In this paper, we propose a human detection method triggered by a visual sensor. In order for moblile robots to work in urban areas, it is important to be thoughtful for people. And they need to recognize people to a certain distance. 3D LiDAR Now widely used has a narrow viewing angle and cannot recognizing nearby obstacles such as curbs. With the wide FoV 3D LiDAR we use, nearby object can be detected, but instead it is not good at recognize distant objects. In this method, we detect a person first by vision sensor, and the position of the person is detected by scanning with 3D LiDAR. The proposed method is shown to enable detecting more 2m distant person than traditional method.

In recent years, research and development of service robots has become active, so we have conducted research on mobile robots that can be utilized for security, guidance, transportation and other services. In this research, we installed a web camera on a mobile robot and controlled robot movement using AR marker created mimicking the alphabet. As the movement control system, it moves toward the target coordinates entered in advance and moves as it tracks the AR marker as soon as it finds the target AR marker. In order to facilitate the task generation of such movement control, Node-RED was used as an upper level system, so that target coordinates can be entered and start commands can be indicated remotely. In order to confirm the validity of proposed system, experiments were conducted to make the model room travel one round.

This paper presents a pose estimation method for mobile robots with vertical walls in urban areas. This proposed method exploits the fact that most of all artificial walls are built vertically. It corrects accumulative error of estimated poses by estimating the gravity vector with the fact without any 3d maps or 3d models. The poses which are estimated with vertical walls are integrated with integration of angle velocity from inertial sensors thorough extended Kalman filter. To evaluate the proposed method, outdoor experiments with an actual robot are performed. It shows the method keeps correcting accumulative error while the robot moves.

In this research, we aim at the autonomous navigation of a mobile robot using braille blocks. Braille block map has information on graph structure and angle of braille blocks. In the graph structure of the Braille blocks, point blocks are represented by nodes and line blocks are represented by edges. The mobile robot navigates to the destination by switching the edge. In this paper, we describe a route switching method using braille blocks around Shinjuku Station. As a result of the experiment, we could switch to the target route by node switching method using Braille block map.

In this paper, we propose a method to apply the method used in speech analysis called Mel frequency cepstrum (MFCC) analysis to tactile evaluation. We construct a tactile mel filter bank that considers the frequency characteristics of the Pacinian corpuscles that mainly feel the high frequency component of vibration and discuss the relationship between vibration and friction feeling.

Visual occlusions caused by the environment or by the robot itself can be a problem for object recognition during manipulation by a robot hand. Under such conditions, tactile and somatosensory information are useful for object recognition during manipulation. Humans can visualize the appearance of invisible objects from only the somatosensory information provided by their hands. In this paper, we propose a method to generate an image of an invisible object’s posture from the joint angles and touch information provided by robot fingers while touching the object. We show that the object’s posture can be estimated from the time-series of the joint angles of the robot hand via regression analysis. In addition, conditional generative adversarial networks can generate an image to show the appearance of the invisible objects from their estimated postures. Our approach enables user-friendly visualization of somatosensory information in remote control applications.

Human appears to change their movements of touch, such as an applied force to an object and a velocity of the hand tracing the object surface, based on a type of material of the object. The investigation of this difference in touch movement can make the haptic discrimination by robotic sensor systems more effective. In this study, we evaluated the relationship between similarity of haptic texture of 7 different materials and movement of finger of participants during tracing the materials’ surface. We focused on the applied force and the velocity of fingertip as a touch movement. The result indicated that the averaged applied force was corresponding with the similarity of haptic textures.

In this study, a new device which amplify user’s tactile sensation in tracing on tiny undulation has been discovered. The device is a thin and slender flat spiral coil. Also, the amplification factor is quantitatively evaluated using a strain gauge mechatron sand. In this experiment, the designed tactile plate is traced in two states with and without a coil with the mechatro sand, and a signal from the sensor device at that time is acquired. The coil proposed in this paper is 1/5 times thinner than the tactile contact lens and 1/5 times smaller in width. In addition, considering the length of the pin, it was found from the experiment result that there is an amplification factor more than twice that of the tactile contact lens.

