Current myoelectric hands have two problems: instability in performance due to low estimation accuracy and individual differences, and deviation from the natural gripping approach of the hand, such as the need to preset operations and strain for stable movement. These problems are thought to be due to the nature of electromyography (EMG); therefore, it is necessary to introduce a new multimodal sensing method that uses control inputs other than EMG. Existing research on multimodal sensing has focused on improving performance rather than replicating the gripping approach of the hand. In order to solve the above two problems, in this study, we developed a system that uses both EMG and distance/infrared images as control inputs, and compared the estimation rate and gripping rate with an existing method that uses only EMG in 10 healthy subjects. As a result, the average estimation rate and gripping rate were improved by about 1.5 times compared to the existing method. The standard deviation was also reduced from 19 to 4.1 for the estimation rate and from 17 to 4.3 for the gripping rate, and individual differences were also reduced.

Currently, myoelectric prosthetic hands are not recognized as part of the body due to the lack of sensory feedback (FB). To address this issue, it is necessary to investigate the factors that influence body recognition. Most existing research focuses on stationary prosthetic hands, such as in the rubber hand illusion, and discusses two concepts: sense of ownership (SO) and sense of agency (SA). SO refers to the feeling that a body part belongs to one's own body, while SA refers to the feeling that one is in control of the movements of one's own body parts. In this study, we developed a wearable and operable prototype myoelectric prosthetic hand equipped with tactile and deep sensory feedback, rather than a stationary prosthetic hand. Furthermore, we investigated the effect of tactile and deep sensory feedback on body recognition through psychophysical experiments. The results indicated that tactile feedback improved body recognition and deep sensory feedback improved SO; however, no effect was observed on SA.

In laparoscopic surgery, large organs are difficult to handle with one forceps, resulting in low surgical efficiency. In this study, to improve surgical efficiency, we developed a five-fingered robotic forceps that can be inserted through small incisions and has multiple degrees of freedom to handle large organs. The results of an experiment simulating sigmoidectomy showed that the proposed robotic forceps could shorten the surgical time. In addition, we confirmed that the physical burden while using the instrument was comparable to that of conventional forceps.

This study aims to develop a robotic hand that can be inserted into the body from a small incision wound and can handle large organs in a laparoscopic surgery. We determined the requirements for the proposed hand based on a surgeon's motions in Hand-assisted laparoscopic surgery(HALS). We identified four basic motions: “grasp,” “pinch,” “exclusion,” and “spread.” The proposed hand has the necessary degree of freedom(DoFs) for performing these movements, five fingers, as in a human's hand, and a palm that can be folded into a bellows when the surgeon inserts the hand into the abdominal cavity. We evaluated the proposed robot hand based on a performance test, and confirmed that it can be inserted from a 20 mm incision wound and grasp the simulated organs.

In this study, we propose a system that detects falls and estimates walking conditions for elderly people living alone. This system uses floor vibration measured by microphone sensors installed on the floor of an elderly person's home, and it detects falls and walking from the frequency characteristic of the floor vibration. Furthermore, it estimates the walking condition using a method of estimating the sound sources with multiple microphone sensors. To demonstrate the usefulness of the proposed system, an experiment of fall / walk detection with one healthy person and a walking condition estimation experiment were conducted in a model room in a facility for the elderly in Fukui Prefecture. As a result, all falls and walks under arbitrary conditions were able to be detected, and the estimation of the walking state was able to be partially achieved.

Numerous robotic devices have been developed to assist hand rehabilitation; however, a majority of these are difficult for stroke survivors to wear. The purpose of this study was to develop an assistive device for treating flexion contracture, which supports the extension of each finger and may easily be worn on a paralyzed hand. To facilitate ease of use, we suggested a new wearing method for this wire-driven device with an elastic skeleton, allowing users to extend the device from the back of the hand onto the fingertip. The functional capacity of this device was measured through fingertip contact force and estimations of supporting torque. Results showed the device provides sufficient torque for finger extension with controlled wire tension. Moreover, experimental results confirmed that the novel design significantly decreased the time it took users to don the device compared to other designs.

