There are three bone graft methods for bone defects caused by diseases such as cancer and accident injuries: Autogenous bone grafts, Allografts and Artificial bone grafts. In this study, meshed pure titanium plates with high biocompatibility are introduced to solve the over stiffness and weight problems of ready-used metal implants. Basic mesh shapes are designed and applied to pure titanium plates using 3D CAD modeling tools. Bending properties of prototype meshed pure titanium plates are evaluated experimentally and analytically. Sample meshed plate specimens with 60°, 90° axis-symmetrical mesh shape types were fabricated for 3-point bending tests. The pseudo bending elastic moduli of meshed plate specimens obtained from 3-point bending test are ranged from 4.92 GPa to 10.15 GPa, within the elasticity range of natural cortical bones from 2.0 GPa to 30.0 GPa. Analytical approach method is validated with high accuracy by comparison with experimental and analytical results for evaluation on bending property of meshed plates. Using the highly accurate analytical method, the influence of various design variables on bending characteristics of pure titanium meshed plates are grasped and reported in this study.

In this research, we aimed to explore the mechanism of carbon filled PTFE forming a carbon film on the sliding surface and reducing the friction. We evaluated the influence of water content on friction behavior using a pin-on-disk tester, and examined the mechanism by surface analysis. Pitch type carbon filled PTFE showed stable and lower friction than PAN type carbon filled PTFE. It was found that there is a difference in the influence of moisture between PAN type carbon filled PTFE and Pitch type carbon filled PTFE, and that effect is small in Pitch type carbon filled PTFE.

This study focuses on the acute influence of an AC electromagnetic field (EMF) exposure (50 Hz, Bmax 180 mT) on flow-mediated dilation (FMD) and cutaneous blood flow volume in healthy human subjects. In a randomized, double blind and crossover design, the sham control (CTL) and the EMF exposures were carried out. For FMD study, exposure of the left upper arm to EMF was performed for 30 min in a supine position. In the case of the measurement of microcirculation, exposure of the left forearm to EMF was conducted for 15 min in a sitting position. The FMD values in the CTL exposure were not significantly changed compared with the baseline value (indicated as 100%). The FMD values were significantly increased from the baseline value in the presence of EMF exposure. The values of blood flow volume in the CTL exposure were significantly reduced compared with the baseline value probably due to immobility stress-induced ischemic conditions by preventing the arm from moving. In contrast, the values of the blood flow volume were not significantly changed by the EMF exposure. There was significant difference between CTL and EMF exposures. When considering the physiological significance of the non-thermal EMF effects, these results imply that the EMF-enhanced vasodilation and microcirculation might help eliminate the metabolic waste products and endogenous pain producing substances inducing muscle stiffness and pain.

This study aims at developing multi-class clustering method of damage/fracture modes using Support vector machine (SVM).Conventional failure mode analyses methods lack the method how the deviation in design specification associate with specific types of failure modes. In our proposed design deviation method, an association table between design/ environmental deviation with the types of damage/ fracture modes has been made. The association table can help practitioners to predict as many as possible failure modes in components caused by damage/ fracture modes of materials. Multi-class clustering using SVM demonstrated that the precision values of each damage/ fracture modes were 77 -100 %. The result demonstrated that the proposed multi-class clustering method can support in selecting specific damage / fracture mode.

Light emitting diodes (LEDs) have been widely used for the luminescent text display in automobiles. The purpose of the current study is to evaluate fatigue reduction characteristics in LEDs luminescent text display. For this purpose, we construct a model to estimate the participant's fatigue from physiological index. In the experiment, participants read Japanese sentences that were represented by backlight emitting for ten minutes. Brain activity were measured for the physiological index, and fatigue evaluation questionnaire was answered. The luminescent text display was implemented with LEDs of six kinds of white color. We constructed the deep learning model that inputs brain activity, and outputs “fatigue”, “no fatigue” or “no change”. The output was calculated from fatigue evaluation questionnaire. The model was able to estimate the fatigue with a validation accuracy of 86.4 %. Thus, the model showed the possibility that estimate fatigue based on the brain activity.

