circular saw disk is often used to estimate the tensioning effect at the stage of industrial production. To clarify the theoretical background of this test estimation method, the relation between the tensioning effect and the deflection of a disk which is loaded by a concentrated force at the outer boundary is studied, analytically and experimentally using normal and tensioned disks. (1) Analytical results of static lateral deflection by a concentrated force at the outer boundary coincide with the experimental results in both cases of normal and tensioned disks. (2) The relation between the deflection and isotropic additional stress parameter (maximum stress, radial location and variance) by tensioning is cleared. (3) Maximum of isotropic additional stress by tensioning has a optimal value for dynamic stability to use as a rotating disk. Because, small additional stresses make instability, and large stresses make low stiffness. (4) The tensioning effect is properly estimated by the method of measuring static lateral deflection of a disk.

抄録全体を表示

The purpose of the present study is to develop a procedure to carry out dynamic calculations when a flexible structure moves along an arbitrary orbit or track. Finite elements' equations of motion regarding a moving flexible structure have been derived considering a geometric nonlinearity. A finite rotational matrix was used to transfer a vector or a matrix to one regarding another reference frame. Computer programs which give an eigenvalue analysis and a time history response have been developed using the derived equations. Natural frequencies and modes of vibration for some analysis models were calculated by the eigenvalue analysis progralm. We also calculated time history responses for a large deformation in the case in which the analysis models moved along a track.

抄録全体を表示

It is desirable to change the shift schedule of an automatic transmission according to the road gradient. We previously developed a drive torque calculation method, and used it to calculate the gradient of a road as which a car was traveling. Gradient resistance torque is calculated by subtraction of the acceleration torque and road load torque from the drive torque. By improvement of the detection method for determining vehicle speed and the masking method for eliminating interruption by braking, a gradient calculation accuracy of ±20% is obtained. This is not a as good as in the case of use of the conventional sensing method, but the new method can be implemented without a torque sensor. We studied the possibility of applying this control with the presnt ±20% error. Control can be realized when this road gradient calculation method is used together with another parameter to judge the control situation.

抄録全体を表示

This paper focuses on the development of practical observers for semi-active suspensions. Many sensors are necessary to measure all the states used for a control system. Therefore, the estimation of states is valid for developing simple and low cost semi-active suspension systems. The nonlinearity, which is included in the absorber of semiactive suspensions, is a difficult problem in estimating accurate states using a usual linear observer theory such as the Kalman filter theory. To solve this problem, this paper presents a new practical observer which has two improvements. The first point is that the observer model includes a data map which reveals the nonlinear characteristic of the absorber. The second is that the observer feedback gain is scheduled. We design the feedback gain L₀ for a linear quarter car model whose damping ratio is minimum, and L₁, for that whose damping ratio is maximum. The feedback gain is determined using linear interpolation of L₀ and L₁ according to the control signal of the absorber valve. Applying this observer for actual systems, however, the problem that this observer needs more quantity of computation than the usual linear observer must be considered. Therefore, we reveal a method which reduces the quantity of the computation. To confirm the validity of this practical observer, simulations and experiments are carried out. These results show that the observer developed in this paper can estimate the suspension stroke velocity accurately using the vertical acceleration sensor on the unsprung mass.

抄録全体を表示

The main purpose of this work was to investigate the effectiveness of active suspension systems in the case that only rear suspension is controlled via a preview control method utilizing the future road surface roughness. The basic concept used here is as follows: The rear wheels of a vehicle will travel on the same road surface as that on which the front wheels have travelled. If the characteristics of the road surface on which the front wheels have travelled can be estimated, then these estimation results can be applied to control of the rear suspension. For design of an estimator which can estimate the characteristics of the road surface on which the front wheels have travelled, the concept of a disturbance observer is used. This approach has the advantage that the estimator can be designed based on a quarter-car model. Since the road surface roughness equivalent to Coulomb's frictional force is included in the estimation results, the influence of that frictional force on the control performance can be reduced. Also, it is easy to construct the feedback controller of the front suspension in order to improve the robustness for the error between the real road surface and the estimated one. From some experimental results, it was confirmed that the translational motion and the pitching motion can be suppressed with this preview control method.

