This paper presents photoelasticity-based force chain analyses for wheel-terrain interaction. The photoelastic experiments can visualize internal force distributions of the terrain beneath the traveling wheel. In this method, the photoelastic disks emulate granular terrain particles. Through the experiments, we evaluated the force chains at large and small wheel slip conditions, then calculated their orientational order. The experimental results show the fundamental characteristics of the force chains at different slip conditions.

The paper introduces a 3D reconstruction of the side wall of the lunar lavatube using multiple time-series images taken by a probe, and quantitatively evaluates the accuracy of the 3D reconstructed shape. First, a three-dimensional CG model of the lunar lavatube is created, and images are acquired by dropping a probe in the CG environment. The 3D model is then reconstructed using Structure from Motion (SfM). In this process, since point clouds with different scales are formed by the upward and downward cameras, they are scaled and aligned in order to integrate them into one. Then, the merged point cloud and the original CG model are divided into sections with regard to the vertical direction in 5-meter increments, and each section is fitted into a cylindrical shape. Finally, the relative error is calculated based on the radius of the approximated cylinder for quantitative evaluation. In the upper and lower layers of the vertical hole, the relative error was less than 10%. On the other hand, the middle layer of the vertical hole had fewer feature points than the upper and lower layers, and the cylinder approximation hardly work well.

Some methods of semi-active vibration control using Magneto-Rheological Elastomers which can change its stiffness are suggested. However, a Magneto-Rheological Elastomer (MRE) can change their stiffness in a very narrow range. Therefore, it is impossible to change elastic force drastically, and accordingly it is impossible to change the vibration characteristics significantly. In this paper, we proposed a mechanism that changes the resonance characteristics by amplifying the elastic force generated from a MRE using a lever mechanism. Moreover, we derived calculation method of amplification rate of the elastic force by using an old type of the Frequency Based Substructuring (FBS) method known as Building Block Approach (BBA).

Experimental modal analysis that is fundamental technology is known to obtain dynamic characteristics of structures. In conventional vibration testing, vibration responses of the structures are generally measured by accelerometers during applying excitation forces by an exciter to ones. The contact type devices cannot be used for untouchable objects such as mirrors to measure the inputs and outputs in vibration testing. If we use a laser Doppler vibrometer (LDV) that is commercially available device, we can obtain the vibration response without contact. The LDV is a point measurement device, so we see the measurement time in a long time as an issue. In this paper, we measure using a surface vibration measurement system incorporating a high-speed polarization camera with an interferometer that is a commercially available one. Moreover, we show that this system can visualize a vibration mode shape of the mirror.

This paper describes the relationship between the shape of Helmholtz silencer and open-end correction. A Helmholtz silencer consists of a neck and a cavity, and its natural frequency is determined by the mass of the neck and the spring constant of the cavity. However, the mass of the neck is slightly increased by the open-end correction. The open-end correction of Helmholtz silencers has been studied in the past, and the open-end correction proposed by Rayleigh and Ingard has some problems such as being inaccurate depending on the shape. On the other hand, the method of calculating the additional mass using modal analysis has been proposed by Yamada et al; however, the condition under which the additional mass becomes large has not yet been investigated. Therefore, we investigated the relationship between the open-end correction and the location of the neck, and the relationship between the open-end correction and the height of the cavity of the Helmholtz silencer in this study. The equations of motion of the entire system of Helmholtz silencers were derived using modal analysis. The location of the neck and the height of the cavity that increase the additional mass were investigated through simulations and experiments.

In this paper, sensitivity analysis focusing on the vibration intensity is proposed. Although research on vibration intensity had been conducted, the method to control it has not been established. In this study, a sensitivity analysis equation to predict the change in vibration intensity due to the addition of mass, stiffness, and damping is derived, and the change in vibration intensity is examined using a finite element method (FEM) model, and appropriate structural changes are discussed. In the FEM beam model, the change in vibration intensity predicted by the proposed sensitivity analysis equation is found to be consistent with the actual change due to structural changes. The structural changes based on the sensitivity analysis allowed us to control the vibration intensity and to reduce the vibration intensity at the target location. This paper confirms the validity of the proposed sensitivity analysis formulae for mass addition and stiffness addition using FEM models.

