In this paper, we propose a design method of a disturbance-accommodating gain-scheduled control taking account of actuator saturation by seismic excitation in vibration isolation for a multi-degree-of-freedom structure. A disturbance model multiplied by an envelope function is introduced for distrubance-accommodating control in order to represent the nonstationary characteristics of seismic disturbance. We formulate a linear parameter varying system according to the amplitude of the disturbance as well as the controller output, and design a gain-scheduled controller. By carrying out simulations and experiments of seismic excitation, it is verified that the proposed controller can maintain the performance of isolation even if a large earthquake with which actuator saturation occurs comes.

View full abstract

This paper is concerned with a new method for avoiding the fictitious eigenfrequency problem in boundary element analysis of two-dimensional acoustic problems govened by Helmholtz equation. It is well known that in solving the external acoustic problem by means of the boundary integral equation, the solution is violated by the fictitious eigenfrequencies of the problem, which correspond to those of the internal problem. The present paper proposes a new method to circumvent the fictitious eigenfrequency problem by using the dual boundary integral equation for nodal points on the boundary. The quadratic boundary element is employed in this study. The usual boundary integral equation is applied to the end points of boundary element connected to other elements, while the derivaive boundary integral equation is applied to the other internal point of element. The proposed approach is implemented, and its effectiveness is demonstrated through comparison of the numerical results obtained by the developed computer code with other solutions.

View full abstract

This paper describes a pointing control of a mobile antenna mounted on a vehicle. For the precise pointing, the antenna must be controlled to point to a target under the influence of continuous disturbances acting on the vehicle. To the problem, we propose an optimal controller design method in a framework of the H_∞ gain-scheduled controller synthesis. By the synthesis, the optimal controller is obtained by using parameter varying notch-filter in order to attenuate the various frequency disturbances. Finally, we apply the design method to an automobile model having rotating antenna.

View full abstract

This research develops the phenomena of chaos and its stabilizing control in following a circular trajectory by considering a time-delay of pneumatic signals in a 2-link pneumatic manipulator. In general, it becomes difficult for such a manipulator to follow a desired trajectory, because chaos phenomena is often caused in manipulators with time-delay of pneumatic signals transmitted through long air tubes. In order to solve this kind of problem, the chaotic behavior is examined by recording the Lyapunov exponent, and a new stabilizing method with neural networks (NN) is proposed. Through simulations and experiments, the phenomena of chaos and the control of stabilization are verified, and the effectiveness of the proposed method is shown.

View full abstract

An integrated design of structure/controller for parallel inverted pendulum systems is presented. In practical parallel inverted pendulum systems, existence of a realizable stabilizing controller is inevitably determined depending on the position of center of gravity of two pendulums. In this paper these parameters are set as structural parameters in structural systems and a descriptor form representation is used to express the dynamics of parallel inverted pendulum systems. A state feedback gain and structural parameters are determined based on a design of linear quadratic regulator (LQR), in which Ricatti equation in LQR problem is reduced to LMI conditions and minimization problem is solved to obtain optimal structural papameters. Main contribution of this paper is to give a method to determine the position of center of gravity of two pendulums for parallel inverted pendulum systems in a sense of suitable for control. Based on the obtained structural parameters, some experimental works are executed. Experimental results show the effectiveness of the proposed design method.

View full abstract

This paper presents a design approach for the synchronization problem of two linear systems. A feature of the presented approach is that the closed loop system is consisting of tow parallel servo systems with a synchronization compensator, and the synchronization compensator is constructed by combining the disturbance observer with direct feedback of the synchronization error. In order to demonstrate validity of the proposed controller, two X-Y stages widely used in the manufacturing systems are exploited for experiment. The experimental results with the comparision to the exsisting control approach shown that a spectacular synchronized performance can be achieved by the proposed controller.

View full abstract

In emergency transportation by ambulance, a patient receives acceleration in various direction. Particularly, longitudinal or lateral acceleration in braking and curve driving causes patient's stress such as blood pressure variation in brain, vomiting or feeling of discomfort. To reduce such patient's stress, we developed the 2-degree-of-freedom (2 DOF) actively-controlled bed (ACB) that reduces the acceleration acting on a patient by controlling posture of the bed. This paper reports its structure, control scheme and control performance. The effect is examined by experiment in terms of cancellation rate of acceleration and ride quality. Using the actual driving data of ambulance, the cancellation rate of acceleration in emergency transportation is also estimated by simulation.

