This paper proposes an innovative pressure imaging approach that utilizes copper foil grids and electromagnetic field variations under different excitation frequencies and conditions, coupled with deep learning algorithms for interpreting voltage signals. The sensor design enables an adjustable pressure detection range along with features such as a concise structure, simple wiring, easy fabrication, and scalability for large-area detection. Multiple excitation frequencies are employed to alter the pattern of the electromagnetic field, facilitating the capture of more comprehensive information and augmenting the dimensionality of the input data, thereby improving the model's capacity to interpret voltage signals. The proposed approach is validated through simulation studies, demonstrating generally accurate imaging of pressure distributions, and proving the concept. In simulated environments, while minor disparities between the predicted and true pressure imaging indicate room for future refinement, the promising outcomes and a mean squared error of 0.102 in validation, with pressure scaled from 0 to 1, affirm the approach's potential.

View full abstract

This paper describes the results of experiments conducted focusing on the influence of the grasping position of the object to be inspected on the results of vibration inspection using a robot hand. In the experiments, pairs of plastic blocks fastened with screws were prepared as inspection objects, which were grasped and vibrated, and the signals obtained from the sensors installed on the robot hand were analyzed. The signals obtained were of two types, force and moment, and each was discussed. As a result, the characteristics of each condition were obtained, and it was confirmed that the grasping position of the object has a strong influence on the results when the robot hand is used for vibration inspection. In addition, it was discussed that the force and moment signals observed by the sensor seems to vary greatly depending on the mass and shape of the object to be grasped, and that the equipment used for the inspection may have vibrated or prevented the observation of the influence of the collision force when the object was rattled. From these results, it is discussed necessary to set optimal inspection conditions such as the grasping position for each part and to discuss the influence of the equipment when conducting vibration inspection. It was also discussed that the optimal grasping position is one that mitigates the influence of centrifugal and inertial forces that may be generated during vibration.

View full abstract

The Brunnstrom stage is an important indicator of stroke patients' rehabilitation status, playing a crucial role in the diagnosis and treatment of patients. In response to the shortage of medical resources and the high cost of rehabilitation, we developed a Brunnstrom stage evaluation system that can be operated on household electronic devices. The system uses Kinema Tracer Motion Capture System to record the motions used by professional doctors to detect the patient's stage and compares the patient's motions with standard motions to determine the patient's stage. The effectiveness of the system is verified by experiments.

View full abstract

Proximity sensing is effective to increase safety in human-robot collaboration. Capacitive proximity sensing can detect objects in proximity over a wide range. However, its detection accuracy suffers from output drift. To compensate for the drift, this study combines a capacitive sensor with an optical sensor that has a narrow viewing angle and estimates the target distance by sensor fusion. A switching unscented Kalman filter was implemented, whose filter setting was switched based on the availability of the optical sensor’s output. A human hand motion was measured by using the two sensors, and an artificial output drift was added on the capacitive sensor’s output. Then, the filter estimated the hand position. It was confirmed that the implemented filter could remove the output drift based on the optical sensor’s output and that it could realize uninterrupted estimation for the viewing angle wider than that of the optical sensor.

View full abstract

In the field of robotics applications, sensing the environment is a crucial step, and recognizing highly reflective surfaces poses a significant challenge. Surfaces with high reflectivity, such as mirrors and glass, can significantly affect sensor data, including depth values from depth cameras and distance values from radar ranging. In this paper, we propose a method for real-time recognition and segmentation of specular surfaces in motion. We extract RGB images, depth data, and motion information captured by a depth camera during motion and utilize the SLAM camera principle to predict feature points. We analyze the displacement characteristics of feature points on reflective surfaces in motion images and subsequently design an algorithm for detecting anomalies in specular features using the DBSCAN clustering method. We integrate the detection of specular surface anomalies and boundary segmentation methods to obtain segmentation results for the specular surfaces in the image.

View full abstract

Current methods for managing swine’s health conditions include visual inspections by breeders, body temperature measurements, and other methods. Among these methods, body temperature is the most frequently used as an indicator of swine health. However, continuous temperature measurement using rectal and vaginal temperatures pose challenges due to hygiene and stress on the swine. In this paper, we aim to develop a new watering device that automatically measures body temperature without causing stress to the swine during watering processes. Specifically, we expect that measuring oral temperature during watering enables continuous temperature measurement without stress to the swine.

