In today’s super-aging society, it is essential to maintain and improve the quality of life of the elderly and to counter the shortage of young workers. Such problems are becoming evident in several areas, not just in the manufacturing sector but also in the fields of transport/mobility and healthcare/welfare. In these fields, there is an abundance of situations that cannot be solved simply by replacing human workers with robots. In the manufacturing sector, for instance, the transmission of skills and knowledge is threatened as the members of the baby boomer generation, who have both knowledge and experience, have reached retirement age at a time when the younger population is declining. Another example is the provision of care with human warmth at nursing sites. As we evolve toward Society 5.0, which is represented by a digital transformation and which encompasses elements such as digital twins, it will be critical to pioneer robotic technologies that allow robots to work with humans to achieve a prosperous coexistence and create an affluent society. For this special issue, which features humancentric robotic systems coexisting and working together with humans, we invited researchers to submit papers covering a wide range of topics, including element technologies as well as operating and evaluation methods.

This special issue on Humancentric Robotic Technology and its Applications for Coexistence with Humans includes two review papers, 29 regular papers, and two development reports that cover the following topics.

• Research on rehabilitation equipment and training systems

• Research on power assist and support actuators

• Research on human-robot interaction systems

• Research on motor characteristics and the musculoskeletal system

• Research on the utilization and development of vital signs and sensors

• Research on IoT technology and smart technologies

We thank all authors and reviewers of the papers, as well as the Editorial Board of the Journal of Robotics and Mechatronics, for their help with this special issue.

Based on the roadmap formulated by “Consortium for Human-Centric Manufacturing Innovation” (here-in after referred to as HCMI Consortium) of National Institute of Advanced Industrial Science and Technology, an industry-academia-government collaborative platform launched on April 10, 2019, this paper introduces initiatives and future prospects for human-centric robotics that collaborate with humans and human-in-the-loop cyber-physical system to support future industrial infrastructure for a sustainable and prosperous society. In particular, in cooperation with humans, it is important to understand the behavior of workers. For this purpose, it is important to have a fundamental technology that can represent models of human in cyber-physical systems and simulate the relationship between human and machines. Based on this fundamental technology, we are proceeding with demonstrations in various use cases.

Technology advancement has enabled the development of robotic exoskeletons that are portable, powerful, and sufficiently smart to be of practical use in the real world. These devices provide partial assistive forces that increase their user’s physical strength to better meet the demands of everyday life and have potential applications in various settings. Examples include helping older adults maintain their independence and preventing musculoskeletal injuries among factory workers. Although great strides have been made to improve the performance and usability of these devices, human characteristics and the way humans adapt to the external assistive forces from these devices are rarely explicitly considered in their development. A common assumption is that if the provided assistive forces are aligned with the intent of users, users can easily “switch off” their muscles and effectively utilize this assistive force. In this review, we demonstrate that human adaptations to external assistive forces can lead to inefficiencies or conflicts that decrease the effectiveness of robotic exoskeletons. We then discuss the motor characteristics of human adaptations to external assistive forces.

An ankle joint stretching device controlled by healthy-side ankle movements was developed for self-rehabilitation. Physical therapists treat their patients to prevent a subject ankle joint’s contracture and improve their walking function. However, sufficient rehabilitation therapy cannot be performed because of the labor demands of ankle joint rehabilitation. There has long been a demand for a self-rehabilitation system to reduce the amount of labor required, with the rehabilitation system operated by a physical therapist using a machine. Self-rehabilitation has not yet been realized. By stretching the affected ankle through the movement of the ankle on the healthy side, a self-rehabilitation device that can be used according to the will of the patient can be developed. An experiment confirmed that the device can realize affected-side ankle joint stretching by moving a foot plate connected to a linear actuator using the angle of the healthy-side ankle joint as a trigger. Ankle joint rotation angles of the affected and healthy sides were measured using two acceleration sensors. Compared with the previously used button-push-type control, healthy-side control can realize a smooth and stable affected-side sole-pushing procedure. The proposed system, which does not require operation by a physical therapist during treatment, makes self-rehabilitation of the ankle joint possible.