View full abstract

This paper describes visual tracking device progressed to measure faster motion of a camera perpendicularly directed to floor. In the device, motion prediction based on extrapolation of the velocity vector is employed and extends the maximum speed of motion from about 45mm/s to 200 mm/s with no increase of search time in the correlation process. Moreover crash detection is brought to this method as a side effect of motion prediction by detecting the change of differential correlation. The device consisted of a FPGA and two SRAMs can process image tracking at frame rate, so this device is useful for localization of mobile robots.

View full abstract

Recently, we need a human friendly robot which can provide human several services. When such a human friendly robot interacts directly with human, it requires flexibility, safety and so on. In order to realize such a flexibility, there is a way to utilize a flexible material and air compressibility. Therefore, it is considered that the pneumatic soft actuator with passive compliance is useful. The purpose of our study is to develop the pneumatic soft actuator in order to realize the robot with flexibility. We have developed a pneumatic bending soft actuator which is made of silicone rubber. We call it soft gripper. The soft gripper can be applied to a finger of robot hand, artificial hand and so on. In this paper, first, the structure, operation, principle and fundamental characteristics of soft gripper are shown. Next, the proposed analytical model of soft gripper is described and it is confirmed that the proposed model is useful for designing the gripper by comparing the calculated results with the experimental results. Finally, as an application of the soft actuator, a rubber robot hand with grippers is introduced.

View full abstract

To obtain specifications for an effective tactile display in virtual reality and tele-existence systems, we developed a matrix type experimental tactile display. It has an 8 by 8 stimulus pin array of 0.3 mm in diameter and with a 1 or 1.8 mm pin pitch. Operators can feel various combinations of pins protruding according to concave and convex portions of virtual textures composed of crossed-axis patterns. We evaluated the matrix type tactile display with two psychophysical experimental methods. In the adjustment method, human subjects adjusted the crossing angle of a comparison texture to bring it as close as possible to a standard texture fixed during the experiment to obtain crossed-axis angle thresholds from the difference between the crossed-axis angles of comparison and standard textures. In the constant stimuli method, many random pairs of comparison and standard textures were presented, and human subjects judged whether the crossed-axis angle of the standard texture coincided with the comparison texture to obtain thresholds with the statistical procedure of the judgments. After seven human subjects performed the abovementioned adjustment and constant stimuli methods, we found that the threshold of the adjustment method was almost the same as that of the constant stimuli method, but the total time consumption of the constant stimuli was about four times larger than the adjustment method.

View full abstract

This paper proposes a method for planning the gait of a humanoid robot in real-time for the force controlled manipulation. Focusing on.the task of pushing an object, a humanoid robot changes the foot placement in real-time according to the amount of pushing an object. By using this algorithm, a humanoid robot can push an object stably regardless of the mass of an object. If an object is heavy, a humanoid robot pushes an object with walking slowly, and vice versa. Also, for planning the gait in real-time, we newly propose an analytical method where the newly calculated trajectory of the robot motion is smoothly connected to the current one. The effectiveness of the proposed method is confirmed by simulation and experiment.

View full abstract

TSP is one of the well-known problems belonging to NP complete problem and it has a lot of applications in the field of engineering. This study presents a new method named LCO for solving a large scale of TSP. The principle of the new method is based on Riccati-type learning equation proposed by T. Kohonen. T. Kohonen developed Self-organizing Map (SOM) theoretically based on the learning equation and SOM is applied to solve TSP, efficiently. However, SOM can only treat TSP, where city costs are represented by Cartesian coordinate system. LCO can solve a general type of TSP, where only city costs are given. Also, LCO can treat much more cities than SOM can. LCO is compared with GA and SOM in numerical experiments and it is proven that LCO solves TSP with high accuracy and fast computation.

View full abstract

As the output of DC motors for electrical power steering becomes higher, noise reduction and temperature rise reduction are increasingly demanded. We previously proposed the short circuit wires is symmetrical position and realized a noise reduction, but the characteristics of the motor are affected by the resistance of the short circuit wires. Therefore, we then examine the relationship between the resistance of the short circuit wires and the brush current and the winding current in detail. As a result, it was proven that the level of the exciting force is large when the resistance is large, and that the maximum value of the brush temperature rise is large when the resistance is small. And we propose a brush current control method which minimizes the maximum value of the brush temperature rise. As a result, by controlling the electric power consumption of a homopolar brush equally, we can decrease the maximum value of the brush temperature rise.