抄録全体を表示

Four analytical models of a series-line resonator, i. e. a distributed parameter model (DPM), two types of combined parameter models and a lumped parameter model (LPM), are introduced. Their resonance frequencies are represented by three variables ζ, η and f_n1 (where ζand η are the length and cross section ratios of the body to the neck, and f_n1 is the fundamental resonance frequency of the neck with a closed-end). These are also evaluated by experiments. It is shown that the DPM is the best model and the available ranges of other models can be presented by ζ and η. For the pressure signal generator, the variable resonance frequency resonators are designed using the DPM, and their fundamental resonance frequencies are evaluated by experiments. The agreement between the analytical and experimental results is good. The systematic error of the acoustic resonance volumeter is also analytically and experimentally considered. The LPM modified by the DPM is the most suitable for solid volume measurement.

抄録全体を表示

We have developed a disturbance suppression method for the focusing control system of CD-ROM drives. Our servo system consists of a stabilizing compensator and a disturbance observer called the "multirate distrubance observer". The control system performs multirate sampling using a digital signal processor because this enables it to overcome the limitation in sampling rate. It makes the sampling rate of the disturbance observer longer than that of the stabilizing compensator. We confirmed that this control system can reduce the focusing error caused by disc vibration.

抄録全体を表示

In the development of sheet-handling machinery, it is important to be able to predict the behavior of the sheets with a high degree of reliability. For automated teller machines and letter sorting machines, in particular, a dynamic analysis of the deformation of each sheet is needed to predict the behavior of the sheets because the sheets are fed and stacked at such a high speed. This paper presents one method of predicting the dynamic behavior of sheets. First, a formula for the dynamic analysis is introduced using a spring-mass-beam model. Then, the calculated results of the sheet deformation due to stacking are presented. These results indicate that the proposed method can predict the dynamic deformation of a sheet.

抄録全体を表示

We propose three core methods for paper currency recognition using the neural network. The first method is to create a small size neuro-recognition technology for paper currency using masks. The second method is to adapt the genetic algorithm to the mask determination. The last method is to develop a high speed neuro-recognition board using a DSP to realize the commercial products. In this paper, we unify three methods for paper currency recognition and realize the neuro-system technology. We show development and effectiveness in an experimental system using current banking machines.

抄録全体を表示

A new type of hard disk system will be put into commercial use in the near future. One point of the system is to keep the contact force between the magnetic head and the moving disk as small as possible. The purpose of this work was to design a structure and controller of the contact-sliding head of magnetic recording systems. We have proposed a new method called a "integrated design", for simultaneous optimization of the parameters of the structure and controller. A numerical example for design of the control system of a magnetic recording system is given to illustrate the effectiveness of the proposed iterative algorithm.

抄録全体を表示

This study is concerned with the active control of a flexible arm using Q parameterization which is as one of form convex optimization. The nominal controller is the observer based controlier. It is designed based on an LQ servo strategy. Q parameterization can satisfy some constraints, for example, control input, overshoot, and settling time. However, Q parameterization becomes a higher-order compensator. It becomes from the 5th to the 40th-order in this study. The Q parameterization controller is also very robust.

抄録全体を表示

In this paper, the equations of motion of the non-contact hold and transfer control system of a spherical magnetic body using a one-pole-type magnetic robot hand system are derived. The robot hand system consists of a two-link robot arm and a one-pole-type magnetic hand. The Iinearized state equation and the linearized output equation of the magnetic robot hand, are derived and the controllability and observability are then clarihed. The non-contact hold control system by the magnetic robot hand is then constructed using H_∞ control based on the normalized coprime factorization method, and the tracking control system of the robot arm is constructed using the disturbance observer. It is confirmed from the experimental results that the non-contact hold and transfer control of the spherical magnetic body can be implemented well using the present control system.

抄録全体を表示

The purpose of this research is to construct a controller for traveling operations of mobile manipulators on irregular terrain. Several works on mobile manipulators have been published, in which the traveling plane was assumed to be flat. However, irregularities exist even on artificial floors in factories and buildings. When the end effector of the mobile manipulator must operate during travel on irregular terrain, it is subjected to the effects of disturbance torques derived from influences of traveling motion depending on the terrain. Then, use of compensation methods against the disturbance torques must be considered. Therefore, a mobile manipulator model traveling on irregular terrain is required to examine the control performance of the controller by computer. In this work, the dynamic model and the simulation method are proposed. First, the mobile manipulator model is formulated as a kind of space robot. Second, constraint conditions are considered for calculation of constraint forces to constrain the mobile manipulator on the terrain. Third, the method of calculation of the constraint forces with the constraint conditions under the influence of gravity is proposed. Then, the mobile manipulator model traveling on irregular terrain is formulated. Finally, the dynamics of the mobile manipulator is discussed through simulation.