View full abstract

Recently, according to Japanese aging society, devices to rehabilitate for the temporally injured elderly and disabled have been actively researched and developed. However, the developed rehabilitation devices are only taken into physical characteristics of human, not technique of Physical Therapist (PT). Therefore, we aim to design the device and actuator with technique of PT. In this paper, based on the technique of PT, a position of the developed Fan-shaped Pneumatic Soft Actuator (FPSA) in our previous study is decided by simulation of MATLAB. As a result, it was confirmed that the angle of ankle joint was approximately 91° when the supply pressure of 20 kPa was applied.

View full abstract

It has been reported that ultrasound can be used for bioengineering research such as cell detachment without enzymes resulting in high cell activity. However, the required forces on cells in such research have not been uncovered since ultrasonic waves generated by out-of-plane vibration on the culture surface develop a complex sound field owing to the reflections of the ultrasonic waves in the culture chamber. Therefore, we developed an ultrasonic device that generates in-plane vibration on the culture surface, which is highly attenuated in the culture chamber to avoid reflections. As a result, it was confirmed that there is an area where in-plane vibration is dominantly excited on the glass culture surface with a driving frequency of 580.3 kHz.

View full abstract

Endoscopic sinus surgery (ESS) is a procedure that requires a high level of skill from the surgeon. Therefore, quantitative feedback is required for efficient surgical training. This study was conducted by regression analysis using the scores evaluated according to the conventional evaluation index ESS-OSATS as supervised information. The surgery using the 3D sinus model was measured by a motion capture system, and the features calculated based on the motion were used for training. Five different regression methods were used to construct the models and Mean Absolute Error (MAE) was calculated to evaluate each model. Comparison results showed that Support Vector Regression (SVR) provided the highest prediction accuracy (MAE = 5.97).

View full abstract

Accurate force and torque measurement is essential in minimally invasive autopsies to avoid tissue damage, aligning with the needs of pathologists and the deceased's family wishes. This research introduces a specialized force-torque sensor designed for this purpose. We designed a sensor using strain gauges, the strain gauges detect structural deformation of the sensor to quantify the forces and torques exerted on a cylindrical cutter. Experimental evaluations of the sensor's performance indicate its capability for basic force and torque monitoring in the autopsy process. However, the results also emphasize the necessity for further development to enhance accuracy. This study contributes an essential tool towards improving the minimally invasive autopsy technique, merging practical utility with ethical considerations.

View full abstract

Trouble with walking restricts many activities in daily life and is one of the factors that cause a significant decline in quality of life. It is expected that the proactive and independent walking with appropriate form is effective for prevention of walking functional decline or recovery of walking functions. Therefore, in this paper, the walking training machine that guides the user’s feet is developed. This device consists only of mechanisms. While the user’s feet are facilitated to walk by the mechanism, the user can exercise to walk within the frame with the caster wheels safely. The power source is the user's own movement with the cane. The rotational motion gained from that action s used as the input to drive the mechanism. The device consists of the arm driving mechanism and the foot guided mechanism that share one crankshaft. Smoothly driving the mechanism is difficult around the limit positions. Therefore, the design is such that the limit positions of the two mechanisms do not occur at the same time. In addition, the operability of this device is improved by adjusting the kinematic constants of the foot guided mechanism and modifying the transmission angle of the arm driving mechanism.

View full abstract

In the workplace, there are many tasks that are performed with the arms raised posture, and this movement causes shoulder and arm disorders. In this research, we developed an assistive device to reduce shoulder muscle fatigue during arm raising. The Remote Center of Motion (RCM) mechanism was used in this device to achieve a structure without a frame around the shoulder and to align the rotational axis of the machine with the rotational axis of the shoulder joint. The assist force of the device was determined by calculation to compensate the load torque generated around the shoulder when lifting the arm. The gas spring was selected as the power source of the device. The placement and selection of the gas springs were also determined by calculation. A prototype of the designed device was fabricated, and experiments were conducted to measure muscle activity during the arm raising operation and to evaluate the effectiveness of the assist. The subjects were three healthy male adults. The experimental results showed that muscle activity in the anterior deltoid was reduced by 38% compared to that without the device in the dynamic posture experiment.

View full abstract

We developed a type of lightweight body-assistive suit designed without batteries and motors. This suit has been shown to assist the biceps brachii and paraspinal muscles, and the interaction of the interlocking rubber belts on the arms and back assists the lifting motion of the worker. On the other hand, not only the arms and back are assisted during lifting movements by that suit, but also the knee joints are required for harvesting work in agriculture, for example, while maintaining a mid- back posture, and for standing up and sitting down for elderly people who have knee problems. If the arm, back, and knee joints can be assisted, it will be possible to contribute to a wide range of workers in various industries and the elderly. Therefore, we have developed a passive assistive suit that assists the arms, back, and knees by means of an interlocking belt. As a result, the evaluation by electromyography and motion analysis confirmed the assistance effect on the assistance effect on the paraspinal muscles, rectus femoris and semitendinosus.