Advances in image processing technology have made it possible to measure the surface shape of the road ahead while driving. We propose a new semi-active suspension control method that considers the forward road surface shape with MR damper. The MR damper uses MR fluid, which is a non-Newtonian fluid, is affected by Coulomb friction. In the simplified semi-active suspension model, the effect of the Coulomb friction is ignored. In this study, we design the semi-active control system considering the dynamics on the Coulomb friction of the MR damper. A vehicle model equipped with the MR damper can be expressed as an MLD (Mixed Logical Dynamical) model. When the shape of the road ahead can be measured accurately, it can be considered that the information on the future disturbances is available before the vehicle undergoes. In this paper, we formulate the finite time optimization problem of the MLD model in consideration of the future disturbances as a MIQP (Mixed Integer Quadratic Programming) problem in the same way as the conventional optimal control problem without future disturbance. The solution to the MIQP problem can be obtained by a commonly available solver software. The performance evaluation of the proposed method was carried out by a simulation. In the simulation study, the proposed method was able to achieve better ride quality, with an equivalent suspension stroke compared to the traditional MLD predictive control and the Skyhook approximation methods.

In recent years, autonomous driving cars have been attracting attention to reduce traffic accidents. Merging on the highway is one of the most challenging problems that need to be addressed for the realization of autonomous driving cars. The problem is difficult because an agent must decide where and how to merge under a complex changing environment. Reinforcement learning (RL) is one promising way for solving decision-making problems. However, it is hard to guarantee the safety of the controller obtained by RL. Therefore, we propose a combined method that decision making is done by RL and vehicle control by model predictive control to ensure safety. The performance of the proposed method is verified by simulation and shows a high success rate of merging.

This paper presents the merging support system considering the trade-off between distance and acceleration under the constraint of merging zone. In merging, drivers consider about safety and longitudinal ride comfort. In addition, drivers must finish merging in the merging zone. To achieve these objectives, we propose the merging support system. It consists of the merging decision, the merging plan, and the vehicle controller. The merging decision compares the selectable merging positions and selects one with smallest acceleration. The merging plan generates the velocity plan to reach to the selected merging position. The vehicle controller follows the velocity plans. The performance of the proposed system is tested by simulations. In these simulations, the merging decision calculates the plan with the smallest acceleration, and the vehicle controller achieves merging in the merging zone.

With the improvement of automated vehicle technology, it is expected that automated vehicles and Human-driven vehicles will share road traffic for a long time in the future. For some low-speed autonomous driving vehicles (<14km/h), which developed for the elderly, one obvious problem is that if they are driven on public roads for a long time, they will affect the normal traffic flow, reduce the efficiency of traffic, and also cause a decrease in social acceptability. In this paper, we aim to study of avoidance strategy for low-speed automated vehicles to give the way for the approaching rear vehicles, which is expected to avoid the dissatisfaction of the following vehicle as much as possible, and without a large delay in the arrival time of the ego-vehicle. A vehicle deceleration behavior dataset is developed by using the naturalistic trajectory data of human drivers collected on the Urban Streets in the United States, and extract the deceleration behavior of normal drivers when they noticed a slower car ahead and built those data into a new dataset. Based on the regression analysis of the database, we propose a novel dynamic avoidance mobility model for the following car. It can dynamically adjust the timing of avoidance according to the velocity and headway of the following car. The experimental results show that our method can achieve a relatively low delay of the ego-car arrival time while the proportion of the rear driver's dissatisfaction is kept at a low level.

To reduce collision shock and injury risk to an infant in an in-car crib (or in a child safety bed) during a car crash, the force acting on the crib must remain constant below a certain allowable value. Hence, we have been proposed a semi-active in-car crib with joint application of regular and inverted pendulum mechanisms. The crib is supported by arms, like a pendulum; and the pendulum system is supported by arms, like an inverted pendulum. The friction torques of the joints are semi-active controlled using a brake mechanism. However, a collision impact is directed toward the infant’s side in the crib, and the resulting motion of the body is relatively complex. Therefore, the in-car crib has been designed to not only reduce the impulsive force, but also to transfer the force to the infant’s back using a spin control system, i.e., the impulse force acts perpendicularly on the crib. In this study, a control algorithm, that combines the semi-active acceleration control and the active spin control and gives priority to acceleration control, is proposed, and, the effectiveness is measured by numerical simulations. Then, it was confirmed that the horizontal acceleration of the crib can be relaxed by 20% by moving the crib horizontally, and that the acceleration force parallel to the crib can be suppressed by rotating the crib.