Recently, the number of older people and the reduction in infant birth rates in Japan have increased. Elderly against the younger increased to more than 28.9% in 2020. According to the Japanese aging society, a welfare pneumatic device to rehabilitate the temporally injured elderly and the disabled has been actively researched and developed. However, the developed rehabilitation device is almost always used for a patient who is not bedridden. Therefore, the joint of a bedridden patient is given only a passive exercise, by a physical therapist (PT). In this study, in order to reduce the burden of PTs, we aim to develop a rehabilitation device that can be used by anyone assisting the bedridden patient. The target joint was decided to be an ankle-joint because it is difficult for patients with contractions to walk. Therefore, a fan-shaped pneumatic soft actuator (FPSA) using a no-stretch hose with weaved bellows and silicone rubber string was proposed and tested. In addition, the characteristics of FPSA with different pitches and lengths are measured. To predict the bending angle of FPSA, a simplified shape analytical model of it is proposed. It can be confirmed that a calculated bending angle of FPSA using the model generally agrees with the experimental result. The proposed analytical model is a first-stage model that does not include the influence of external forces on FPSA. The ankle-joint rehabilitation device using FPSAs is proposed and tested, where an ankle-joint is changed from a plantarflexion to a dorsiflexion motion by pressurizing both FPSAs on the tested device. And it can be confirmed that a range of motion (“ROM” for short) is satisfied in an inversion and eversion driven by operating one or another FPSA on the device.

Rehabilitation robots and rehabilitation braces are in demand, as they aid in reducing a patient’s need for therapist’s attendance and guidance during therapy. This study focuses on the bending of fingers under the tension of the fingertips using clinical equipment and demonstrates the development of a small and versatile rehabilitation brace.

Owing to aging populations, the number of elderly people with limb dysfunction affecting their daily lives will continue to increase. These populations have a great need for rehabilitation training to restore limb functions. However, the current numbers of rehabilitation hospitals and doctors are limited. Moreover, people often cannot go to a hospital owing to external conditions (e.g., the impacts of COVID-19). Thus, an urgent need exists for telerehabilitation system for allowing patients to have training at home. The purpose of this study is to develop an easy-to-use system for allowing target users to experience rehabilitation training at home and to remotely receive real-time guidance from doctors. The proposed system only needs a monocular camera to capture 3D motions. First, the 2D key joints of the human body are detected; then, a simple baseline network is used to reconstruct 3D key joints from the 2D key joints. The 2D detection only has an average angle error of 1.7% compared to that of a professional motion capture system. In addition, the 3D reconstruction has a mean per-joint position error of only 67.9 mm compared to the real coordinates. After acquiring the user’s 3D motions, the system synchronizes the 3D motions to a virtual human model in Unity, providing the user with a more intuitive and interactive experience. Generally, many telerehabilitation systems require professional motion capture cameras and wearable equipment, and the training target is a single body part. In contrast, the proposed system is low-cost and easier to use and only requires a monocular camera and computer to achieve real-time and intuitive telerehabilitation (even though the training target is the entire body). Furthermore, the system provides a similarity evaluation of the motions based on the dynamic time warping; this can provide more accurate and direct feedback to users. In addition, a series of evaluation experiments verify the system’s usability, convenience, feasibility, and accuracy, with the ultimate conclusion that the system can be used in practical rehabilitation applications.

With the increase in the aging population, the demand for healthcare devices has increased. Among the aging population, many experience joint pain, muscle weakness, or poor balance. Without external help, such persons may have difficulty walking, fall easily, or experience poor quality of life. To provide this category of the population with the opportunity to exercise and walk safely and take care of themselves in daily activities, a stepper motor-powered walking assistive device is proposed and built. The device enables users with different heights to walk in suitable gaits; assists users to stand up, sit down, and step up or down stairs; automatically detects user’s intention using pressure sensors and potentiometers; and can be used outdoors for long periods. In this study, the effectiveness of the device is verified using muscle-activity measurements.