View full abstract

One of the most widespread vehicle aid for movement of individuals with mobility impairment is the wheelchair. Although manual wheelchairs are intuitive in propulsion and effective in manoeuvrability, they might lead to physical disorders and pushing fatigue. Among various power-assisted devices, rear add-on solutions are good trade off in portability and thrust effectiveness, but lack of handling in narrow spaces and control reliability. A novel read add-on assistance device based on a controlled propulsive and steering wheel is here presented. Functional design and control laws are analysed. Performance evaluation is carried out through a dynamic model. Simulations of curved trajectories with or without add-on assistance show how the device may reduce user’s effort in propulsion, providing longitudinal thrust and yaw torque consistent with the user motion request, detected by means of sensorised handrims, still preserving the intuitive pushing pattern as manual standard wheelchair.

View full abstract

This paper presents the working principle and development of a personal mobility platform that is a part of a hand-free wheelchair project. The wheelchair design integrates two control levels, motor steering wheels, cameras, and IMUs, enabling precise motion control based on the rider's movements on a seat. The mobility platform, comprising two motor steering wheels and a passive caster wheel, is designed to satisfy wheelchair requirements of omnidirectional maneuverability. In this article, detailed descriptions of the platform components, including dimensions and functionality, are provided. Additionally, the motor steering wheel design is discussed, emphasizing its crucial role in providing motion and supporting the platform's weight. A preliminary experimental test result demonstrates the platform's capability to navigate in the environment. The results highlight the system's potential to enhance independent mobility and address challenges faced by individuals with mobility limitations.

View full abstract

A novel flexible two-finger hand mechanism that can grasp an object with various sizes and locations with easy control is proposed. The hand is a planer 17-bar underactuated mechanism with one driving motor and five prismatic joints with springs. Mechanism dimensions are designed to set angular velocities of phalanxes to 0 when the hand has just grasped the object, and are optimized by maximizing the difference of relative angles between phalanxes. In order to calculate motion of the hand and grasping force, a kinetostatic analysis of the hand that has contact with the object is carried out. Finally, a prototype of the hand mechanism can grasp objects that have different diameters and locations with input rotation of the motor at uniform speed.

View full abstract

A prototype of a tool electrode holder driven with a voice coil motor (VCM) was fabricated to contact a tool electrode with a workpiece at a small thrust force in a lathe-type electrochemical discharge machine. The top and bottom of a plunger on which the tool electrode was mounted was supported with thin leaf springs so that the tool electrode was fed in only the axial direction. The relationship among the shift of the tool electrode, the driving current of the VCM and the thrust force was calibrated. Then the driving current of the VCM was determined based on the relationship at a present shift of the tool electrode. A thrust force of 0.05 N was stably applied.

View full abstract

By attaching arm robots to mobile robots, it is expected to develop systems for automating household chores, assisting the physically disabled, and so on. However, commercially available arm robots are large and heavy, making it difficult to use them in a limited space such as a home. Therefore, in this study, we focus on the parallel link mechanism and develop a parallel link robot with 3-DOF posture. In this paper, we describe the inverse kinematics solution by numerical analysis and the motion control of posture.

View full abstract

To support handicapped and elderly people in their daily lower-limb activities, a multi-legged robot that changes its form according to the application is designed. Each leg of the robot can be separated to operate as a cane robot with spherical wheels. The cane robot moves on a plane by itself and touches the user to stabilize the user's walking balance and prevent the user from falling. Since walking motion of a multi-legged robot causes some shock when touching the ground, the cane robot must have a more shock-resistant structure than a conventional spherical-wheel robot. In this paper, we propose a cane robot with a built-in shock absorber that can be applied to a multi-legged robot.

View full abstract

When striving for the enrichment of students' learning experience in Mechanism and Machine Science fundamentals, a significant barrier can be the level of abstraction of such topics. Concepts in, e.g., spatial kinematics and dynamics often pose challenges for students in terms of visualization and practice-oriented outcomes. Employing simulation software and physical tools, also in combination, can effectively address this issue. This is where this work comes in, exploring and describing a learning-by-doing practical teaching approach in the Smart Mini Factory laboratory of the Free University of Bozen-Bolzano (I). The laboratory environment, together with the pool of mechanisms, mechatronic and robotic systems, tools, and software available for students, is described; examples of practical applications are also provided.