These days, lacrosse has become increasingly popular. The role of goalie is very important in lacrosse. If goalie makes a save, there is more chance of winning the match. The save motion of goalie has instantaneous movement of the body and the stick. Despite the fact that efficient forms must be learned and are difficult to master, there are no previous studies on save at this point in time. There is a lack of knowledge on efficient save form and teaching methods are limited to repetition and strength training. Therefore, the purpose of this study is to propose the most suitable form of saving motion for the top of the cross to reach the goal point in the shortest time. In this study, we use a rigid body link model with stick and a human body, set up an evaluation function and solve a time optimal control problem to calculate the form. In this report, a simulation of the pendulum model as a cross is performed as an initial study. In addition, we compared and validated two types of evaluation functions because the items to be evaluated in calculating the optimal form are not clear.

In this paper, we discuss the supporting method when using a standing assistance device. This device is for elderly who are hard to stand up, and chair seat is designed to draw the trajectory which decreases their joint moment. Standing motion depends on each person, so it is necessary to assist at the optimal speed for each person. However, it is hard for elderly to experiment standing motion for many times, so we created a standing motion simulation and analyzed standing motion with various chair seat velocities. We prepared two types of chair seat moving conditions, the first one is based on only one moving velocity and the second one is based on two different moving velocities for former and latter halves.

We investigated the effect of chair seat velocity on human standing motion.

This study discusses autonomous mobile robot (AMR) navigation with a manual operation system to carry equipment and product parts in manufacturing factories of the aircraft industry. Fully autonomous navigation for the AMRs has not still been achieved due to large number of parts and types, and complicated passages. In our operation strategy, when the AMRs cannot perform autonomous navigation, the AMRs have to be manually controlled by operators. Immediate stops of the AMRs have to be avoided to ensure efficacy of the transportation during the switching of the autonomous and manual navigation. An operation device which can be controlled from all the directions around the AMR is required. We focus on that the body of the AMR sways and utilize the sway to control the AMR. Additionally, we implement a power assist function according to longitudinal velocity-based impedance control based on the sway. In the power assist system, we design a dynamic observer based on a mathematical model of the AMR to estimate the hand force from the operator. Experimental analysis of the AMR shows the efficacy of the proposed power-assist control scheme.

In this study, we conducted a correlation analysis between the joint moments and the variables obtained from the experimental data to estimate the joint moments of the driver in order to simulate the driver's behavior during turning considering the driver's intention of a stand-up type personal mobility vehicle (PMV). First, in the vehicle experiment, human behavior, handle reaction force, and floor reaction force were measured using motion capture, 6-axis force sensor, reaction force meter attached to the foot, and reaction force meter attached to the vehicle. From the obtained external forces and human behavior, the joint moments of the driver were calculated using inverse kinematic analysis and inverse dynamic analysis. Then, correlations such as external forces and vehicle speed to the calculated joint moments were investigated. As a result, it was found that the correlation with external forces was high when both the driver operated the vehicle by himself and automatically. In addition, when focusing on the direction of centrifugal force, the correlation between the speed of the vehicle in the direction of centrifugal force and the angle of the joint of tibia, ulna, humerus and pelvis, and the joint angular velocity of the pelvis joint with respect to the joint moment was also high when the driver operated the vehicle by himself.

It is considered that energy consumption can be reduced by selecting appropriate stride length and cadence during walking. In order to verify the assumption in outdoor walking, it is required to estimate the stride length accurately in real-time. The purpose of this study is to build a system to estimate the stride length in real time using inertia measurement units (IMU) fixed on the shank. We estimate the ankle acceleration from the IMU and integrate it based on the Zero Velocity Update (ZUPT) concept. The accuracy of the estimated stride length was verified by comparing with the result from the three-dimensional motion analysis system. As a result, the maximum standard deviation of the estimation error was 41.5 mm. We calculated the stride length by averaging 10 steps to keep the estimation error to around 10 mm. We conducted overground walking tests using the developed system. In this test, the walking speed was kept constant at the individual preferred speed, and the stride length and cadence were designed in five patterns. The test measured the heart rate as an index of energy consumption. Due to the developed system, we achieved overground walking with a stride length variation of less than ±5cm. The heart rate tended to be the lowest when the stride length and cadence were close to the free walking.