Total knee arthroplasty (TKA) is the primary treatment for knee osteoarthritis. However, TKA is highly likely to result in prolonged chronic postoperative pain. The one-foot-one-step walking style is likely to induce fear of movement because of pain, leading to catastrophic thinking about the actual activity and consequently limiting movement. The aim of this study is to develop a system to induce the sensation of stair climbing through the interaction of the visual and kinesthetic senses. By controlling the amount of movement of the foot and the point of view in virtual space, the system can present a visual image of stair climbing even when the patient steps in a fixed position. This system enables easy motor imagery intervention even for early postoperative patients who have difficulty with the actual stair climbing movement. The clinical intervention experiment confirmed that the smoothness of the knee joint motion during descent was improved by intervening with motor imagery during stair ascent and descent for TKA patients.

This study proposes a method for improving the performance of a manipulator driven by pneumatic artificial muscles. Although the straight-fiber-type pneumatic artificial muscle (SF-PAM), a kind of pneumatic artificial muscle, is lightweight and exhibits high contractile force and contraction percentage, its contractile force decreases as contraction increases. To compensate for the decrease in the SF-PAM contractile force, we developed a noncircular pulley and integrated it into the manipulator driven by a wire pulley mechanism. Because this noncircular pulley is designed in accordance with the output characteristics of SF-PAM, the contraction force of SF-PAM can be converted into manipulator torque efficiently. In addition, the radius of the noncircular pulley is expressed as a function, which can be incorporated into a numerical model for the manipulator’s controller. Subsequently, simulation and experimentation to verify the proposed method showed that, when using the same actuator, the manipulator with a noncircular pulley can optimize both output torque and range of motion better than that with a conventional circular pulley. However, a few differences between simulation results and experimental results were observed. These differences were caused by SF-PAM stretching which was not considered in the model. This drawback can be overcome by improving the SF-PAM and the numerical model in future studies. We believe that this study will provide designers of robots that coexist with humans with a high degree of freedom.

In a typical pneumatic soft actuation system, the control valve is often heavy and expensive, making a compact and economical control valve highly desirable. In addition to improving the valve itself, reducing the number of valves required can also be beneficial. To address this requirement, this study proposes and tests a multi-port switching valve based on a slide-gate mechanism. This valve is designed to enable the supply or exhaust of multiple actuators simultaneously, thereby reducing the number of valves required to drive a complex pneumatic actuator with multiple chambers. The main components of the proposed valve are a rotational cam, multiple ports, and 12 gate pins. The construction and operating principle of the valve and a design method for a rotational cam are described in this study. Furthermore, the results of the designed valve for radial bending and extension of a flexible robot arm are also reported.

Because of the elderly society in Japan, a passive exercise device that can be used in homes or hospitals without assistance is desirable. In the device, a soft actuator that can move with both longer strokes and a larger force is useful. In the previous study, a pneumatic linear stepping actuator (PLSA) that can push and pull a flexible rod while changing the gripping position of a tube was developed. In this study, a variable step type PLSA is proposed and tested. Moreover, the development of a passive exercise device for the hip joint using three PLSAs and one connecter connected with three flexible rods and a knee supporter is described. The tracking position control of the knee supporter based on an analytical model is also described. The control system using developed actuators is also introduced.

In the case of the treatment of low back pain, the lumbar corset which is the orthotic therapy is mainly used. However, there is a problem that the trunk muscle group is not used continuously by wearing the conventional lumbar corset for a long time, and muscle decline may occur. In this research, we propose an intelligent pneumatic corset that detects the forward bending posture with a sensor and automatically drives the soft actuator according to the attitude angle. It detects a forward-leaning posture in which the burden on the lumbar region increases, and automatically pressurizes the soft actuator. And, in the corset developed so far, the effect of supporting the muscle of the lumbar region was expected by compressing the abdomen and increasing abdominal pressure. In this paper, the operation support method by the restorative force when the soft actuator is deformed and the design of the intelligent pneumatic corset with built-in them are described.

In recent years, the aging of society has led to challenges such as ensuring adequate facility capacity and promoting health maintenance for the elderly. Consequently, there has been a growing demand for rehabilitation and fitness equipment suitable for use in limited spaces such as homes. In a previous study, a hexapod mobile robot, equipped with six tetrahedral-shaped pneumatic soft actuators (TSAs), was developed as a core training device to address these issues. However, the robot’s gait was investigated experimentally via a trial-and-error process. Hence, it is necessary to examine the stability of the gait. In this study, a straightforward model is presented for gait analysis of the mobile robot. Furthermore, the stability of the gait is demonstrated based on the analysis, and a more efficient gait with a sufficient stability margin is introduced.