In this paper, we propose and demonstrate a new type of soft tactile sensor. This sensor consists of multiple air pressure sensors attached on to the same single air chamber. In this sensor, based on the phenomenon where sensors are subject to different delays depending on positions relative to a source of a pneumatic pressure changing, the contact position can be estimated from the difference of arrival time. In addition, the air chamber functions as a bumper to protect this sensor and underlying mechanism. The experimental results show that the validity of the core idea and the applicability of this sensor.

In quality inspection, it is important to find deformation of parts surface. However, skillful techniques are required for tactile detection. It takes a long time to master skills.

In this research, we propose a planar scanning tactile sensor which detects deformation of the surface. It has a rectangular shape. And two strain gauges are placed in a V shape. And it is embedded in rubber. We conducted an experiment to detect deformation of the surface. As a result, deformation position was detected by the sensor. By improving the accuracy of the sensor, high accuracy and high efficiency of the surface inspection can be achieved.

In this research, we employ spatially controlled thermal radiation to display a virtual shape. Using controlled-thermal radiation, our proposed system displays the virtual 3 dimensional area (e.g. a dangerous area). In this paper, we propose the laser-emission-based thermal radiation system for farther range and precise projection, and discuss about the radiation control method.

Monitoring of biosignals plays important role in health and safe management in daily life. We have developed wide range load sensor based quartz crystal resonator (QCR load sensor) for casual sensing of biosignals. Expansion of the measurement range of the sensor is required for enhancing the usage of the QCR load sensor as casual sensing. We proposed sensor fabrication using vacuum sealing technique for expansion of the measurement range. The sensor fabricated by resist bonding had a measurement range of 0.27 mN - 1180 N, and the measurement range was improved compared with conventional sensor. Proposed sensor has wider dynamic-range compared with commercial product. Eventually, multiple biosignals detection of pulse and weight is successfully achieved with proposed QCR load sensor.

Many studies on material estimation by a robot have been proposed, but appropriate movement on target object is required. On the other hand, when humans obtain material information, they can touch the object with appropriate contact force and scanning velocity according to the shape and mechanical properties of the object. In this paper, material estimation by using a wearable skin vibration sensor is proposed. The sensor allows users to touch an object by the bare fingertip. Ten kinds of materials were used in the experiment and estimation with convolution neural network was adopted. The correct answer rate of the learning model was about 60%, and it was seemed that the material information estimated with the sensor was relevant to the information on roughness and friction.

This paper proposes a compact High Dynamic Range six-axis Force/Torque sensor (HDR six-axis F/T sensor). The high dynamic range of the force/torque measurement is important for robots to acquire environmental information in various operations. However, conventional HDR F/T sensors were too large to attach to robots. Therefore, we developed a compact HDR six-axis F/T sensor with a novel structure named cross-arch strain body. The developed F/T sensor showed linearity with respect to the force component. Since the multiaxial interference component was also linear, it is considered that it can be removed by calibration. It was also shown that the sensor could detect 0.2N to 500N force.

Force feed back devices have been studied. Many of them are desktop type fixed to a heavy object such as desks, and they can not be used for applications with movement. Wearable devices have been developed, however, other problems such as discomfort caused by the reaction force of the device at another part of the body are occured. In this research, a force feed back method using compressed air jetting for wearable force feed back device is proposed. In this paper, concept of the method is introduced and the effectiveness is verified with prototype.

Tactile information might be relevant to communication. In recent years, tactile technologies have been developed and it is becoming possible to share tactile sensations to other persons. However, psychological effects of perception of other person’s tactile sensations is much less clear. In this study, we aim to investigate psychological effects of receiving tactile information of other person. By using a wearable tactile sensor and a vibrator, image with tactile sensations is compared with that without tactile sensations through questionnaire evaluation.

The visually impaired have to obtain shape information through tactile graphics. We have developed a tactile graphics recognition system with electrostatic tactile display, and have investigated recognition rates of displayed figures in our previous research. Based on the result, we improved the rendering methods. In this paper, we conducted a pilot study to explore a better combination of stimuli for an improved method. Furthermore, we investigated the recognition rates of displayed figures with the proposed method.