Delaying of progress of labor have been occur due to the lack of basic mother’s physical strength in Japan. The delay could be the risk of affecting the prognosis of the mother and fetal. In addition, that could cause urinary leakage and prolapse of the pelvic organs after childbirth due to the burden on the pelvic floor during childbirth. For smooth delivery avoiding such risks, it is desirable that the pelvic floor muscles located near the fetal delivery exit are as relaxed as possible, even when the mother is exerting abdominal pressure during childbirth. In clinical practice, the baby is generally delivered in a supine position on the delivery table with both legs open. In recent years, various delivery postures have been recommended. Thus, it is required that verifying study about the progress of labor in some posture, especially focus on the relaxed state of the pelvic floor muscles. In addition, it is difficult to confirm the contraction and also to be aware of the contraction by oneself since the pelvic floor muscle is a deep muscle. Therefore, contraction state of the pelvic floor muscles during pushing was evaluated by using of an ultrasound imaging system in this study. Then, some indexes were calculated for assessment of state of the pelvic floor muscles. Furthermore, the relationship between the delivery posture and the change of state of the pelvic floor muscles was examined by using a statistical analysis method.

It is important to measure and analyze the motion of human body in various fields. Recently, a new motion measurement method using an inertial sensor has been used. This method is suitable for real life conditions measurements because the measurement space is unrestricted, while the accuracy of the analysis depends on the configuration of the model. For these reasons, it is often applied to the lower limbs where relatively simple movements are performed. In addition, measurements involving complex shapes, such as the upper body, are considered to be difficult. Therefore, we propose a method to construct a detailed upper body model including back and waist from the output of inertial sensors. In other words, we aim to calculate the joint positions of the human upper body using only the output of inertial sensors, and to construct a detailed upper body motion measurement model by connecting them. As a result, the upper body model was constructed from the calculated joint positions, and the motion including bending and twisting was reproduced, which enabled more accurate motion measurement than the conventional method. In addition, in order to make the upper body model more detailed, the shape of the spine was estimated using the accurately estimated joint points on the spine, and the estimated spine had an anatomically correct shape. In conclusion, we have developed a more accurate and detailed upper body motion measurement model than the conventional one.

The guide roller is a vibration control mechanism installed between the guide rail and the car in the elevator shaft, and has the role of guiding the elevator up and down and reducing vibration caused by unevenness on the guide rail. In recent years, elevators have been required to have higher speeds and longer travel distances due to the increasing height of buildings. However, if the conventional guide rollers are used, the vibration generated between the guide rollers and the guide rail will increase and affect the ride comfort of the elevator. Therefore, optimization of the guide roller mechanism is required in the future. This study focuses on the vibration problem in the elevator guide roller part and aims to investigate an effective guide roller mechanism that is both comfortable and durable. For this purpose, an analytical model for calculating the time history response analysis of elevators was developed, and the specifications of the guide rollers were examined through a simulation incorporating springs and damping elements based on previous studies. In addition, a prototype guide roller with the specifications obtained from the simulation analysis is fabricated, and the prototype guide roller is evaluated using a small-scale test apparatus that simulates the running condition of an elevator. This paper describes the results of the simulation analysis and the running test for evaluation the vibration reduction effect of the damping element installed in the coil spring of the guide roller mechanism.

In Japan, earthquake damage such as the Great East Japan Earthquake(GEJE) and the Kumamoto Earthquake occurs frequently. Facilities such as semiconductor manufacturing factories, which require expensive and high-performance equipment, carry economic risks due to damage to those equipment caused by the earthquake. Some general industrial equipment has anti-vibration technology applied to the foundation, but since they target the vibration region with small amplitude and high frequency, they cause great damage to the input of large displacement caused by seismic motion. There is a risk of it occurring. Therefore, in this study, we examine a method of floating the equipment by air and physically insulating the equipment from the foundation. In this study, we study a device that reduces seismic input by using aerodynamic levitation technology, aiming at avoiding damage to the device and maintaining the function of the system for mechanical structures such as anti-vibration devices. In this study, a vibration experiment was conducted to evaluate the effect of reducing the acceleration response. The test equipment assuming a mechanical structure is floated with air, and the vibration response of the test equipment is measured and evaluated by inputting seismic waves. In addition, in order to examine the parameters of the device in the future, we will perform a reproduction analysis in this study. The behavior during aerodynamic levitation is analyzed by time history response analysis that considers non-linear factors such as friction.