As the population ages, the number of elderly people suffering from systemic diseases such as stroke increases. To address this problem, various wearable walking assistive robots have been developed to promote physical exercise for stroke prevention. Wearable assistive robots have shown the ability to improve human mobility. However, most of these robots are heavy, bulky, and impractical. In this study, we developed a compact ankle assistive robot for elderly users to promote walking exercise. By informing the user of correct motion and timing, the robot can guide the user to achieve a healthy gait by only assisting their ankle joint. The robot provides faster-than-ankle motion to allow the user to feel supported while walking. Users can adjust the robot’s assistance parameters through a graphical user interface (GUI) according to their needs. Furthermore, we proposed a gait-adaptive method for ankle assistive robots to adapt to the user’s changing gait. Hence, the robot can automatically adjust the parameters to provide more accurate walking assistance. Finally, the results of an evaluation experiment demonstrated the feasibility of human gait adaptation. The proposed methods have the advantages of low cost and easy implementation.

One of the serious problems that Japan has been facing is its growing aging society, which causes a shortage of young workers in the nursing field to support care recipients and in other industries that require heavy labor. Many workers in these workplaces experience back pain. To reduce the burden on the lower back, the squat method is generally recommended when workers lift heavy objects. Many assist devices have been developed to reduce the burden on the body; however, most wearing type devices have difficulty performing the squat method owing to their mechanical constraints. In a previous study, we developed a non-wearing type pneumatic power assist device that considers the squat method and evaluated its support effects during a simple lifting motion. In this study, we developed a passive type assist device and a previously developed active type device and quantitatively evaluated their effectiveness in lifting motions. In addition, we analyzed muscle activity using a musculoskeletal simulator.

In this paper, we introduce an adaptation to steering assistance systems for providing impedance control. We also suggest using the damping ratio as a design guideline for the parameters of the impedance control. In our research, experiments were conducted with experimental participants using a driving simulator implementing the suggested steering assistance system. As a result of both subjective and objective evaluations, it was found that the suggested steering assistance system is better accepted by drivers than the conventional system. Moreover, the use of the damping ratio as an evaluation index facilitates the design of the impedance parameters. These results indicate the effectiveness of our suggested system.

In recent years, there has been strong focus on sustainable development goals (SDGs). In Japan, Society 5.0, which is proposed as a future vision to be achieved towards the realization of SDGs, is being promoted by various organizations. In particular, “i-Construction” is being promoted in the construction industry. As a result, hydraulic excavators are becoming increasingly automated and semi-automated. Furthermore, if operators achieve high productivity, it helps them maintain a sense of accomplishment and motivation to work. In this study, a control system that results in the desired output is proposed for a hydraulic excavator; in this system, the degree of interference depends on the human input and controller input. Attachments on hydraulic excavators cause interference owing to the characteristics of the hydraulic system. Therefore, a control system that can adaptively adjust the controller input to the human operation, considering the interference with information caused by human operation stored in the database, was constructed. The proposed method was implemented on a hydraulic excavator and its effectiveness was verified.

In recent years, collaborative robots able to work together with humans have become widespread in production sites and factory automation fields. In this context, the goal is to provide a production site where cooperative robots can share various types of tools with humans. However, we have not yet reached the point where collaborative robots and humans can share tools for advanced skills. Therefore, it is very important to investigate how a collaborative robot can control advanced tools requiring human skills and the processes for realizing such control. Musical instruments are some of the most difficult tools to handle. It is important to focus on musical instruments because they allow us to evaluate the robot’s manipulation not only by evaluating its movements, but also by evaluating the resultant sound. In this study, we consider a flexible and large deformable musical saw as an unknown and advanced tool difficult for humans to manipulate. In a previous report, a support method for imitating a human was adopted and provided manipulation control based on striking sound feedback using a cooperative humanoid robot. In the present paper, the support method is improved at the tip of the musical saw to realize an advanced wrist motion. In addition, we discuss skillful manipulations based on striking sound feedback control while considering the positions and postures between the collaborative humanoid robot and musical saw.