This work reports research and development of a lightweight neuroprosthesis, which can control the impaired motion by using voluntary biological signal. The total weight of the upper limb neuroprosthesis is 900 g, which is 40% lesser than the commercially available ones. For a trans-humeral amputee who had targeted muscle reinnervation (TMR) surgery, pattern classification of five motions was possible by using surface electromyogram (s-EMG) extracted from four dry electrodes.

A myoelectric prosthesis, which can be used as a replacement for a person's upper limb, is able to control many hand motions optionally via surface electromyography (sEMG). Although many researchers have developed myoelectric prostheses that realize various motions, evaluation of their practicality has only been carried out on small numbers of patients. To solve this problem, a myoelectric prosthesis simulator that is able to move like a real prosthesis is needed. In this study, such a simulator that enables persons to grasp virtual objects with virtual prostheses was developed using Augmented Reality (AR). The results of comparative experiments conducted to evaluate the developed system using a pick and place task show that although it is not at present as effective as real prostheses, a virtual prosthesis is able to grasp virtual objects in the proposed simulator.

The purpose of this study is to investigate how sensory feedback affects the cognitive strain of grasping using a prosthesis and to develop an efficient method for decreasing the cognitive strain of grasping. We divided a grasping action into two phases, an “approaching phase” and a “grasping phase”, and assumed that the tactile feedback affect for the cognitive strain induced by grasping phase and the deep sensory feedback affect for cognitive strain induced by approaching phase. Using dual-task method, we compared the effect of sensory feedback method using strength-changed simulator or using spatially changed stimulators. As a result, we concluded that the sensory feedback can decrease the cognitive strain of grasping action, and the sensory feedback from one strength-changed stimulator is better than the one from spatially changed stimulators to decrease a cognitive strain caused by grasping action.

For use as five-finger myoelectric hands in daily living, robotic hands must be 1) lightweight and human sized as well as possess 2) human-like multiple degrees of freedom (DoF) and a high grip force. The compatibility of these requirements is a trade-off; ideal robotic hands have yet to sufficiently satisfy both these requirements. Herein a power allocation mechanism is proposed to improve the grip force without increasing the size or weight of robotic hands using redundant DoF during pinching motions. Additionally, this mechanism is applied to an actual five-finger myoelectric hand, which can make seven types of motions that are necessary for activities of daily living (ADL) and realizes about a 60% improvement in fingertip force, allowing three fingers to pinch objects exceeding 1 kg.

Although a cellular manufacturing system can produce diversified products flexibly, its assembly efficiency is mainly limited by the operators' assembly performance. To improve the operators' assembly performance without longtime training, an assembly-skill-transferring method was proposed to realize the assembly skill transfer process from skilled operators to novice operators. In the cellular manufacturing system, the assembly skills consist of cognition skills and motor skills. Taking a “Peg-in-Hole Task” as an example, the effect of the proposed assembly transfer method was verified for motor skills. The results show that the proposed assembly-skill-transferring system can improve the novice operators' assembly performance greatly.

A cellular manufacturing system is good at producing diversified products flexibly; however, its assembly efficiency depends mainly on its operators' abilities. As the workforce shrinks in Japan, cellular manufacturing systems are difficult to maintain. In this case, a new assembly system has been developed since 2006 that combines both the dexterities of human operators and the advantages of automatic machinery. Its characteristics consist of three aspects: collaboration between an operator and twin manipulators on a mobile base, assembly information guidance, and safe design for collaboration. To meet the rapid changing tastes of customers, operators have to assemble various products without longtime training. This requires an effective assembly skill transfer system to extract assembly skills from skilled operators, and then transfer them to novice ones. Considering the characteristics of a cellular manufacturing system, an assembly skill transfer system was proposed and used to extract and transfer assembly skills in both cognition and execution aspects. Taking a cable harness task as an example, the proposed skill transfer system was applied in a developed assembly system. The results show that working under the developed assembly system with physical support, informational support, and the assembly skill transfer system, novice operators' assembly performance was greatly improved. This verified the effect of the proposed solution to maintain the cellular manufacturing system in the aging Japanese society.