To create intuitive tactile displays, collecting vibrotactile information is important, though, the collection procedure requires manual scanning of textures. Thus, a collection of vast information is difficult. However, by employing machine learning technology, there is a possibility to generate further virtual data from existing collected data. In this paper, we made a generation model of vibrotactile signals from the collected acceleration data by using Deep Convolutional Generative Adversarial Network (DCGAN). We held two kinds of experiments to evaluate the performance of our DCGAN model and we consider the result of experiments.

We have been researching a rendering method to improve tactile reproducibility in our haptic display that can control vibration in two directions. In this paper, we proposed and evaluated a method to improve tactile reproducibility by using the association between the image feature of actual texture and recorded vibrotactile information.

This study analyzes the recognition process of examining the texture of fine particles through tactile sensing using human fingers and aims at establishing the structure of texture recognition. Characteristic values of the tactile sensing are studied in order to investigate a relationship between tactile sensing and its verbal response. In this study, the SD method (Semantic Differential method) is used to execute the language estimation (the evaluation value) of fine particles. This paper is intended as an investigation of a feature of Kansei Evaluation in around 60 years old using multivariate analysis.

It is important to develop robot systems which can encourage social activities of elder people and the handicapped people. We propose a mobile robot system that follows a human and keep close eye on the human when he or she goes out of home, becomes a chair when he or she wants to sit for break, and assists standing up motion when he or she resumes walking. In this paper, we propose a mobile robot system with the functions of providing sitting and standing assistance to elderly users. Distance between the robot and a user is obtained by depth camera, which measures the distance between the robot and the user’s hips. The control method to beep constant distance between the robot and a user is development. The modeling of motion is standing assistance is introduced in the paper.

This study reports the development of a novel additive device for conventional walkers to enhance gait function of elderly. When the elderly is getting over a step, weight balance is changed in a basket because of the inclination. This causes unintended twisting moment, and falling of the elderly. Further experiments of getting over a step were conducted using a force gauge to observe the effect of an additive device. Results showed that this device make it easier to get over a step.

We will propose a gaze information acquisition system that helps the instructor teach the teaching plan by preserving the measurement data of the driving situation and the gaze information of handicapped children during the training using the electric wheelchair.

In this study, in order to improve an easiness of use for the power assist suit used in nursing care, a hands-free activation device for supplying air to the power assist suit which was developed in our previous study was proposed. After detecting that the caregiver stooped down by using the sensor, compressed air is supplied to the artificial muscles by performing another action. As another action, 3 methods were proposed, which are (1)Use of time lag, (2)Use of bone conduction microphone, and (3)Use of proximity sensor. Mental workload of each activation method was evaluated using CSTLX method, and accuracy of the detection of air supply timing in each activation method was investigated. In addition, effectiveness of the power assist suit was verified through evaluation experiments of lifting and taking down the heavy object.

In this paper, reduction of shoulder burden by the power assist suit developed in our previous study is discussed. There was a problem that the user felt shoulder pain when working for a long time by wearing the assist suit. Therefore, a method of wearing the assist suit was improved. The wearing method of the assist suit by combining shoulder belts at chest front was proposed, so that the burden acting on the shoulder part was distributed to the shoulder part and the chest front. To compare the proposed wearing method and the conventional wearing method, the measurement of variation of the muscle hardness and the evaluation by the visual analogue scale (VAS) were carried out. From the results, effectiveness of the proposed wearing method was verified.

In mountainous region, medical examination bus is often used for medical examinations of the elderly, especially for the ultra-elderly who are usually the users of wheelchair. A slop between the floor and the bus entrance is necessary to let the users of the wheelchairs enter into the bus, in which, the gradient of the slop is as high as 15 degree in some cases. To make the wheelchair run into the bus, the caregivers or the examination staffs have to exert all of their forces to push and/or pull the wheelchair. In order to alleviate the burden of the caregivers or the staffs, in this research we are developing an electric powered assistive wheelchair for examination bus. We designed the wheelchair chassis, developed its control system and HMI (human-machine interface), and conducted the experiments. The experimental results show that the caregiver’s power is reduced by 72% on a 15 degree slope.