In recent years, an increase in space debris on orbits around the Earth is a problem for safe and reliable utilization of space activities, and removal of space debris is required. In this paper we assume capture of large space debris by robot arm. Since attitude control is not working, most space debris is rotating. For this reason in capture of the space debris, there is a complicated dynamic interaction between the capture target and the robot arm, and there is a possibility that big load may occur to the robot arm. As a buffering method of big load generated at the time of capturing space debris, we proposed a buffering method using impedance control and a flexible mechanism with physical softness, and confirmed the effectiveness by simulation. We also designed and manufactured a full-size prototype of robot arm. We made a simulation model in one joint and compared with prototype.

Since the number of space-debris (hereinafter, referred to as “debris”) in earth orbit is increasing year after year. High risk of collision that debris and operational satellites. Peculiarly, the density of low orbits is high. Debris fall into atmosphere and disappears naturally. However it requires a lot of time. Orbits of altitude of 800 or higher, it'll be necessary for hundreds of years. Each research institution reported even when new satellites aren't launched, the number of debris automatically increases due to collision and crushing of each debris (Kessler syndrome state). If the number of debris continue to increase, it is evident that will a serious impact on future space development. Therefore, it is urgent to actively remove debris from onorbit. The removal method of dropping debris to the Earth, in the same as natural extinction is promising. As above, require deceleration to drop debris. However, propulsion system of the debris is not working. For this reason, mounting propulsion device is necessary. In this study, we assume that the upper of the H-series rocket will be the removing target and capturing it using robot arm. In this paper, results of examining capture strategy and a trial product of gripping mechanism.

In the present study, a method to scan the front side, the lateral side and the back side of an object by a two-dimensional laser range scanner and mirrors is described. Distances and angles from the scanner to the object at a constant angle are measured. Laser reflecting points on the surface of the object at a constant angle are gotten. An experimental equipment was designed to evaluate a performance of the scan method. The experimental equipment was built using an actual scanner and mirrors. Experimental methodologies using the experimental equipment were set. Experimental results using the experimental methodologies were considered.

In the present study, a method to scan an object in the blind area of a two-dimensional laser range scanner by mirrors is described. Distances and angles from the scanner and the mirrors to the object at a constant angle are measured omnidirectionally including the blind area. Laser reflecting points on the surface of the object at a constant angle are gotten. An experimental equipment was designed to evaluate a performance of the scan method. The experimental equipment was built using an actual scanner, an obstruction and mirrors. Experimental methodologies using the experimental equipment were set. Experimental results using the experimental methodologies were considered.

A ranging and vibrator module and a sensing vest composed of the module were made. A condition of contact with a vibration plane and the skin surface was checked by a person in the made sensing vest. The frequency of the vibration was measured using the person in the made sensing vest. The vibration or non-vibration of the module was checked at an object behind the person by the person in the made sensing vest. The results have shown the usefulness in the sensing vest for feeling the presence or absence of the object behind the person in the normal direction of the skin surface by haptic sense of back.

This paper presents a guidance control method that avoids large obstacles without collision of multiple mobile robots with other robots and obstacles. In this research we have conducted a lead guidance by the Leader-Follower method. A plurality of sonar sensors is incorporated in the robot, and the formation shape is controlled using only the distance information by the sonar sensor. In addition, when it is difficult to avoid formation while keeping the formation at the time of obstacle avoidance, we introduced an algorithm to change the ranks of robot groups to make it easier to avoid. In this case, in order to realize a method of temporarily changing the follower to follow and returning to the original shape again after avoiding the obstacle, an algorithm for estimating the position of the reader machine so as not to lose sight of the robot as the reader machine is introduced. We demonstrate the effectiveness of the proposed method by actual machine experiment using three robots.

This paper presents a method to grasp deformable objects with position-controlled robot hands based on caging. We derive a sufficient condition for caging-based grasping of two rigid objects connected by a string, as an example of deformable objects. We also verify the condition by experiments with an industrial robot. This shows an example of geometry-based manipulation, which requires no mechanical information.