View full abstract

Animal therapy, known for alleviating stress and mental disorders through interaction with animal pets, has disadvantages such as infection risks and difficulties in care. On the other hand, robot therapy, utilizing animal robots like aibo®, emerges as an alternative solution. This study explores the use of aibo® for robot therapy, in order to reduce personal stress, dementia, and mental illnesses. A developed program, utilizing aibo's ability to detect human gestures, aims to enhance the healing effects on individuals by controlling the robot's actions through the “aibo Web API”, to overcome practical disadvantages associated with alive animals in therapy.

View full abstract

In this study, we propose a method for imitating dance movements in biped humanoid robots utilizing a Model Predictive Control (MPC) framework. By employing the MuJoCo MPC framework, we design and implement an MPC controller that serves as a dynamic filter by predefining the cost function. This filter can transform physically unfeasible whole-body motions into feasible movements within the physical constraints of the system. Experimental results demonstrate that this controller can be applied to dance motion capture data, successfully generating feasible dance motions.

View full abstract

In this study, the authors examined the effects of human stress relief using a four-legged dog type robot, aibo® (hereinafter referred to as aibo) by measuring (1) heart rate by a smart watch and (2) changes in biological signals such as the amount of amylase contained in saliva. In addition, the authors tried to develop the action control for aibo to improve the stress relief effect. Although the appropriate experimental time for contact between aibo and examinee was identified, it was found that the individual value obtained from the biological signal differed greatly from person to person. Therefore, in the future, it is necessary to make examinee well-rested before the measurement, and to search for a more effective stressor.

View full abstract

We describe our experiences in attempting to find ways to acquire behavioral data in child-robot interaction group events in the wild where video recordings are prohibited or limited. We discuss the use of questionnaires and behavioral coding and the limitations we have encountered with both methodologies and highlight the need for more investigation in this area.

View full abstract

Magnetorheological fluid (MRF) actuators have been developed for fine haptic generators. In this study, we report a measurement and modeling method for the rheological properties of the MRFs for their optimal designs. Additionally, we estimated the torque characteristics of a fine haptic device using MRF. Two different MRFs were selected as tested samples. Firstly, the shear stresses of the MRFs were measured and their profiles were modeled with quadratic function of the magnetic flux densities. Next, these two MRFs were filled in a compact MRF clutch and their torque properties were measured and compared with the estimated values.

View full abstract

In this study, the authors attempted to enhance the effectiveness of robot therapy using aibo® (hereafter referred to as "aibo") with natural language processing (NLP) applying BERT to recognize arbitrary command sentences and respond accordingly. During this process, the authors verified that the command sentences were accurately inferred using the question-and-answering task with BertForMaskedLM. However, it was found that complex commands were still not accurately predicted. Then, the necessity of further data augmentation was signified.

View full abstract

In the field of robotics, teleoperation technology is of great significance, which facilitates efficient, safe, and sustainable solutions for human’s work and life. This paper proposes a novel teleoperation approach for mobile manipulators that allows operators to control this robotic system by using one hand. We integrate a hand skeleton detection technology named MediaPipe Hands with the RGB-D camera to obtain 3D keypoint coordinates more accurately. After that, we analyze the various features of our hand in space to form corresponding teleoperation commands. To enable the utilization of particular gestures for transitioning between controlled objects, we introduce a particular hand gesture recognition model leveraging Long Short-Term Memory (LSTM) architecture, which has achieved 100% recognition accuracy for three types of gestures. Ultimately, we perform practical experiments to validate the effectiveness of our approach.

View full abstract

By working under conditions different from those of everyday life, a person can discover new potential abilities. As part of the ability mining research, in order to further understand the potential for manipulation by a toe, it is necessary to investigate the ability of a toe to manipulate when a resistance force is applied to the toe movements. This paper first discusses, on the basis of previous literature, the possibility that a resistance force may facilitate manipulation by a toe. An input device that provides a resistance force to toe movements is developed.

View full abstract

Vibration-induced kinesthetic illusion is a phenomenon in which a person feels as if his or her body is moving when it is not actually moving by mechanical vibration stimulation of human muscles. This illusion can be used in virtual reality and rehabilitation because it makes a non-moving body feel as if it is moving. In this paper, we introduce an experimental device that has been used in the authors’ previous study of kinesthetic illusions. The improvements that the authors applied to the device to facilitate the generation of kinesthetic illusions are described.

View full abstract

Magnetorheological (MR) fluids are suspensions of ferromagnetic particles, medium oils, and several types of additives. Their rheological properties change rapidly, stably, and repeatedly when applying magnetic fields. Therefore, MR fluids have been used to develop several haptic devices. Consequently, we focus on developing a wireless and lightweight handheld haptic gripper with MR fluid device. Subsequently, to reduce the weight of the wireless haptic gripper, we designed the miniature MR fluid device. According to the design result, the weight and torque performance of the miniature MR fluid device were approximately 60 g and 0.27 Nm at 1A, respectively.