To classify the five individual finger motions from an electromyogram (EMG) signal, a classification system that hybridizes EMG signals in both the transient and converged states of a motion is proposed. The classifications of finger motions are executed individually in each state by a well-established artificial neural network (ANN). Then, the outputs of the two classifiers are combined. The efficacy of the result is evaluated via a piano-tapping task, in which the subjects are instructed to tap a keyboard with each of their five fingers. We use this task to compare the proposed hybrid system and a conventional converged system that uses an EMG signal only in the converged state. For five of the six subjects, the accuracy ratio of finger motions was better in the proposed method: approximately 85% for each finger except the second. Further analysis suggests two remarkable advantages of the hybrid method: (1) the output of the ANN is more credible, and (2) finger motion in the transient state (i.e., the early phase) is more predictable.

To improve productivity of assembly in cell production, human robot collaboration is introduced. In the collaboration, robot manipulator supports human operators by handing over an assembly part in the assembly process. However, when the human operator works coexisitently with the moving robot in close distance, it might induce high stress to the operator. This paper studies on strain measurement caused by the hand-over motion of an industrial robot, and discuss design criterions of the robot arm motion when working with a human operator. Several basic stresses are experimentally measured: motion speed at robot's movement, motion trajectory and so on. Experimental results indicated following two points. First, low motion speed (less than 500[mm/s]) or non-straight motion is preferred to design hand over motion which induce low mental strain. Second, it is recommended for a human operator to be notified of robot motion speed before it moves to avoid mental strain.

Even if a cellular manufacturing system is robot assisted with assembly support, the main operator is human. A system designer cannot therefore ignore the mental strain experienced by the operator as a result of improper assembly support. In this paper, authors analyzed operators' mental strain induced by physical and informational support in an advanced cellular manufacturing system with human-robot collaboration using physiological parameters related to cognition and emotion. Our results showed that higher speeds of an approaching robot's motion and closer distances between human operator and robot increase mental strain related to emotion, while augmenting text instruction with voice guidance and optical instruction increase mental strain related to cognition and emotion.

Long-time exposure to repetitive or monotonous work is associated with increased risk for musculoskeletal disorders that are due to muscle fatigue. Previously, researchers reported that muscle fatigue can be estimated using a low-frequency band of an SEMG signal. However, these studies were performed exclusively during static contraction tasks. The objective of the present study was to evaluate and determine the frequency range for a low-frequency band. In addition, the performance during dynamic contraction tasks was analyzed. A group of healthy university students (8 males) was recruited, and endurance handgrip tasks were conducted. SEMG signals were captured from the forearm muscle. The frequency range for the low-frequency band was redefined as 5 - 40 Hz. The results from a dynamic contraction task indicated that a low-frequency band is a reliable method for indexing muscle fatigue from SEMG signals.

The objective of this work is to enhance man-machine collaboration in cell production by task modeling approach. Task modeling provides a method to define and study the collaboration between man and machine. Six requirements have been derived as basis of development. In this work, the task modeling has extended classical HTA method to represent collaborative operation in a hierarchical structure with the classification of operation task levels and extended operation details as task properties. A modeling tool is developed to provide a development and data management platform. The man-machine collaboration is well defined in a structural format by the ability of task analysis method to address human task in a more flexible way. The collaboration modeling and operation resources information have significantly facilitated operation planning in both assembly and control levels. From the case study, the improvement in safety aspect has proven the ability of this approach in assisting safety design and development of production system. The integration with operation control system has shown the execution potential in real-time operation. The unique development of operation data management has expanded this work into operation information support. Through the multimodal information support system, the human operator is well guided by the operation information corresponded to the task components. The operation information also can work as evaluation functions to measure the human working performance and system performance as a whole.