Discussions on human-robot interactions abound in the fields of human-agent interaction and human-robot interaction. Although studies on communication through conversations and physical actions have been conducted, many of these studies focus on robot gaze while neglecting human gaze. In this study, we developed Pocketable-Bones, a mobile robot that can move with the human gaze in mind. This robot can fit in a breast pocket and follows and turns in the same direction a person faces. Notably, we have been investigating the realization of joint gazing in which two persons share interests and concerns through gazing. This study’s experiment results showed that Pocketable-Bones’ gazing behaviors satisfied the components of self-determination theory in well-being, notably autonomy, competence, and relationship.

This study investigated a pregrasp strategy for hooking and caging ring-shaped objects. Through-hole features enable the robot hand to hook an object with holes by inserting its finger into one of the holes. Compared to directly grasping the ring, an inserting motion is more convenient to allow the uncertainty of positioning errors and avoid collisions between the hand and the object. Instead of recognizing the exact shape of the object, we only detected its ring-shaped feature as a through-hole to be inserted and estimated its approximate center position and orientation from the point cloud of the object. The estimated geometric properties enabled the approaching motion of the robotic gripper to complete insertion. The proposed perception and motion-planning method was demonstrated for rigid and deformable objects with holes.

A novel gaze-based user interface (UI) is proposed for the remote control of robots and electric wheelchairs. A task-based experiment showed that this UI is more suitable for remote control than a conventional interface. The UI was applied to the control of a commercially available low-cost electric wheelchair. By using a 360° camera and an eye tracker that can be used outdoors, the visibility of obstacles and the usability of the gaze-driven electric wheelchair were greatly improved. The gaze-driven electric wheelchair exhibited good performance in a task-based evaluation experiment.

Muscles in the musculoskeletal system can only transmit forces in the tensile direction, thereby resulting in redundant actuation. This redundancy creates an internal force among the muscles. The musculoskeletal potential method uses a potential field generated by the internal force among muscles and performs a step input of the muscle tension balanced at the desired posture to achieve the point-to-point (PTP) position. This method is extremely simple and does not require any sensory feedback or complex real-time calculations, as long as the target muscle tension is achieved. However, it is known that convergence to a desired posture is strongly influenced by muscular arrangement. In a previous study, we limited our analysis to a specific structure with two joints and six muscles and explained the conditions for convergence to a desired posture. However, when the structure of the target system, number of joints, and number of muscles are different, the convergence conditions cannot be clarified using the previous method. In this study, we extend the previous method to a musculoskeletal system with multiple one degrees-of-freedom (DOF) joints driven by monoarticular and biarticular muscles. In this study, we clarify the conditions that must be satisfied by the muscular arrangement to converge to a desired posture in the musculoskeletal potential method and verify the results through simulation.

We considered the change in joint stiffness as a motion strategy in human lumbar extension movement. We developed a musculoskeletal model that can reproduce this motion strategy and attempted to clarify the individual characteristics and differences in motion strategy. This model reproduces changes in lumbar joint stiffness during movement by appropriately adjusting parameters such that they represent the motion strategy. Using this model, we identified parameters in lumbar extension exercises for six participants with differences in lumbar joint stiffness and examined changes in lumbar joint stiffness and motion strategies. The results showed that participants with similar changes in joint stiffness had commonalities in their motion strategies.