The purpose of this study is to reduce the vibration of a repulsive magnetic levitation system using adaptive control method. Active vibration control is necessary to ensure the dynamic stability of repulsive magnetic levitation systems because it is repulsive magnetic levitation system is levitated. Therefore, the vibration is virtually undamped. The purpose of this study is to control the vibration with electromagnetic force generated by using the coil as an actuator. In doing so, we must consider the strong nonlinearity and amplitude dependence of the repulsive magnetic levitation system. This is attributed to the levitation force generated by the magnet varies with the position of the levitating body due to Coulomb's law. Therefore, in this study, simple adaptive control (SAC) method is appropriate to vibration reduction. The experimental device used in the study was designed according to a single-degree-of-freedom system model, and the levitating body with the coil moves up and down using linear guides. The result of the study, it is confirmed that SAC method improves vibration and amplitude dependence.

Vision-based sensors can obtain two or more information at the same time. A camera can be used instead of displacement sensors for measuring displacements in multi directions. In this research, a USB camera is used for measuring the vertical and horizontal displacements of a floator in a two-degree-of-freedom magnetic levitation system. The floator is controlled based on the information from eddy current sensors and levitates stably. A black circle marker is attached to the floator, then an image of marker will be taken by the camera. The center of the marker is calculated in real-time from the pixel intensity in a specific region of a captured image. The displacements by the visual position measurement are compared with the measured values by the eddy current sensors. The results show that the multi displacements can be measured by the camera. However, the displacements are affected by the rotation of the floator, therefore, compensation is required for the levitation control.

This study investigates optimal arrangement of Active Wheel Dampers (AWDs) based on vibration control performance and static deflection amount when multiple AWDs are mounted on sprayer beam modeled as a cantilever beam. Static deflection amount is evaluated as the sum of static deflection amount depending on the gravity affecting on the cantilever beam and static deflection amount depending on the masses of AWDs. The peak gains of primary mode to fourth modes are evaluated from the magnitude of deflection towards disturbance of various frequency at each AWDs mounted position. And the sum of the peak gains that primary mode to fourth modes is normalized by the gain in the static state when the frequency of disturbance is zero. We defined the normalized gain as the magnitude of deflection towards disturbance. The magnitude of the amplitude of output torque of AWDs is calculated, and vibration control performance is evaluated by the magnitude of deflection towards disturbance and the magnitude of the amplitude of output torque of AWDs. Static deflection amount and vibration control performance when multiple AWDs are mounted on the cantilever beam are shown by numerical simulation. Optimal arrangement of AWDs based on static deflection amount and vibration control performance found from the numerical simulation is investigated.

The burden of vibration on passengers during vehicle operation is significant, and prolonged exposure can cause vibration problems such as back pain. However, it is difficult for passive dampers to provide good insulation against steady and transient inputs. In this study, we develop a semi-active vibration control technique using the magnetorheological fluid damper installed in the seat suspension and adaptively changing the viscosity by applying magnetic field. The viscosity is switched according to the on/off and fuzzy controllers based on the skyhook theory, which discriminates the appropriate magnitude of the damping force from the combination of input signals. The effectiveness of the variable viscous damper is evaluated by simulating the system response under two excitations: steady-state excitation and shock input. The results show that the vibration reduction performance of on / off control and fuzzy control is not much different from that of highly viscous one, and acceleration reduction can be achieved at the same time.

This paper describes an energy harvesting technique that uses a wave receiving plate with piezoelectric elements. In this research, a resistive circuit is coupled to the piezoelectric elements. The optimum value of the resistive circuit that maximizes the amount of generated power has not been theoretically derived in previous research. Therefore, the purpose of this research is to derive the optimum value of the resistive circuit considering the effect of water on the receiving plate. In this research, the frequency of the wind wave was regarded as constant. The optimum value of the resistive circuit was derived based on impedance matching. In case where impedance matching condition cannot be satisfied, the optimum value was derived under the condition of maximizing the damping effect of the electrical system. The effectiveness of the optimum value was verified through simulations and experiments.