View full abstract

In this paper, we describe about the measurement technique of vehicle sideslip angle using stereovision system. The vehicle sideslip angle is one of the vehicle state quantities. Moreover, this parameter is very important for control system of vehicles such as the dynamics stability control system. We propose measurement technique for sideslip angle using stereovision system. This technique uses the disparity and the optical flow to measure the relative movement of the vehicle. Moreover, it only uses two cameras, PC and the rate gyro. So, this measurement system can be realized inexpensive. Additionally, we propose using the predicted image based on vehicle motion model, in optical flow analysis. The result of the optical flow analysis has been improved by proposal method. It is confirmed that proposed technique can measure the movement distance and the sideslip angle by the experiment.

View full abstract

The results of vehicle performance tests often vary depending on differences in the road surface temperature. It is believed that these changes can partially be attributed to the effect of tire surface temperature. The aim of this study is to develop a tire force model considering the dependence of tire surface temperature. The newly developed tire model is composed of two functions, the thermodynamic function that allows us to consider changes in the tire surface temperature and the force function that allows for the effects of tire surface temperature. The parameters of this model are identified using the measured data obtained with an indoor test facility. To prove the validity of the developed model, the surface temperature and the tire force values are predicted using this model. The simulation results agreed rather well with the experimental results.

View full abstract

Wheelchair users need lifts to climb up/down steps at entrances with small spaces. Lifts driven by motors or hydraulic equipments are large and expensive. They also need switches to start/stop actuators. The aim of our study is to develop a compact and non-powered lift for wheelchair users. We propose a lift driven by wheels of a wheelchair on it. Coil springs are built in the lift to decrease driving torque and some mechanism are attached so that small front caters can pass through the lift stage and large rear wheels can drive the lift. After the lifting speed, the driving force, and the condition to drive rollers were analyzed, a prototype lift was designed and made. The developed lift succeeded to raise a wheelchair with a user and continuous motion of a wheelchair from going into the stage to going out of it was executed smoothly.

View full abstract

This paper proposes an algorithm for detecting objects representing potential hazards to drivers, based on the combination of local information derived from optical flows and global information obtained from the host vehicle's status. The algorithm uses artificial neural networks to infer the degree of danger posed by moving objects in dynamic images taken with a vehicle-mounted camera. This approach allows more flexible adaptation of the algorithm to many drivers with dissimilar characteristics. Experiments were conducted with a model and real vehicles using video images of multiple moving objects. The results show that the algorithm can infer hazardous situations similar to the judgments made by human drivers. The proposed algorithm provides the foundation for constructing a practical driving assistance system.

View full abstract

Weightless environments such as bedrest and spaceflight are associated with profound losses of muscle bulk and strength which hinder patient's or astronaut's health and reintegration into daily life. To prevent the losses, we developed a practical and effective training method with electrical stimulation, named “hybrid exercise”; electrical stimulations are added to antagonists at joints in order to produce resistance forces in one' s own body. The environment was simulated to evaluate the effects of the hybrid exercise on human body. An unilateral lower limb suspension (ULLS) was introduced as a simulation model of the weightless environment. Two kinds of experiments have been done : (1, ULLS experiment) only ULLS was executed to make disuse of lower limb. : (2, ULLS-Hybrid experiment) ULLS combined with the hybrid exercise to prevent losses of lower limb functions. We evaluated experiments by four items, maximum volitional contraction (MVC), cross-sectional area (CSA), bone mineral density (BMD), and electromyogram (EMG). The ULLS experiment showed dramatic losses of these parameters, whereas the ULLS-Hybrid experiment showed good effects of the hybrid exercise to prevent the losses.

View full abstract

A novel biochip is developed for culturing stem cells. Biochip is made of Polymer (PDMS), and cells can be loaded by gradient strains in one chip. They grow well on a hydrophilic membrane and differentiation is promoted by cyclic strains. In this paper, we propose the method for culturing and monitoring of stem cells such as bone marrow stromal cells (ST 2 cells) and myoblasts (C 2 C 12 cells), and the results of culture. First we analyzed strains on a membrane when an air hole is decompressed, and clarified their range. From experiment, bone marrow stromal cells grew well in a narrow range, and we quantified their ALP activity as a measure of differentiation. As myoblasts, the direction of their differentiation was perpendicular to a groove, that is, the same direction of uniaxial strains.