View full abstract

Parallel Continuum Robots (PCRs), composed of interconnected rods and disks, offer a blend of flexibility, dexterity, and safety, rendering them promising for tasks within intricate environments such as inspection and repair. This paper focuses on a novel type of PCR which is capable of twisting motion and evaluates its performance with experiments. First, the novel mechanism and the prototype fabrication are introduced. Next, experiments were conducted to measure the pose of the end-effector of the prototype, and its positioning accuracy by comparison with a kinematic model and repeatability were examined. Last, their results were discussed, and future work was suggested.

View full abstract

Pneumatic actuators are expected to be usable in special environments such as dusty, explosive atmospheres and strong magnetic fields. In our laboratory, a pneumatic wobble motor that has air chambers made of polyimide films has been developed. Polyimide film has excellent mechanical and thermal properties among resins. Wobble motor extracts rotational motion from an oscillating wobble ring. However, previous pneumatic wobble motors require a spring mechanism around the wobble ring to restrain it, resulting in a large size in the diameter direction. In this study, to make motors smaller, two functions were assigned to the air chamber, which serves as the driving source for the motor. These functions are to act as a spring and to fix the wobble ring. As a result, the mass and volume of fabricated motor were reduced by 90% and 78%, respectively, compared with the previous motor. A maximum speed was 0.84 rps and a maximum torque was 7.23 mNm. The performance was higher than that of the previous one.

View full abstract

Conventional McKibben artificial muscles struggle to perform passive elongation movements from their natural length by external forces when air pressure is not applied. This feature limits the application of these muscles in musculoskeletal robots, where mutual interactions between artificial muscles occur. To address this problem, we proposed a back-stretchable McKibben muscles (BSM), which can perform the passive elongation movements. We investigated the static characteristics, however, the dynamic characteristics of BSMs were not discussed. Its dynamic characteristics are a significant evaluation index when applying the BSMs to actual musculoskeletal robots. In this paper, we investigate the dynamic characteristics of BSMs by comparing the length of the BSM at pneumatic release and supply.

View full abstract

The Artemis project is being pursued primarily by NASA. However, such manned bases entail extensive operations during construction and operation. Therefore, the utilization of space robots is crucial to alleviate the workload on humans. Humanoid robots, particularly as partner robots, are employed to reduce residents' psychological stress through communication with humans and their own performances. Hence, this paper focuses on the dance performances actively utilized by ground-based partner robots and explores the feasibility of achieving new types of lunar-specific performances that are not easily achievable on Earth by actively utilizing the low-gravity environment. In particular, the utilization of the central pattern generator enables the creation of a variety of dance steps with minimal parameter manipulation.

View full abstract

In recent years, seabed exploration has been attracting the attention of scientists and engineers, and underwater robots equipped with screw propellers play an active role in this field. However, they have some problems such as cavitation and damage to marine animals. Therefore, this study aims to imitate the shrimp’s swimming motions and elucidate its paddle-type propulsion mechanism. First of all, dissection and motion analysis of shrimp were carried out to comprehend the features of shrimp. Then, a drag force propulsion model was developed, and simulations was conducted. By conducting extensive simulations, the effectiveness of this model was confirmed.

View full abstract

In the field of service robotics, wheeled mobile robots have a central role in precision agriculture, logistics, healthcare, inspection and maintenance and cleaning. This paper proposes a kinematic model for swerve-drive robots having two or more locomotion units. The kinematics of swerve-drive systems have already been addressed for specific robots, but a general approach is still missing.

View full abstract

The animal tongue can perform various roles such as feeding, swallowing, and speaking by taking advantage of its flexible and free-moving features. The complicated and dexterous movements of the tongue are acquired by three-dimensionally oriented muscles, and mimicking the muscle orientation can be expected to realize a general-purpose manipulator. Previously, the authors developed the tongue muscle robot which mimicked the human tongue muscle structure by pneumatic artificial muscles. In this paper, as a prototype for the realization of a general-purpose tongue-mimetic manipulator, the performance of the developed tongue robot was reported.

View full abstract

In this study, we propose a method based on the Model Predictive Control (MPC) framework for generating dynamic gaits for bipedal robots. Utilizing the Crocoddyl optimization library, we designed and implemented an optimization strategy that minimizes a predefined cost function while satisfying the robot's dynamic constraints. Experimental results show that our method can effectively generate various gaits, including walking, jumping, and standing, demonstrating the potential of this approach in motion planning and control for bipedal robots.

View full abstract