In this study, we take a uniform beam fixed at both ends as a simple example of a continuum, and the optimum design of multiple dynamic dampers is carried out such as to minimize the vibration amplitude in the primary mode against locally subjected external force. to steady vibration. We dealt with the problem of optimally designing characteristic parameters. The number of dynamic dampers was set to 5, and the arrangement was fixed. The masses of the five dynamic dampers were set according to the amplitude of the natural vibration mode of the primary mode. The natural frequency ratio and damping ratio of the dynamic damper were searched by the round-robin method so that the vibration amplitude of the point of interest was minimized. To compare the vibration suppression performance, we also calculated the result when a single dynamic damper was attached to the center of the beam. As a result, when evaluating only the vibration suppression effect, the single dynamic damper is more advantageous, but when the comprehensive evaluation is performed in consideration of the stress generated in the beam, it is found that the multiple dynamic damper is superior.

Two DOF Frahm damper attached to two DOF system is investigated. The anti-resonance frequencies of whole system correspond to the natural frequencies of the two DOF Frahm damper. Two combinations of the stiffness coefficients of the two DOF Frahm damper are derived for same natural frequencies. These combinations are different in the mode localization of the two DOF Frahm damper, and the differences between the natural frequencies of whole system on the both side of the anti-resonance frequency. The wide difference results in the efficient Frahm damper because the increment of the amplitude is suppressed although the excitation frequency is perturbed. In our previous study, same masses of the two DOF Frahm damper have been considered. However, the mass ratio of the two DOF Frahm damper affects the characteristics of the vibration suppression. In this study, the mass parameter α is introduced, and the two masses of the two DOF Frahm damper are expressed as (1−α)mα and (1+α)mα. The relationship between α and the natural frequency of whole system are obtained. To suppress 1st natural frequency of the main system, the stiffness coefficients without localized modes is more effective. However, to suppress 2nd natural frequency, the stiffness coefficients with localized modes is more effective.

Seismic and wind excitations may cause severe vibration in structures such as highrise buildings and bridges. In this study, we propose a wave absorption control method for reducing vibrations in multi-degree-of-freedom systems using a magnetorheological elastomer-based dynamic absorber. The stiffness of the absorber is changeable according to the applied magnetic field strength. The value is tuned adaptively so that the mechanical impedance at the boundary meets an absorptive boundary requirement. Numerical investigation clarified relationship between the excitation frequency and absorber stiffness that could eliminate the wave reflection from the boundary and maintain no resonant state in the system.

This paper presents an adaptive controller for a class of nonlinear systems referred to as strict feedback form with matched uncertainty. The basic structure of the proposed controller is based on the I&I (Immersion and Invariance) adaptive control proposed by Ortega, Astolfi et al. This approach has many superior features compared to other conventional adaptive approach based on the so-called certainty-equivalence. In some practical applications, however, it can be difficult to design the important manifolds included in the control law. This paper proposes a design method of the I&I adaptive controller using reinforcement learning to overcome this problem. The basic effectiveness of the controller is demonstrated through numerical simulations of a simple 2^nd order system.

This paper presents a control strategy for a wheeled mobile robot with variable pitch blades which make it highly mobile over irregular terrain. In our previous work, a mathematical model of the pitching motion for placing the front wheel of the robot on a step has been developed, and a nonlinear controller for the pitching motion was proposed. In this paper, a mathematical model for placing the rear wheel on the step, which is the latter half of the step-climbing, is developed., and then a nonlinear controller is designed applying the input-output linearization method together with integral sliding mode control to provide robust performance against model uncertainties such as road adhesion property. The validity of the model and the effectiveness of the controller are shown through numerical simulation.

In this report, we apply the stable manifold method, which has been proposed by one of the autors, to the problem of stabilizing a parallel-rotating inverted pendulum. A parallel-rotating inverted pendulum is an underactuated system with strong nonlinearity, and the validity of the theory is confirmed by simulation results using a designed feedback controller. The Hamilton-Jacobi equation has a non-unique solution, which is utilized to solve the problem of swinging up the pendulum with several swings.