抄録全体を表示

We describes a framework for the acquisition of perception-based navigational behaviors in autonomous mobile robots. Globally coupled chaotic system is applied to learn reactive action-rules. The control architecture of a robot consists of chaotic elements each coding an action-rule. The elements have the dynamics, which are designed so that the elements can collectively execute reinforcement learning. We carried out simulations on a navigation task in a static environment. Then, we observed the reproduction process and utility transition of the action-rules to examine how the robot acquire behaviors. Simulation results demonstrate that the robot successfully acquired behaviors such as goal-reaching, wall-following, and collision avoidance without any prior knowledge of the task space. The observation indicates that the stability of the acquired behavior depends on the types of the credit assignment and that the progress of the generalization of the state space correlates to the learning performance in a trial.

抄録全体を表示

A framework for the acquisition of perception-based navigational behaviors by autonomous mobile robots is presented. A globally coupled chaotic system is applied for learning of reactive action-rules. The control architecture of a robot consists of chaotic elements each encoding an action-rule. The elements have the dynamics, which are designed so that the elements can collectively execute reinforcement learning. The proposed method has been tested in simulations dealing with a navigation task in changing environments. Simulation results indicate that the robot acquired behaviors such as light-seeking and wall-following. The robot became able to move along a wall with its head swinging reciprocally so as to keep perceiving the wall on its left side. This strategy is flexible enough to enable the robot to search for, keep following, and turn along the wall. The process of behavior acquisition is also discussed from the point of view of credit assignment.

抄録全体を表示

This paper is concerned with a novel type of small 6 D. O. F. RTS fully parallel manipulator with elastic joints. The elastic joints are used to realize highly precise operation of the manipulater. Rotary input is applied to drive six groups of four-bar linkages in this mechanism. We establish two displacement equtations of the mechanism and present the inverse kinematic analysis including the motion of four-bar linkages. The bending and torsional deformations of elastic joints are further investigated from the kinematic viewpoint of the mechanism. Then we derive elastic moments and torques for the case that the motion can be described by the linear model in the elastic deformation range. Finally, we obtain an equivalent force-balance-equation of the 6 D. O. F. parallel mechanism where all the elastic moments and torques and all the input torques and loads of the upper platform are included.

抄録全体を表示

We describe the fatigue phenomena in bearings for automotive alternators. Characterics of this fatigue are as follows. (1) Flakings are concentrated in the outer rings on the pulley side. (2) Flaking occurs faster than the calculated bearing life. In order to analyze and investigate the effect of engine vibration and rigidity of the housing on the bearing life, fatigue tests were carried out using actual engines and alternator simulation test rigs. Eight different test conditions were used to derive the new equation which includes a new load factor. It was suggested that the new equation can be applied to engine accessory bearings.

抄録全体を表示

A general numerical method of calculation of the dynamic characteristics of fluid film lubrication in irregular bearing shapes and a high bearing number region is described. The method represents an expansion of the formulation of the divergence formulation method to the triangular element used in the finite element method. Weighted upstream schemes which are commonly used in the divergence formulation method on structured grid systems can be introduced in unstructured grid systems very easily. Numerical results for stiffness and damping coefficients of oil and gasjournal bearings show the usefulness and accuracy of the method.

抄録全体を表示

We investigated the lubricating characteristics between the vane and piston lubricated with an oil-refrigerant mixture. The theoretical analyses were performed using the mixed EHL theory, considering metallic contacts. Lubricating conditions were estimated by comparing the theoretical resuits of the piston dynamics based on actual surface roughness distributions measured by SEM, with the experimental results measured under a practical operation. We concludes that (1) lubricating conditions between the vane and piston are estimated to be within a mixed lubricating zone in which the coefficient of friction changes from 0.04 to 0.08, and the contact force is almost equally supported by oil film and metallic contacts, (2) the coefficient of friction decreases as the contact force increases because the ratio of pressure rise of the oil film to the decrease of the clearance is greater than that of the metallic contact pressure, and (3) it is important not only to measure the actual surface roughness but also to select the appropriate contact model in the mixed EHL analysis.