It's possible to measure static load using piezoelectric ceramics by sticking 2 pieces of identical ceramics together. When one of the ceramics is oscillated at its resonant frequency, the other ceramics generates voltage by following up to the motion of the actuator ceramics. The vibration of the body is suppressed due to the impressed load, and then the sensor voltage drops according to the magnitude of the load. Vibration analysis proved that the configuration of the vibration of the sensor body is simple 1^st mode bending of a thin circular plate. It's also turned out that the characteristics of the sensor voltage are dominated not by the magnitude of the load but by the contact area of the load with the ceramics. Therefore, introducing the cushion material, which alters the contact area according to the magnitude of the load, will improve the sensitivity and the linearity of the sensor voltage to the load.

In the existing master-slave system, the design of the control parameters is performed in the heuristics, and veteran designers and workers are needed. In this paper, we aimed to realize the ideal behavior for each person more easily, devised a method incorporating DeepQ-Network (DQN) in the virtual internal model. As a result of verifying the method using simulation, we could not realize master - slave control. but could show the learning possibility of the model.

This paper describes the mathematical modeling for jumping motion of a one-legged jumping robot that consist of a hip, a knee and an ankle. The jumping robot has six mono-articular muscles mounted antagonistically in pairs across two joint and three bi-articular muscles mounted across each joint. Moreover, the vertical jumping of the robot has three phases such as a stance phase, a lift-off phase, and a flight phase. It is difficult to obtain the output of each actuator because each joint torque is produced by plural pneumatic actuators. Thus, a method converting the reference torques into the output of each actuator is developed. Vertical jumping simulations are conducted, and the effectiveness of the mathematical modeling is verified.

In general, CNN (Convolutional Neural Networks) is used as the method with high recognition accuracy. In CNN, however, several tens of thousand images are required as learning data for each category. Also, huge learning time is required and the reason that misrecognized is not explained. In contrast, after a human just look at several objects in a category he can get something like its general object concept. Furthermore, a human can represent the concept by words. Also, a human can explain the reason that he chooses an object as an identification object. In this article, a new concept learning method based on clustering and logistic regression using object concept based on color, shape and size is proposed, which requires low dimensional multi features, small training data and short learning time. Generated object concept was evaluated in comparison with the result of recognition using real world objects included in RGB-D Object Dataset.

In recent years, autonomous mobile robots are expected to be useful in outdoor environments. However, it is difficult for the robots to respond to changes in the outdoor environment, which is an important problem. For an autonomous mobile robot, it is important which navigation system to use. In recent years, simultaneous localization and mapping (SLAM) is often used. This method has good performance if accurate maps is used, but it takes time and effort to create accurate maps. However, in case of human being, such accurate maps are not necessary. Our previous study of human environmental recognition ability demonstrated that human uses scenery and road information. In this study, we propose an autonomous mobile navigation method with image processing based on human environmental recognition ability. This method uses GIST feature for localization, and SegNet for road detection. We validated our method in outdoor using a mobile robot. Experimental results demonstrate that the proposed method enables autonomous mobile navigation without accurate map.

We have developed the human friendly robot that consists of two arms, a trunk, a waist, an omnidirectional mobile base. Each arm of the robot has 2 DOF shoulder, 1 DOF elbow and 2 DOF hand, and its waist has 3 DOF. Its height is 1.42[m], its arm length is 0.76[m], and its total weight is 82.4 [kg]. In this paper, a new passive collision force suppression mechanism is described, which consists of two release air bags, a clutch gear, a drum, and four compression springs. Also, a new air cushion bag is developed to cover the entire outside of the robot. When the collision occurs between the robot and an object, the joints of the arm and the waist are disconnected by the collision force suppression mechanism depending on the magnitude of collision forces. Collision experiments are conducted, and the effectiveness of the collision force suppression mechanism is verified.

