Abstract This paper proposes a new state transfer method for geographic state machine replication (SMR) that dynamically allocates the state to be transferred among replicas according to changes in communication bandwidths. SMR improves fault tolerance by replicating a service to multiple replicas. When a replica is newly added or recovered from a failure, the other replicas transfer the current state of the service to it. However, in geographic SMR, the communication bandwidths of replicas are different and constantly changing. Therefore, existing state transfer methods cannot fully utilize the available bandwidth, and their state transfer time increases. To overcome this problem, our method divides the state into multiple chunks and assigns them to replicas based on each replica's bandwidth so that the broader a replica's bandwidth is, the more chunks it transfers. The proposed method also updates the chunk assignment of each replica dynamically based on the currently estimated bandwidth. The performance evaluation on Amazon EC2 shows that the proposed method reduces the state transfer time by up to 47% compared to the existing one. In addition, we apply the proposed method to dynamic replacement of replicas, which can mitigate latency degradation caused by network trouble, and evaluate how fast the method can relocate a replica.

Massage therapy is crucial in the field of physical activities and sports to recover muscle fatigue and reduce stress. However, the physical workload on massage therapists is demanding, and the number of therapists is limited. To solve these problems, a massage robot that can move with human-like flexibility and delicacy was develop using a five-fingered robotic hand that mimicking a human hand. Moreover, the aim of this research is to develop a method of online estimation of a user's fatigue based on the muscle hardness in order to evaluate the effect of massaging from tactile information of the robot hand finger-tips to perform appropriate treatment. In this paper, we constructed a system about to measure muscle hardness based on reaction forces and measuring changes in muscle hardness and subjective fatigue over time before and after exercise. Then, the experiment of measuring muscle fatigue was conducted. As a result, the correlation coefficient between changes in muscle hardness and subjective fatigue highly correlated with five of the seven experimental participants. The result suggests the possibility that changes in the reaction force of muscles can be used as a measure for muscle fatigue recovery.

In this paper, we examine the effect of wrinkling and bending generated in coils when used in wearable devices and investigate how they affect the wireless power supply system. A wrinkle is modeled as a semi-cylindrical deformation in a coil, and a bend is modeled as the deformation of an entire coil plane. The effect of these deformations w evaluated by simulation. The result showed that the maximum efficiency is decreased by both deformations, and wrinkling decreases average efficiency more than bending. In addition, efficiency decreases more from multiple wrinkles than from one bigger wrinkle.

This paper presents a sensor node designed to measure the tension of a bolted joint for the purpose of unattended structural health monitoring. The electronic circuits are embedded inside a custom designed stainless steel washer. LoRa is used for wireless communication and data transfer between the sensor node and the base station. The proposed design eliminates the necessity of using specialized bolts with mechanical modifications to monitor tension making it an easy add-on to current and future mechanical structures. Design of the prototype is discussed and evaluated by means of simulation to confirm the proper functionality and adherence to the set requirements.

In this study, with the aim of improving bystander CPR support using a smartphone and a smartwatch, we evaluated six feedback methods considering practical situations. Since CPR is sometimes required to be conducted in a noisy place, each method was evaluated with 50dB and 80dB noise environments, which correspond to a silent office and a noisy construction site, respectively. Also, considering the requirement for a bystander to maintain the safety of him/herself and in order to give appropriate care to the patient, the capability of noticing change in patient condition during CPR was evaluated. From the evaluation results, the best feedback method is a method that uses voice, a metronome sound and a graphic display on a smartphone and vibration and graphic display on a smartwatch if both a smartphone and a smartwatch are available. Interestingly, if only a smartphone is used, the result shows that feedback using only voice is better than the one using voice and clicking sounds. These results are the same no matter whether the noise is loud or quiet. Moreover, with regard to the subjective feeling, feedback using only a smartwatch is worse than other methods, and it is recommended to use in conjunction with a smartphone.

A pointing gesture has an important role in human-human interaction. From our observation, we predict a pointing target before finishing the gesture in human-human communication. This paper, thus, proposes a method for a robot to predict human pointing gesture using the minimum-jerk model. Analytically, the final position of a wrist can be obtained by detecting the first acceleration peak, which corresponds to first 21\% of the entire gesture. We implemented and evaluated the method with a desktop size robot named Robovie-W. The result showed that our method improves naturalness and smoothness.