Click sensation is the feedback by sound and haptic feeling during pressing down an input device such as a button or a switch. In our daily lives, perceiving a click sensation causes us to recognize completion of an operation. Recently, however, input devices using touch screens, such as smartphones, have become widespread, and these devices do not cause us to perceive a physical click sensation. To solve this problem, a device that can generate a click sensation without stroke was proposed and consists of a plate and piezoelectric elements. The device vibrates at the frequency at which mechanoreceptors in human fingertip are most sensitive to vibration. The device causes the perception of a click sensation. Experiments on human subjects were conducted using this device and subjects were able to perceive a click sensation during pressing the plate.

Recently, touch screens and physical buttons that can be operated with small amount of force are used to operate information devices. Some of them provide feedback on completing their operation through somatic sensation. An example of that is a button that has a click sensation induced by stiffness change. However, the quantitative evaluation of the relationship between pushing sensation and perception of the operation through buttons is still insufficient. The purpose of this study is to quantitatively evaluate the relationship between each element of a button and recognition through somatic sensation. In this article, an experimental apparatus for the quantitative evaluation and its control method are reported. The apparatus consists of a voice coil motor as an actuator, leaf springs that limits the movement direction to the vertical direction, a displacement sensor, a force sensor and a key top. The vertical displacement of the key top can be controlled. A control method of reproducing nonlinear button-push profile including basic factor of pushing sensation is shown. It is experimentally confirmed that the target button-push profile can be reproduced when the pushing force is quasi-static.

In Hokkaido，the vegetable processing workers are getting older，and the shortage of manpower is becoming a problem. In particular, the process of peeling pumpkin and removing the leftover peel is labor-intensive，so there is a need to develop a pumpkin peel removal device to automate this process．In our previous research, the pumpkin surface peel remover robot was able to automatically perform the processes of pumpkin chucking，peeling, and removal, but the peeler part sometimes fell over during the peeling operation, interrupting the operation. This study improves the pumpkin peeler to make the peeling operation more robust and stable．

In Japanese livestock farming, the number of cattle has been declining for 58 years with respect to both beef and dairy cattle due to the aging of the workforce and lack of successors. However, in terms of facilities and equipment for dairy farming, the introduction of milking robots and other have been actively promoted the scall up and automation of dairy farming. As a result, the idea of breeding cows while keeping them healthy has become to be considered less important than before because it is difficult to take care of more number of animals in detail. In particular, accidental deaths of calves and mothers during cow calving cause significant economic damage, and are an issue that needs to be addressed by modern livestock farmers. In this study, we aim to build a system that detects signs of cow calving and alerts the farmers through continuous monitoring using AI image recognition processing. We use an AI tool developed by SCSK Corporation to build an AI model for detecting the signs of calving in cows, and realize automatic detection of calving signs.

This research develops an algorithm that generates an optimal travel path by introducing turning paths using relaxation curves. The travel path for a GNSS-based automatic steered tractor have to be generated with taking the tractor's travel characteristics into account. The algorithm can be applied to the field of the shape without discontinuous side consists, and is implemented as an android application the controlling terminal that with is mounted on a tractor. The outline of the field is obtained by touching the vertices on the field displayed by OpenStreetMap to obtain the location information of the vertices. The polygons are generated from the acquired vertices, and the width of the headland is determined according to the turning path. A clothoid curve is used for the turning path. The sequence of the work path is determined so that the turning can be conducted only forward direction (without backward) movement as much as possible, and finally the overall travel path is generated. The time required for 15 manual tractor turns driven by a male farmer was compared with the simulation time required to trace along the output turning path generated by proposed algorithm. When the minimum turning radius was set to 2.5 m, the time was reduced by about 17.2%, when the minimum turning radius was set to 2.8 m, the time was reduced by about 5.3%, and when the minimum turning radius was set to 3.0 m, the time was extended by about 8.7%.