We propose a six-axis force sensor with a simple structure that can measure small forces while affording a high load capacity. For a robot to perform complex motions in an unknown environment, the force sensor used must exhibit a high resolution and high load rating. Various studies have been conducted to simultaneously satisfy these two requirements. However, the high processing cost due to the complicated sensor structure is problematic. Therefore, we develop a column-type high dynamic range (HDR) six-axis force sensor using two types of strain gauges. The sensor was composed of a drawn pipe to solve structural problems that arise during manufacturing. By attaching strain gauges to the surface of the drawn pipe, the forces and torques in the six axes can be measured. HDR measurement was realizable using a semiconductor strain gauge for small loads and a metallic foil strain gauge for large loads. Based on the simulation results, the rated loads of the sensor were 1400 N in the F_x and F_y directions, 9000 N in the F_z direction, and 120 Nm in the M_x, M_y, and M_z directions. The performance of the fabricated force sensor was evaluated. The maximum nonlinearity errors of the semiconductor strain gauge and the metallic foil strain gauges were 1.21% and 1.81%, respectively. In addition, when comparing the S/N ratios, the minimum measurable values were 0.035 N and 0.13 N for F_y and F_z, respectively.

This paper discusses an indoor positioning technique aimed at human-centric services such as pedestrian navigation or service robots. The method is called “CEPHEID” and uses a light flickering pattern as an environmental fingerprint. The authors found that each lighting fixture has unique and distinguishable flickering characteristics. In this paper, CEPHEID is introduced as a “classifier” and its validity is shown based on experimental results. Additionally, an approach for improving the positional precision is proposed. The classifier and regressor are combined to create a zone-classified regressor model for CEPHEID. The basic performance of this concept is also tested using an experiment.

The progressive aging of society has increased expectations for the spread of nursing care robots to support long-term care and welfare services. This research had the goal of developing a communication system as one of the elemental technologies of nursing care robots, along with a method that allows care robots to consider a user’s emotions. The estimation of emotions based on a user’s electroencephalogram and heartbeat has attracted attention. However, users may experience stress when wearing the sensors needed for such measurements. To prevent this system from causing stress in users, we had the goal of developing an estimation model for emotions based on the pulse, which is relatively easy to measure. Various autonomic nervous activity indices (pNN50, RMSSD, LF, HF, LF/HF) were adopted for the estimation model, and transfer functions were established. These indices were considered in time domain and frequency domain analyses of the heart rate variability. The pulse was measured while the user was watching a video and converted into an accelerated plethysmogram using second order differentiation. Then, the autonomic nervous activity indices were calculated. The transfer function from the input to output was identified using these autonomic nervous activity indices as inputs and the responses to a questionnaire that was administered after watching the video as outputs.

In this paper, we report the development and evaluation of an excretion detection device which is to be used with a cushion. Remarkably, the device does not require air suction and can be used during the daytime. The device can be placed on a chair or wheelchair and used whenever a cared person sits. Reflecting on the views of caregivers and the observations of the cared persons, the device was designed to be as thin and flexible as possible. In experiments to evaluate the response, it was confirmed to respond not only to simulated urine but also to actual excrement. From the results of the clinical experiments, the excretion detection model achieved an accuracy rate of 76%. Although this figure is close to the approximately 80% accuracy rate of an existing product H, it must be kept in mind that this accuracy rate is limited to urination owing to the limited training data in this study. These findings suggest that the device used with a cushion provides a valid means of excretion detection during daytime hours in a non-wearing and non-invasive manner, although its use is currently limited to urination.

The transversus abdominis muscle is important for various human functions, including the prevention of back pain and the use of muscle activity values as an indicator for the treatment of urinary incontinence. Measuring muscle activity is an effective means of evaluating whether correct training has been performed. The transversus abdominis muscle stiffness sensor developed in a previous study used a method in which a belt with a sensor attached was wrapped around the lower back, which made measurement simple but also caused problems owing to the influence of phenomena other than muscle activity on the measurement data. Therefore, we propose a novel correction method to calculate the muscle activity of this sensor accurately, which can be used for simple measurements. We propose a formula to compensate for effects other than muscle activity based on a model equation that includes contact conditions and belt tension, which affect the muscle stiffness sensor value. We evaluated the validity of the correction equation from simulations and the change in similarity owing to the correction by simultaneously measuring the muscle action potentials and muscle stiffness sensor values during breathing movements with muscle activity. The proposed correction method improved the muscle stiffness data during respiration and contraction. It is possible to improve the measurement accuracy by improving the calibration method and hardware.