View full abstract

This paper is concerned with the development of a wireless electric stethoscope system and cardiac sound analytical method for cardiovascular disease diagnosis. This system consists of an electronic stethoscope, data acquiring amplifier/filter circuits, wireless transmitter and receiver modules, and a notebook computer. The cardiac sound acquired by the electronic stethoscope and can be easily transmitted to the computer by wireless transmitter and receiver modules. The input-output characteristics of the electronic stethoscope and wireless modules are investigated in detail. Furthermore, one-DOF analytical model based on the consideration of the eardrum property is proposed for abnormal cardiac sound analysis. With the analytical model the original heart sound signal can be converted to a simple characteristic waveform which fits the behaviors of some cardiovascular diseases. Some case studies using the characteristic waveforms show that the normal sound can be distinguished visually from the abnormal cases such as the mitral regurgitation, aortic incompetence, allorhythmic pulse and so on.

View full abstract

The thickness of a lubricant is 1 to 2 nm for a current magnetic disk, and this means that the lubricant is a monolayer or sub-monolayer. The two-layer model has been widely accepted for rather thick lubricants. In that model, a lubricant film consists of the first layer bonded to the disk and the mobile layer on the bonded layer. However, the conventional two layer model is not applicable to a current monolayer lubricant film for a magnetic disk. In this paper, based on the investigation of the molecular conformation and spreading characteristics, we clarify the diffusive motion of mobile molecules in order to build a tribological model for a monolayer lubricant.

View full abstract

Slide-screw (SCR) test using resin nuts was conducted together with Pin-on-disk (POD) test for the same resin materials, in order to know the relations of friction coefficient and specific wear rate of resin materials for SCR between both of tests. The results showed that the friction coefficients were almost the same (-0.27) in both of the tests, while the specific wear rate tended to become larger in SCR test than in POD test. The larger specific wear rate in SCR tests was ascribed to the longer contact length (the length of contact area along sliding direction), which was 550 mm between the screw and resin nuts and 3 mm between the disk and resin pins. In practice, the specific wear rate was decreased in the SCR test using the resin nut with 18 mm contact length, which was shortened by cutting the screw thread away over 1.5 mm width at every 180 degree seen from the axial direction.

View full abstract

It is well known that a bolt is apt to loosen by slipping between the contact surfaces of joined parts. A loosening test using real parts involves precisely expressed real phenomena which are under the influence of multiple factors such as slip distance, surface roughness/waviness, and coefficient of friction, etc. However, estimating the influence of individual factors is very difficult because the behavior of real contact surfaces correlates to the variation of friction coefficient. Therefore, it is required to apply simulations to control the coefficient of friction. In this study, FE analyses were carried out with various levels of certain factors, and the self-loosening of a bolt was investigated. As a result, the calculations revealed that slip distances and coefficients of friction strongly influence the self-loosening of a bolt, but roughness/waviness of the pressure flank of threads had hardly effect. These tendencies coincide with the real nut's phenomena that adhesive is effective, and roughness/waviness on the surface is not active for prevention of loosening.

View full abstract

In the high stiffness hydrostatic thrust bearing in which the land supported elastically could displace instead of shaft movement against load fluctuation, a new elastic supporting construction without friction while land displacement is proposed. With regard to the static performance, this bearing shows infinite stiffness within the load carrying capacity of 0.4 to 0.9 under the optimum condition and this agrees well with the calculated results. Meanwhile, the self-excited vibration takes place sometimes in this bearing for the lowering of damping capacity. The stable region for the vibration is examined and the factors which influence it are clarified. Then in order to suppress the self-excited vibration, a squeeze film damper which does not affect the static performance is installed inside the bearing body. By this damper, stable operation under the optimum condition becomes possible with the proper matching of the squeeze clearance and oil viscosity or oil temperature. The effect of the damper is also confirmed by the stability analysis.

View full abstract

This paper shows a thermohydrodynamic lubrication (THL) analysis method to simulate slider bearings with steps on the bearing surface such as Rayleigh step bearing. To satisfy the flow continuity on the step region, virtual clearance and its derivatives are defined on the edges. This method allows the finite differential equations (Reynolds equation and energy equation) to be solved at both continuous and dicontinuous clearance region. The simulation result is in good agreement with the CFD calculation result. This paper also shows a simulation result of a finite width Rayleigh step bearing with side dams.