Most of the current autopilot technologies for agricultural tractors use Real Time Kinematic-Global Navigation Satellite System (RTK-GNSS) that provides positioning accuracy of several centimeters. However, the positioning accuracy of GNSS can be degraded in the area around near windbreaks or buildings. This paper proposes an automatic steering system based on Artificial Intelligence (AI) image recognition using a camera. In this paper, since it is difficult to install mount a large high performance PC for AI image recognition on a tractor, the high performance PC for AI processing is set at a remote data processing room. In addition, the image data obtained by the camera on the tractor are send to the PC for AI processing through the 5th Generation (5G) mobile communication system. This study realizes automatic steering of an agricultural tractor by using AI image recognition method. The results obtained by AI image processing are transferred back to the small PC on the tractor through the 5G communication network. The results show that crop row can be correctly recognized by the AI model based on Semantic Segmentation when applying organic dent corn and organic onion. The guidance line are calculated by the least squares method from the crop rows which are recognized by the AI image recognition. By using a 5G connection, images are able to send and receive stably at 10 frame per second (fps).

In this research, we considered the final-state control for time-varying systems with a stiffness dependent on the displacement. In such systems, the resonant frequency changes during positioning. A frequency-shaped final-state control and a time-varying final-state control were applied, and the obtained control performances were compared by performing simulations. The simulation results indicated that the time-varying final-state control achieved a good nominal performance while the frequency-shaped final-state control method achieved a good robust performance against parameter variations.

In this paper, we propose a control method of quadrotor based on the almost strictly positive real (ASPR) based output feedback control and backstepping strategy in order to realize safer and more accurate automatic flight. By adding parallel feedforward compensators (PFC) to the control system, an augmented system that satisfies ASPR property can be configured, and by controlling this augmented system, the number of steps in backstepping strategy can be reduced. In addition, in order to attain output tracking to arbitrary trajectory of the airframe position, the attitude angles are utilized to the input design in the backstepping strategy. Furthermore, we are constructing a two-degree-of-freedom control system with feedforward input added to the attitude angle control for accurate control performance. Finally, the usefulness of the proposed method will be verified though numerical simulation.

This study presents a construction method of an active vibration control system which does not require mathematical models of controlled objects. A virtual controlled object (VCO) which is a user-defined single-degree-of-freedom (SDOF) system is introduced between an actuator model and an actual controlled object. The appropriate parameter selection of the VCO yields the model-free vibration control. A state equation for the model-free controller design is derived based on the actuator model and the VCO. A model-free state feedback controller is designed via the traditional linear quadratic regulator (LQR) theory. A reference controlled object (RCO) is defined on behalf of the arbitrary controlled objects to tune the VCO-based model-free LQR. An automatic controller parameter tuning scheme is constructed based on the RCO and the simultaneous perturbation stochastic approximation (SPSA) algorithm. The effectiveness of the proposed method is examined by vibration suppression simulations.

Various kinds of devices employing inerter technologies have been proposed for the purposes of structural control and energy harvesting. Especially, in recent years, considerable efforts have been made to develop variable inerter mechanisms to improve their performances. In this paper, model predictive control (MPC) algorithm is applied to the tuned inertial mass electromagnetic transducer (TIMET), one of the inerter devices taking advantage of resonance effect proposed by the authors, to control the variable inerter mechanism in addition to the motor. Numerical simulation studies of free vibration of a single-of-freedom model are carried out, and the obtained results show that the MPC can be applicable for inerter devices in real time and has a great potential to improve the performance of vibration control.

A human body support robot has been attracted, because a robotic technology is expected to support these aging people for walking trainings or rehabilitation trainings, and workers for heavy lifting. These robots tend to mount a single assist control system, walking assist or lifting assist. In this case, the robot does not assist for human when human perform the action that does not mount an assist control system in the robot. When the robot does not assist for human, human suffers from the robot load. When the robot has a multiple assist control system, human needs to change the assist control systems, and this assistance is not seamless. The purpose of this research is to develop the function that automatically changes assist control systems by detecting a human action. This paper describes the method for detecting human action using machine learning. We build the prediction models based on decision tree. We verified a percentage of correct detection and a tree structure for mounting in the robot. In the result, the percentage of correct detection is more than 90%, and we are able to build the prediction model, which the tree structure is simple.

The function of robot manipulation can be extended by attaching tools to end-effector. In addition, deep imitation learning can be used to make the robot arm imitate the use of tools. However, deep learning requires a large number of iterations. Therefore, It is not suitable in actual clearing scene. In this paper, a method to reduce the number of training iterations is proposed by loading deep learning parameters that represent the usage of another tools as initial values. In this paper, the reduced number of iterations and the effectiveness of the method are confirmed. A cleaning experiment is also conducted with silica sand.