Validity of the watchband-type device for feedback of prosthesis hand grasping was investigated. The device is shaped like an index finger and thumb and the bands made by soft material using a three-dimensional printer can bend like a finger grasping motion by feeding air into inside of the band. Therefore, by wrapping around the wrist with a force proportional to the hand prosthesis grasping force, these finger-like bands are able to provide pressure on prosthesis user’s intact wrist. Thus, the device can convey the prosthesis grasping power to the user intuitively. In this study, recognition rate of grasping power by the device was investigated in tests of examinees. The results show that the recognition rate of device was high to some extent in the experiments.

It is hard works to move heavy loads for health person and elderly people. Carriages and walkers are generally utilized, however it is hard to move them at uneven area and they take much space. Therefore, wearable devices with bipedal locomotion are desired. Because existing devices adopt electric motor and pneumatic actuator, they have problems about battery and air cable. So we have developed simple wearable device with only passive mechanical devices.

In this study, we have developed a biomimetic knee joint (BKJ) for assisting human standing motion. In the previous study, we developed an active-type BKJ with artificial pneumatic actuator. However, its weight and size are not suitable for daily use. Thus, in this article, we proposed a passive-type BKJ. We also proposed a shear-force sensitive sheet (SSS) to evaluate unnecessary friction between human and machine in shear direction. We conducted an experiment to evaluate the SSS with testing apparatus which fabricated in order to explore the optimal shapes and materials for the SSS. We performed images processing with images obtained from experiment. It was done to calculate five values which are area ratio, line length, 99% line length, average gradation of line, average radius. We performed multiple regression analysis based on those values and created a model formula.

Upper-limb rehabilitation training for hemiplegic patients has been primarily conducted by human therapists, and, hence, their use of training methods and conditions strongly depends on their expertise. The force control and motion sensing functions of rehabilitation robots are expected to be used for the qualitative training/assessment in the next-generation computerized rehabilitation. We have developed a desktop rehabilitation robot for upper limbs (D-SEMUL). In this study, we investigated the effect of the resultant view after rehabilitation games of the D-SEUL by measuring electroencephalogram (EEG). The EEG was analyzed with sLoreta, which is a kind of a solution of inverse problem. The result shows that the target line shown on the resultant view can improve motivation of training, however its effect cannot continue for a long time.

In our research group, we have been conducting a research on the Self Transfer Assist Robot that realized the transfer by care receivers themselves. A prototype of the transfer assist robot was able to transfer from the sitting position to the sitting position. In this research, our interest are concentrated on the advanced research on "Self Transfer Assist System", which is capable of the transfer including the posture conversion from the supine position to sitting position. This posture conversion is carried out by collaborative work with a plurality of robots. This system consists of "the leg part support robot", "the motorized bed with body position changing function", in addition to "the transfer assist robot". In this paper, considering the load on the leg part supporting robot, we improved "the mechanism that lifts the leg support part" and "the leg support part" so that leg part supporting robot can support posture conversion of supine position to sitting position.

According to a 2018 World Health Organization investigation, worldwide, there are around 253 million visually impaired persons. Technological support is required to help such people walk safely. This paper focuses on the fact that many pedestrians are forming a flow in the actual walking environment. We believe that behavior decisions based on people flow lead to safe and reliable walking of the visually impaired people. The purpose of this study is to assist a visually impaired person’s recognition of surrounding dynamic environments by presenting people flow. To realize this, it is necessary to establish methods of extracting surrounding people flow information from sensing data and presenting the information to a user. In this paper, we propose a method of people flow presentation using vibrotactile stimulation. After identifying the vibration perception characteristics of the presentation positions on a human body, we design and verify the vibrotactile stimulation for people flow presentation.

We have studied a new interface that allows you to "click" real objects, just like a PC's familiar GUI, using a TOF laser sensor on a pan-tilt actuator. In this paper, we propose an intuitive robot operation instruction interface by user 's hand movements without having a device such as PC mouse. Based on our real world clicker, we confirm the validity of the proposed idea to grasp an object with a life support mobile robot.

Children with impaired motor function may be difficult to move on their own. Our research group prototypes an Omni-directional mobile electric wheelchair that has high degree of freedom and can move close to the movement of healthy people and offers it to children with disabilities. However, we must consider that there are individual differences in the operating abilities of children. Therefore, in order to compensate for individual differences and to further enhance the mobile experience, we are working on the development of “operation support function” that supports children’s operation.