抄録全体を表示

The hobbing machine of recent years has positively introduced a computer numerical control (CNC) system, but many traditional hobbing machines with gear trains are still being used to produce gears. These traditional machines generally prepare many indexing change gears for gear cutting. But when we cut the gear of a prime number of teeth, the indexing change gear is not prepared for more than about 100 teeth of the gear. In such a case, we have to correct the indexing of the number of teeth using a differential gear to cut a helical gear, but when we apply the prime number of teeth on the helical gear, we cannot choose an arbitrary feed speed. Therefore, in this paper, we report on a new indexing mechanism for the prime number of gear teeth using a double differential gear device, which reduces the number of indexes the change gear must go through.

抄録全体を表示

This paper gives some ideas how to calculate tooth load distribution of one pair of mating gears with assembly errors. The main procedures are as follows. Firstly contact points surrounding the geometrical contact line on the tooth face of one gear are assumed. Secondly position changes of these contact points under a proposed special three-dimensional coordinate system are calculated by using three-dimensional coordinate transforming method when the gear shaft has assembly errors. Thirdly points on the tooth face of the other gear of the mating gears which to be considered to contact with the contact points on the region of the former gear are looked for through a so-called section method suggested in this paper. Finally clearances and influence coefficient of these pairs of contact points are calculated by the three-dimensional finite element method and tooth load distribution on these pairs of contact points is calculated by the mathematical programming method. The calculated tooth root strains and tooth contact pattern are well agreed with the measured ones.

抄録全体を表示

The present paper deals with the thermal behavior of the main spindle system of an ultraprecision machine tool as affected by a change in the temperature of the ambient air. First, it is found that the thermal deformation increases with an increase in the rotational speed of the spindle, which is considered to be caused by an increase in the heat transfer coefficient. The transfer functions between the ambient temperature variation and the thermal deformation of the machine are obtained experimentally for various rotational speeds, and the thermal behavior of the machine is analyzed. Second, a simple method of estimating the thermal deformation for future control of machining accuracy by utilizing the measured transfer functions and the convolution integral is proposed.

抄録全体を表示

An error compensation system for turning operations based on prediction of machining error is described. The machining error in longitudinal turning processes is mainly dependent on the deflection of the workpiece due to the cutting force and the thermal expansion of the cutting tool due to cutting heat. Furthermore the cutting force and cutting temperature vary with flank wear. The machining error can, therefore, be predicted with cutting force, cutting temperature and flank wear. A neural network which uses input information including cutting force and cutting temperature enables us to predict the machining error with distance from chuck to cutting point. The machining error is then compensated for by modification of the depth of cut at each cutting location. The compensation system is proved to be effective in the turning of cylindrical products.

抄録全体を表示

By repeating the stretch draw forming, the side wall thickness of a cup which is made of steel laminated with polyester film on both sides can be reduced significantly and this method has lead to the development of a container having a higher H/D (cup high/cup dia.) ratio. The purpose of this paper is clarifty the effect of die radius conditions on thickness reduction of the multi-stage stretch draw forming. From experimental results, higher side wall reduction can be obtained by using two-stage stretch draw forming. The available range of 2nd redraw expands by making the side wall thickness in the prior redraw stage as thin as possible. The optimal conditions for the highest conditions are to set the 1st redraw to have a Rd₁/to in the range of 1.4∼1.6 and the 2nd redraw a Rd₂/to of 1.1∼1.6. The stretch draw method is different from the conventional redraw method in terms of properties of plastic strain. The maximum wall thickness rate obtained in two-stage stretch draw forming is approximately 35% using high strength steel.

抄録全体を表示

The study deals with automatic polishing performed by a six-degree-of-freedom industrial robot equipped with two types of automatically selected rotational polishing tools. Little progress has been made in the autolmation of polishing since it requires much skill and experience. An industrial robot and a CAD system are introduced to cope with this problem. However, most products are so complicated that one type of polishing tool is not enough to provide a good finish. The aim of this study is to device a program that will enable the robot to automatically select polishing tools suitable for each workpiece surface and to polish it according to the type of tool. The system automatically selects the I-shaped (straight) rotational tool or the L-shaped one according to the surface shape of a workpiece using strategies to discriminate on the basis of CAD data. After the selection, the system generates a polishing path for both tools and performs the polishing. From experimental results, the system is found to be effective in polishing workpiece surfaces of complicated shapes.

抄録全体を表示