View full abstract

When gear pairs have the shaft angle 90°, the radius of gear 95mm and the inclination angles of straight paths of contact given by a pair of 20° (drive side) and-20° (coast side) on the normal plane S_n which is perpendicular to the relative velocity at P₀, variations of the characteristics of the design point P₀ and its contact conditions (the radii and the centers of curvature at P₀) of tooth surfaces caused by those of gear ratio i₀, offset E and spiral angle ψ₀ of bevel and hypoid gears are discussed through the numerical examples to examine how to choose E and ψ₀ according to the given i₀, which is summed up as follows. (1) Variations of the condition of contact are greatly influenced by i₀ and ψ₀, but not so much by E. Therefore, E should be chosen mainly to design the adequate diameter of mating pinion. (2) In the case of small i₀ (for example, i₀=1.0), E should be chosen smaller to avoid too large diameter of pinion, which gives convex-convex contact on both drive and coast Surfaces at P₀ in the ordinary range of ψ₀ (<40°). (3) In the case of large io (for example, io=20), both E andψ₀ must be chosen larger enough to realize sufficient diameter of pinion, where the drive surfaces have convex-convex contact while the coast surfaces have convex-concave contact. (4) In the ordinary case of i₀=4.1, the bevel gears with ψ₀=35°and the hypoid gears with E=35mm and ψ₀=50°have convex-convex contact on the drive surfaces and have convex-concave contact on the coast surfaces. The difficulty of designing bevel and hypoid gears is in those unsimilarities of the conditions of contact due to i₀, E and ψ₀ mentioned above, so that E and ψ₀ must be chosen individually according to i₀.

View full abstract

A systematic scheduling method is required to modify the predetermined production schedules in order to cope with unforeseen changes in manufacturing systems. The objective of the present research is to propose a scheduling method based on GA (Genetic Algorithm) for modifying delayed production schedules due to the additional jobs, reactively. A prototype of reactive scheduling system is developed and applied to scheduling problems. The effectiveness of the proposed method is verified through the numerical expreriments of the reactive scheduling problems.

View full abstract

This paper shows a new safety strategy of the hazardous point nearby operation such as the process confirmation, adjustment, trouble shooting, maintenance, repairing or cleaning of the machine. About 40% of fatal accidents by the machine is caused by such operation because the operator approaches to the machine without stopping hazardous moving parts. New concepts such as the risk manegement division (I : safe condition, II : uncertain condition, III : hazardous condition), the accident prevention division (0 : restriction of speed and force, 1 : fixation of human and machine space, 2 : machine stop, 3 : hazardous point nearby operation) and the supporting protective device (TYPEA : priodic check for the protective device, TYPEB : priodic training for the operator) were proposed. These results can be used as a complement of ISO 12100 standard.

View full abstract

This paper proposes a new selective area exposure method for stereolithography using liquid photoinitiator and base resin (photopolymer resin without photo-initiator) separately. First, base resin is supplied as a layer, and then a mask pattern is drawn onto the surface with photo-initiator using inkjet printing. When the surface is exposed with a UV lamp, only the drawn pattern is cured by the photo-initiator. In this process, the photo-initiator works as a positive mask and the uncured base resin can be used again because it is not mixed with the photo-initiator. The merits of this method are : (a) an area exposure instead of dot scanning enables fast fabrication, (b) no use of a laser or a mask device such as LCD leads to low cost, (c) keeping the base resin and photoinitiator separately leads to longer pot life of materials, and (d) low viscosity and no curability of liquid photo-initiator itself enable more stability and liability of inkjet process. The basic idea, fabrication system implementation and the results of some fabrication experiments are presented.

View full abstract

This study deals with automation of metal hammering working on the basis of CAD data as a forging type rapid prototyping system. In this article, in order to analyze the phenomenon in metal hammering working, a CAM system for metal hammering is developed and experiments of incremental metal hammering based on the system are examined under several conditions. From the experimental result, influences of the mode of a tool path on the deformation behavior, a possibility to improve a forming limit by distribution control of strain on a workpiece and a capability of the system to perform the forging type rapid prototyping are found.

View full abstract

The dificiency in the variety of available magnetic abrasives results in a narrow range of finishing performance of the Magnetic abrasive finishing process. To break through this situation, this study develops a Ni-based flaky magnetic polishing tools, which consists of thin layers manufactured using an electroforming technique, and which is made without conventional abrasive. Because of the flaky structure, the edges broken during the finishing process effectively renew the cutting edges. It has an easy axis of magnetization and is magnetized such the flat surface is normal to the workpiece surface. This localizes the contact points of the magnetic polishing tools against the peaks of the uneven surface of the workpiece. While existing magnetic abrasive removes material following the overall form of the surface, the Ni-based magnetic polishing tools remove the material from the peaks of the surface. Accordingly, the Ni-based magnetic polishing tools achieve precision finishing of SUS 304 stainless steel tubes despite having lower material removal than the existing magnetic abrasive.

View full abstract