In this research, we focus on the fact that children respond to the reactions of surrounding caregivers and enjoy movement experience, and we have worked on the development of a “human following system” that enable a pleasant movement experience to children.

This study aims to develop a fluid actuated pillow in order to provide comfortable sleep. We proposed an active deformed pillow system by air pressure and its operation system. When inner pressure of the pillow was low, the actuators became flat because stuff inside of the pillow was nothing. Due to the property, sometimes the pillow was uncomfortable. We conducted sensory evaluation of a new actuator which contained urethane form. Almost subjects showed that the pillows with urethane were highly rating. Contact pressure concentration was reduced in case of containing urethane. As a result, actuators which contained urethane improved feeling of the pillow. We plan to select better urethane materials and these thickness.

A survey on recognition of seal robot PARO was conducted in Japan. PARO is a nursing and medical robot imitating a baby seal intended for robot therapy and was released in 2005 in Japan. However, compared with other countries, the spread of the PARO is not advanced yet. We distributed questionnaires on PARO’s recognition rate at lectures and seminars. As a result, PARO was well-known by women, and as the age rose, the recognition rate tended to rise. Based on this result, we would like to consider actions to be taken for the spread of PARO for aging society.

In this study, we have developed walking assisting shoes by using elastomer-embedded flexible joint (EEFJ). The EEFJ is a kind of compliant joint, but has non-linear mechanical properties and adjustability for its mechanical properties. This non-linear mechanical characteristics is suitable for plantar and dorsiflexion of ankle motion in walking. In this report, we newly developed an EEFJ with a function for adjusting its initial angle. In addition, we experimentally investigated the effect of shoes on gait with the new EEFJ on healthy elderly subjects with and without the shoes.

Robotic rehabilitation support systems for paralyzed patients have been developed. The rehabilitation support system can provide various training and gradually shift to a difficult task depending on the ability of the trainee. Force displays are classified into active and passive types. A passive-type system using brakes has less risk of harming a user. In this paper, we develop a rehabilitation support system with nano-particle MR-fluid brakes and carry out experiments.

According to survey by cabinet office, there are about seven thousand disabled children with lower limb in Japan and many of them cannot move by themselves. Early moving experience affect their later feelings and exercise capacity, so it is ought to early recover their walking ability. In previous researches, we developed a mobile assist device to support a walking for disabled children by a hybrid driving combining with pneumatic soft actuators and electrical motors. It can use everywhere for children who are difficult to undergo walking rehabilitation. As a result of trial ride with the device, we confirmed it is necessary to extend step that the device support and have capacity to support with various motion, for example route change. In this paper, we will describe about a mobile assist device with new traction mechanism which is capable of supporting longer step than former one and changing step mechanism.

In a nursing home care facility, it is one of the factors that the burden of a caregiver is increasing by guiding many cared persons on a wheelchair before and after a meal. Although many autonomous electric wheelchairs have been researched and developed, it is not reasonable to introduce them for the entire care receivers in a nursing home from the viewpoint of cost, time and trouble of maintenance. We are developing a detachable robot which makes a manual wheelchair run autonomously. This report describes the experimental machines were prototyped by defining the basic concept, target specifications and design conditions, and the results of basic experiments such as straight running stability experiment, speed experiment, measurement experiment of obstacle sensor. In future, we will proceed with implementation of straight follow-up control function and autonomous driving function.

Artificial Arm is an assistive product which is used by people who lost an arm because of illness or accident. It compensates for the appearance and supports various daily performance as part of the user’s body. Recent developments of Artificial Arm are significant. Today, the latest thing in Artificial Arm which applies the human biological information such as Myoelectric Hand has appeared. However, the cost, weight and so on are a heavy burden on users. And deterioration of maintenance by becoming complicated structure is a problem, too. So new Artificial Arm which can satisfy various needs was developed using Pneumatic Artificial Muscles and 3D Printer.

The aging of Japan is currently still going on, and senior citizen households have a tendency to increase. stand-up movement, which is one of daily motions, is a heavy burden on elderly people whose muscular strength tends to decline. Therefore, we aim to develop " a supporting equipment for human stand-up movement" targeted at elderly people who can live independently to support the stand-up movement in order to reduce the burden in daily life, and conduct research toward practical application. In our laboratory, a movable handrail type stand-up movement measuring equipment has been produced experimentally. we analyze the stand-up movement of the user and examine the evaluation method of the support effect from the measured values of the sensors installed in this measuring equipment. In this paper, we focus on the change of the sole-floor reaction force and verify the effectiveness of the support effect on the evaluation.