Due to the high development of artificial technology, mechanization of robotic technology and automation of household chores or building maintenance, our living quality has been dramatically improved. It is expected that introduction of autonomous robots to indoor is more rapidly proceed than ever by this trend. In this work, we propose the integration of two of the most widely used approaches for the implementation of autonomous navigation systems: the reinforcement learning for path finding, along with SLAM (Simultaneous Localization and Mapping). These two approaches are integrated to address the problem of how a robot should explore a complex environment while it collects perception features in order to locate itself and, at the same time, to obtain information clues about cost traversability of an area. Therefore, this research is a new suggestion for improving the efficiency of current clearing robots.

Research aimed at energy conservation was developed as a single device. Although research on mobile robots has been done a lot, research and development focusing on energy conservation has not been done much. If it is an autonomous mobile robot, energy saving can be realized in the exercise plan. Therefore, in order to make energy saving of the mobile robot possible, we have designed and manufactured hardware capable of mechanical and electrical energy saving operation. In addition, energy saving can be realized by combining exercise planning, for example, dedicated software based on navigation and exercise control. In this paper, we calculate the efficiency using a simple DC motor model and estimated the efficiency of the conventional independent drive steering module and the differential drive steering module. As a simple estimate, comparison was made between the conventional independent steering module and the differential drive steering module using the proposed differential mechanism. The results are reported because the possibility of improving efficiency is suggested for the differential driven steering module using the proposed differential mechanism.

In this study, the tracking controller of a mobile robot is designed by using the method based on the time state control form and the image processing techniques. The proposed controller uses the relative position and angle of the robot with respect to the target trajectory which are computed from the images of a camera mounted on the robot. In the experiments, the trajectories of the robot were investigated when changing the control gains. The large gains in the rotational control input make the robot to turn rapidly and to lose sight of the target trajectory. Therefore, we propose to regulate the rotational velocity input automatically by adjusting the translational speed such that the rotational velocity is kept within the limit. By using the self-adjustment of the speed, the robot tracked on the target trajectory without missing the target.

The purpose of this study is to develop a pig house cleaning robot system that is driven by a pulley and a wire to extend the arm length. The developed cleaning robot consists of an articulated arm that is constituted by a telescopic arm mechanism. The experiment performed a comprehensive evaluation of two input devices including the proposed mobile device.

We developed a tele-operated mobile robot with a high expansion camera arm using convex expansion and contraction mechanism. A webcam which is installed to the top of the robot gets the information abound the robot and sends the camera images via the internet constantly. The robot is controlled by a game pad using RSNP system and RT Middleware remotely. Here, we newly developed the expansion system to get a wide range view using the tilt angle controlling which adjusts each length of convex mechanism and verified the system by basin experiment of tele-operated mobile robot.

In this paper, we propose the method of grasping "soft" objects with 3 fingered-Robotic hand. In order to realize this, we newly developed fingertip sized Young's modulus sensor. Measuring pressure on robotic hand fingertip with the developed sensor and using expanded Hertz contact theory make it possible to estimate the Young's modulus of various softness objects. Finally, we indicate about effectiveness of proposed method through the verification experiment.

In the previous study, we developed a worker assistance system to support the assembling work on the desk. This system projects the support information on a desk with a projector, and it shows the movement of a worker captured by Kinect V2 on a display. This system supports only a person. However, in the factory, assembling machines sometimes requires multiple people. Therefore, in this study, we will develop a new system that can provide support information for large scale tasks performed by multiple people. In the new system, Kinect V2 detects multiple people, and the support information is given on the floor.

Roots are essential parts for plants to support plant body and absorb water and nutriments. In addition, the roots grow under various influences from the soils during the growth process. In this paper, we describes the method of analyzing the behavior of the growing roots using a micro force-sensor. We fabricated the microchannel device integrating a micro force-sensor that measures the driving force exerted by root growth and a microchannel measuring the growth rate. This device enables us to conduct plant growth experiments in soil-like environment. By using this devices, we analyzed the influence of soil nutrients on root growth. From the experimental results, we found that the smaller the amount of nutrient, the thinner the root and the smaller the driving force. It is expected to contribute to improvement of cultivation technology of agricultural crops under various soil environments by applying this analytical method.