Activity recognition using multiple body-worn sensors can directly monitor the movement of each body part and can recognize various activities accurately. However, using multiple sensors increases the chance of sensor failure or communication failure, and most current activity recognition algorithms do not work when failure occurs due to the difference (reduction) of the dimension of the feature vector from that of complete sensor data expected in system design time. Therefore, we compared three possible techniques to solves this problem on the feature value level: a classifier trained with reduced feature values, feature value compensation with multiple regression, and feature value compensation with kernel regression, in a no failure situation. All of these techniques do not depend on classification algorithms. While creating a regression model, which is in the training phase, requires relatively high computational power, compensation itself can work with low computational power. As overall results, kernel regression had the best performance that was the closest to the no failure situation. Also, the results imply that each sensor position has its own effective method and more accurate coping can be viable with the appropriate choice of the method.

This paper proposes a direct anthropomorphization method to improve interaction between human and an agent. In this method, an artifact is converted into an agent by attaching humanoid parts to it. There have been many studies that have provided valu

A method has been developed to improve response time of accelerometer based activity recognition. Our method firstly detects a point of change in activity from sensor data and then applies a classifier trained with short window data immediately after the point. Results of experiments show that our method achieves a response time more than one second faster than that of a conventional method.

In this study, we investigated the boundary for recognizing robots. Many anthropomorphic robots are used for interactions with users. These robots show various body forms and appearances, which are recognized by their users. This ability to recognize a variety of robotic appearances suggests that a user can recognize a wide range of imaginary body forms compared with the native human appearance. We attempted to determine the boundary for the recognition of robot appearances. On the basis of our previous studies, we hypothesized that the discrimination of robot appearances depends of the order of the parts. If the body parts of a robot are placed in order from top to bottom, the user can recognize the assembly as a robot body. We performed a human-robot experiment in which we compared the results for robots with ordered parts with those for robots with inverted parts. The result showed that the users' perception of the robot's body differed between the two groups. This result confirms our hypothesized boundary for the recognition of robot appearances.

A system has been developed that can project an embodied agent's image and sound anywhere in a room. It can thus overcome the problems inherent to other embodied agents in dealing with 2D on-screen information and 3D physical information simultaneously. An experiment demonstrated that this `PROT' agent can effectively present both on-screen information and real-world physical information. Because the PROT agent combines the advantages of a robot agent with those of an on-screen agent, it should improve human-robot interaction.

In this article, we propose a model for the addition of user information to sensor data. The problem is that it is difficult to identify a user of objects that are used by multiple people and to define a unique way to identify the user. The proposed model identifies the user assuming that the objects are used by multiple people. Further, it introduces three policies. By our experiment, we have proved the following: (1) Addition of user information to sensor data by the model is valid. (2) By presenting a difference in policies, it is possible to draw someone's attention.

With this video, we propose a new human-robot interaction that anthropomorphizes a target common object and transform it into a communicative agent using attachable humanoid parts. The user perceives the target to have its own intentions and body image through the attached body parts. This video shows examples of anthropomorphization method as below.

In this paper, we propose new robotics presentation method called, Self Introducing Poster that uses attachable humanoid parts and explains its contents through a self introduction style. Presentation by a conventional robot sometimes fails because the robot presenter is often too attractive and distracts from the presentation itself. In our method, the poster is anthropomorphized and explains its contents. Due to this self presentation, users can more easily understand its meaning because the information's contents and information provider are strongly related. We designed and implemented our system and evaluated it in the field. The results suggest that the self-introducing system is useful for gaining users attention and effectively presenting information.

We have constructed a sensor network in a real hospital environment to develop a system that prevents medical accidents by monitoring nursing activities. The network has been carefully designed to be dependable for practical purposes even in a network-unready environment with considerations of safety and stability as well as the consistency of the obtained sensor data. This paper describes the design and implementation of our sensor network. Through an experiment of about one week, we confirmed that our design performs well in a practical environment and obtains consistent data among the sensors worn by nurses and installed in the environment. Since a hospital has strict limitations on the use of sensor network equipment, the design described in this paper provides a practical solution for the construction of sensor networks in many indoor applications

This paper proposes a scheme to recover the state of peripheral devices in non-volatile main memory systems so that the system resumes its execution after an unpredictable power failure. Our scheme is software-based approach and accomplishes the recoverable system without complex hardware equipments. First, the requirements for maintaining consistency are discussed based on the concepts of distributed message-passing systems. Next, the application to a device driver is illustrated using an example. Experiments with UART and network devices showed that the system restarted properly after a power failure, and that system performance was barely decreased.