In this paper, we describe a fuel debris picking-up mobile robot for the Creative Robot Contest for Decommissioning. Our robot has an extendable telescopic arm with gripper utilizing the convex rule, and has 2 crawlers. In addition, it has an omnidirectional camera equipped on robot ’s body and a micro USB camera. The former is for surrounding view, and the latter is for forward view. Our robot is remote-controlled through wired Ethernet by utilizing ROS system. In the contest, our robot could not achieve all competition task. However, it had shown the ability to extend its arm to the bottom of the pedestal to confirm the existence of simulated debris, and return to base safely within the time limit.

This study aims to develop a robot for surveying the interior of a nuclear reactor and sampling fuel debris. To sample fuel debris that has melted in the lower part of the primary containment vessel, it is necessary to extend a rigid arm up to 3 m through a guide pipe with a diameter of 100 mm. To achieve these requirements, a sampling arm using a low melting point alloy as a locking mechanism at joints was proposed. In this paper, we describe an improvement of the mechanism of sampling arm using low melting point alloy and outline of compressive fracture experiment.

In this paper, we propose maneuverability index focusing on the “sense of unity” as a new performance evaluation method for remote control systems. We used the Crossmodal Congruency Effect (CCE) for the quantitative evaluation of the “sense of unity”. We verified the validity as a maneuverability index of “sense of unity” evaluated by CCE. We compared the measured CCEs between the systems which are different in terms of operation force of the master device and camera view orientation of the remote environment. As a result, task performance and “sense of unity” results were consistent. It was suggested that “sense of unity” could be used as an appropriate maneuverability index. Finally, we propose a method to compensate for a bias effect of CCE due to the appearance similarity of target objects to human hand.

This study proposes a new method to reconstruct 3D radiation distribution considering objects in measurement environment. A mobile robot equipped with a gamma-ray detector and a 3D LiDAR is used. The gamma-ray CT is applied for image reconstruction. The mobile robot observes radiation at different points and localizes the position of the robot. In observing process, the robot makes environment map which is made from point clouds at multiple observing points. By using this map made from point clouds, precise reconstruction area is set and reconstruction is conducted in the designated area. Experimental results show the effectiveness of the proposed method.

This paper proposed test methods for the robots for nuclear emergency response. Our motivation is to enhance the development of the robots and the improvement of the operator proficiency for decommissioning of Fukushima Daiichi Nuclear Power Station. We describe some proposals of developed test method for nuclear disaster response ground robots. Subject of the robots’ performance to be evaluated and required conditions are described. Practical example courses and testing procedure utilizing the mockup facility and developed test field are also introduced.

In this paper, we propose an autonomous motion control method for an underwater vehicle using only image information. LED marker which mounted on the vehicle is captured by the camera, then we can automatically recognized its yaw angle and position. Using this information, we can control the position and posture of the vehicle. To show the effectiveness of our proposed method, experimental results are given.

There are many difficult tasks to solve for grasping and transporting operations, specially, one of them has difficult tasks to grasp every object with one type of grasping mechanism. Therefore, robots with gripping mechanism cooperate based on recognition technology to try and solve this problem by grasping and carrying various objects. In this research, we propose a system that shares objects gripped by a robot with a relatively simple gripping mechanism based on recognition technology. Recognition technology based on deep learning recognition using RGB information and recognition technique based on depth information and autonomous mobile robot with different gripping mechanism can handle more objects by sharing grasping transportation for each object. Experimental results have shown the effectiveness of the proposed system.

In this study, we propose a prototype test system for variable flexible spacecraft by using multi robots. The system is designed for gravity compensation of the prototype model of spacecraft which has deployable membrane structures or flexible structures. The system is composed of multi robots. Each robot has parallel mechanism to follow the fast motion of the structural vibration and deployment. The robots are controlled by the force control to support the structures as well as the decentralized autonomous control. The validity of the proposed system and the most suitable support point for this system are verified by hardware experiments.