We have developed a controller design method for dual-stage actuator systems in hard disk drives with the Robust Bode (RBode) plot. The RBode plot represents robust performance criterion with gray area on the Bode plot. By using the RBode plot, control engineers can easily design controller which satisfy robust performance with visualized guidelines. In this study, we designed an H_∞ controller for dual-stage actuator system, and improved the H_∞ controller with the RBode plot. Simulation results show utility of the proposed method for the truck-following control in the hard disk drive.

The simple expectation for safety would be "to prevent accidents". Accident is probabilistic concept. And there is no definitive way to prevent it. Therefore, instead of proactively avoiding by control, we have to take a method of stopping (function itself of original control) before accident (event), at least not to cause critical damage. The title theory (ex ante-safety control) is a system that ensures safety by stopping "before the accident". In this thesis, rather than positively preventing accidents, safety is to stop against accidents that can not be avoidable, at least to avoid causing critical damage. In this thesis, in advance safety theory fundamentally different from safety based on the probability theory so far is under consideration.

This paper describes a real-time tire-road friction estimation system. In this method, the lateral force is obtained by two methods. One is obtained by the centrifugal force, and the other is obtained by tire model using elastic deformation of tire. It is possible to measure the tire-road friction by these values are equal. For measuring, a deferential GPS device is used. Experimental examination using a real vehicle was carried out and the feasibility of the measurement system has been verified by the results.

Since urban railways have many sharp curves, it is necessary to pay attention to the running safety of railway bogies in sharp curves. A value called as derailment coefficient is used as an index of running safety. As the value of the derailment coefficient increases, the afford for wheel climb-up derailment decreases. For measuring the derailment coefficient, a special wheelset called PQ wheelset is generally used. By attaching this wheelset to a vehicle traveling on the commercial line, it is possible to measure the derailment coefficient of the entire route. However, since the running test using this PQ wheelset is not carried out frequently, there is a problem that automation of data analysis is not sufficiently progressed. In this research, we propose an interface that makes data analysis easier for data obtained from PQ wheelset. This system easily detects the point to be paid attention from time-series data of the derailment coefficient. Also, the system is capable of analyzing factors that cause local increase of the derailment coefficient.

Railway operators regularly inspect track conditions and perform running tests in order for the vehicle to run safely and comfortably. It is desired to inspect track frequently to realize safer operation. However, it is difficult to conduct inspections and running tests frequently because of enormous cost. Recently, development of the measurement system using commercial vehicles with onboard sensors has been considered. In the development of the system, it is important to estimate the vehicle velocity and position precisely. Some methods to estimate velocity and position use a tachometer and a GPS. However, the error of the position is accumulated by the integration of the velocity. GPS is effective to reduce the error but it cannot be used in the subway. In this research, a method to estimate the position of the railway vehicle with only an inertial sensor and map is proposed. In this paper, a method to estimate a highly accurate position detection of the railway vehicle with estimated velocity with an inertial sensor while combining with the information of map which is officially provided by Technical Report of the Geospatial Information Authority of Japan. Since the proposed method and measurement system can be implemented on the commercial vehicle, large-scale data can be easily obtained. In order to make the most of such large-scale data, a method to adjust the position of several data along with running distance is also described in this paper.

In developing a pantograph of a high speed train, reducing aerodynamic noise of the pantograph is one of the most important subjects. For reducing this noise, suitable configuration of pantograph head, which has a smooth cross section profile, is investigated. However, the pantograph head with smooth cross section profile and conventional support system of contact strip sometimes wreak undesirable lift force characteristic. In this presentation, multi-segment pantograph head, which has a new pantograph head structure, is introduced to improve contact performance without deterioration of lift force characteristic. The support system is that the pantograph heads are segmented in lateral direction. This paper describes the validation result of this support system with carbon contact strip by some bench test.