The application of vibration stimulation to muscles, via the skin surface, can generate the sensation of movement, when actually there is no motion. This phenomenon is called kinesthetic illusion. Recently, in the fields of rehabilitation and virtual-reality technology, research has been conducted to utilize kinesthetic illusions to feel body movements, when there are none. To apply kinesthetic illusions in the above fields, it is necessary to develop techniques to improve the occurrence rates of the kinesthetic illusions and shorten the latency, which is the time lag from the onset of stimulation to the occurrence of the illusion. In a previous study, the authors reported that the occurrence rate of kinesthetic illusion could be improved by simultaneously applying vibration and electrical stimulations to the antagonistic muscles. In this study, the influence of this technique on the latency of the generated kinesthetic illusion is investigated by applying a combination of vibration and electrical stimulations. Three different electrical-stimulation voltages are used in the combined stimulation to induce the kinesthetic illusion, and the latency is studied for each voltage condition and the vibration-only condition. The effects of the voltage change on latency are evaluated from a regression analysis performed using the generalized linear mixed model. The results suggest that the change in the electrical stimulation voltage can shorten the latency of kinesthetic illusion.

The rise of single-person households coupled with a drop in social interaction due to the coronavirus disease 2019 (COVID-19) pandemic is triggering a loneliness pandemic. This social issue is producing mental health conditions (e.g., depression and stress) not only in the elderly population but also in young adults. In this context, social robots emerge as human-centered robotics technology that can potentially reduce mental health distress produced by social isolation. However, current robotics systems still do not reach a sufficient communication level to produce an effective coexistence with humans. This paper contributes to the ongoing efforts to produce a more seamless human-robot interaction. For this, we present a novel cognitive architecture that uses (i) deep learning methods for mood recognition from visual and voice modalities, (ii) personality and mood models for adaptation of robot behaviors, and (iii) adaptive generalized predictive controllers (AGPC) to produce suitable robot reactions. Experimental results indicate that our proposed system influenced people’s moods, potentially reducing stress levels during human-robot interaction.

This study aims to provide listening training and comprehension support using a communication robot for young and overseas staffs who have problems understanding dialects in medical and nursing care settings. In this study, we conducted an impression survey of university students in a region where dialects are used using a robot and a person who speaks standard Japanese and a dialect (Mikawa dialect). Moreover, a nationwide impression survey using only a robot was also performed. In the dialect-using region, the robot speaking the dialect gave the impression of being cute but difficult to understand, whereas no difference in impression was observed in the nationwide survey. However, the analysis of the free text showed that the dialect and motion of the robot gave a positive impression. Based on the factors contributing to the positive impression of the dialect, we aim to construct a training program that motivates young and foreign personnel.

Recently, the increasing social isolation of the elderly has caused major social problems, such as loneliness and the progression of dementia. A human-centric system could be a solution to these problems and promote coexistence with humans. Therefore, we aimed to develop a robot system using smart devices, which are essential for the Internet of things (IoT) technology, to provide services, such as information support and monitoring. As the development and application of smart devices become more sophisticated, a hyperconnected society will finally be realized through the development of smart-device-centered robots and their connection to peripheral devices. A hyperconnected society is one in which people, things, and data are connected. Personal mobility is developing and converging with robotic technology to the point where a large mobile robot can board a person. These robot technologies can be connected to wireless networks to provide organically connected services. In the era of Society 5.0, the connection among smart devices, robot systems, and mobility technology is still developing and will be a new paradigm in the development of human-centric systems in the future. Therefore, this study introduces the creation of a human-centric system using a robot system and a mobility system based on the IoT. Finally, we present several examples of the effectiveness of the proposed system and discuss its applicability.

In this work, we have incorporated an electronic travel aid (ETA) as a smartphone application that alerts fall risks to the visually impaired. The application detects negative obstacles, such as platform edges and stairs, and occlusion using a grid map including height information to estimate fall risk based on the distance from an area’s edge to the user, and the area ratio. Here, we describe a grid map correction method based on the surrounding conditions of each cell to avoid area misclassification. The smartphone application incorporating this correction method was verified in environments similar to station platforms by evaluating its usefulness, robustness against environmental changes, and stability as a smartphone application. The verification results showed that the correction method is, in fact, useful in actual environments and can be implemented as a smartphone application.