Recently, the appearance of electric vehicles has changed the structure of vehicles. Therefore, it is necessary to create design guidelines tailored to the structure of the electric vehicle. In this research, we focus on the relationship between the vehicle structure and the dynamic performance of the vehicle, and create design guidelines for the new generation vehicle structure. As the first step of this research, vehicle structural parts having rigidity which greatly affect the dynamic characteristics of the vehicle were identified. This evaluation was carried out with a stability factor derived by constant radius analysis using flexible multibody dynamics. For the target model, a solar car frame with a general electric vehicle body structure added was used. When pillar part stiffness is high, the elasticity transforms decrease by the vehicle frame cornering force. It is understood that, the vehicle alignment changes, the vehicle stability increases.

This paper considered the yaw resonance mode at general vehicle speed. The equilibrium position is on the extension line of the vehicle speed vector at the position where the sideslip angle becomes 0 in steady turning, and the rear wheel moves toward this line. Furthermore, this paper points out the response parameters that govern the two-stage transient movement at the general vehicle speed.

In this research, we control the active suspension of railway vehicles. Currently the generated force of the actuator in railroad is about 5 kN, so we decide the upper limit and lower limit of the generated force of the actuator by 5 kN, and we will aim for improvement of ride comfort by controlling in it. Model predictive control is handled as control in constraints such as determining the upper and lower limits of the actuator generated force. The model predictive controller consists of optimization and prediction model, and optimization is to minimize or maximize the evaluation function, and the prediction model predicts the future behavior of the controlled object. Solving these optimization problems at high speed and determining the control input at each time is a control method. As a result of simulation by model predictive control, the ride comfort level was 2 (about 83 dB), and good control effect was able to be obtained. As a future task, the model predictive control is a control method that decides the optimal control amount at each time, has a disadvantage that it takes much time to calculate since the amount of calculation is extremely large, so it is necessary to reduce the amount of calculation There is. Also, since the design parameters are very large, it is considered necessary to find out how to determine them.

The part of rail joint becomes the cause of the disadvantage points such as ride discomfort, undesired sound and track maintenance, etc. For restraining those points, the long rail made by welding the standard length rails has been developed. However, the problems of the buckling caused by the axial force of the rail have been pointed out. Though some evaluation techniques for the buckling management are proposed, the development of the technique with non-destructive inspection which quantitatively evaluates the force at the good accuracy and reduces the labor for measurement worker is desired. In the previous study, the compressive stress was increased in 10 [MPa] interval, and the echo at each stress was detected. From that experimental results, it was found that there is the correlation between the ratio of maximum wavelet coefficients (R_MWC) of a wavelet transform (i.e. a kind of time-frequency analysis) result of each echo and the applied stress. In this study, the extraction of the others characteristics led from wavelet transform result is tried, and the accuracy of the stress evaluated by those characteristics is reported.

This paper presents a control of propulsion force of a self-motion on liquid using chemical Marangoni effect by applying electric field to the liquid surface. We used imidazolium ionic liquids as surface active materials to induce Marangoni effect for propelling a locomotor. It was confirmed that the propulsion force of the locomotor was reduced by applying the electric field to the aqueous surface where the ionic liquid was expelled from the locomotor.

In developmental disorders, it is often difficult for a person himself / herself to express intention to apply for consideration due to its characteristics. From the present circumstances, with the aim of supporting expressions of intention in requests for rational consideration, to encourage retrospect from the response process to their own needs, We report an e portfolio system with structure conforming to ICF-CY including objective information. Particularly in task research, wearable terminals were examined from a non-invasive point of view. Specifically, we tried to compare it from the viewpoint of noninvasiveness and measurement accuracy in the portable terminal, the terminal attached near the heart, the wristwatch type terminal, and the headband type terminal.

It is required to provide reasonable accommodation by constructive consensus building using individualized educational support program which is a portfolio of support for persons with disabilities. However, even if it seems that reasonable accommodation is necessary, it is difficult to expression of intention in the request of reasonable accommodation from himself. Within this context, we develop an e-portfolio system including heart rate information having a structure conforming to individualized educational support program, which urge reflection from correspondence process for own request for the purpose of supporting expression of intention in the request of reasonable accommodation. In this paper, we report about the examination using the system in the work experience.