The challenge of this work is to study the design and development of human-robot collaboration (HRC) in cellular manufacturing. Based on the concept of human collaborative design, four main design factors are being identified and developed in an active HRC prototype production cell for cable harness assembly. Human collaborative design aims to optimize the system design for the advantage of collaboration between human and robots based on human considerations. Task modeling approach is developed to study and analyze the task in order to identify the collaboration tasks to develop the collaboration planning. In the collaboration safety development, five safety designs, cover both hardware and control design, are proposed and developed in the prototype system. Risk assessment is conducted to verify the safety design. Two main experiments were conducted as preliminary study to investigate mental workload in HRC. A multimodal information support system is developed in the study of man-machine interface in this work to provide a comprehensive human-robot interface to facilitate human operator. The system performance evaluation had proven the improvement of prototype production cell with HRC design for cellular manufacturing.

The purpose of this work is to conduct human factors studies in the development of an information support system for human-robot collaborative cellular manufacturing system. Multimedia technologies are being utilized to enhance the support system to function as operational information support and interface between human operator and robot system. In the study, five experiments were carried out in both system hardware and support information developments to investigate the human factors design of the support system. In the information display approach study, comparison of information presentation ability among paper manual, LCD TV and projector was conducted. Similar information was displayed in different visual areas by vertical and horizontal LCD TV configurations to study the effect on human visual working area. The comparison experiment on image, text and voice formats had resulted varying modality effectiveness of information. Different information formats require different human cognitive processes even though the information contents are similar. The investigation on multimedia displays in five combinations of information elements was performed in multimedia display design study. The information support system is implemented in a human-robot collaborative prototype production cell. The cable harness assembly experiment had proven the effectiveness of the information support system in guiding the operators on the assembly operation and also to collaborate well with the robot system.

Nowadays, the Man-Machine interfaces are becoming more important in our interaction with robotic systems. One particular case is on prosthetic devices, where the need of "biofeedback" is of vital importance to achieve subconscious control of these devices. In order to provide this feedback, so the device can be included into the user's body image, we use electrical stimulation as a substitute for tactile feedback. In this study we analyze the effects of the electrical stimulation together with the intention, provided by the electromyography, for the sensation generation, and its inclusion on the user's body schema. To evaluate our system, we use functional Magnetic Resonance Imaging, to measure the changes in the brain due to the use of an EMG controlled prosthetic hand with electrical stimulation as tactile feedback.

This paper describes a new real-time learning method for the development of a robust motion discriminating method from an EMG signal, to adjust to the change in user's characteristics. This method is done under the assumptions that the input motions are continuous, and the teaching motions are ambiguous in nature, therefore, automatic addition, elimination and selection of learning data are possible. Applying our proposed method, we conducted experiments to discriminate eight forearm motions, with the results, a stable and highly effective discrimination rate was achieved and maintained even when changes occurred in user's characteristics

We developed a new adaptable EMG prosthetic hand, which executes recognition process and learning process in parallel and can keep up with change in the mapping between an electromyographic signals (EMG) to the desired motion, for amputee. EMG-to-motion classifier which used in proposed prosthetic hand is done under the assumptions that the input motions are continuous, and the teaching motions are ambiguous in nature, therefore, automatic addition, elimination and selection of learning data are possible. Using our proposed prosthetic hand system, we conducted experiments to discriminate eight forearm motions, with the results, a stable and highly effective discrimination rate was achieved and maintained even when changes occurred in the mapping. Moreover, we analyzed mutual adaptation between human and adaptable prosthetic hand using ability test and f-MRI, and clarified each adaptation process

The purpose of our research is to build a system for mutual adaptation between a user and assistive devices for restoration of motor function. To build such system, it is necessary to know human's motion patterns. In this paper, as the first step, we investigated human motion characteristic on human-machine system like EMG (electromyogram) prosthetic hand and EMG to motion classifier system. In the experiment, we measured the EMG signals and investigated a difference between motion patterns of teaching motion, i.e. user's intended motion, and that of actual motion using the proposed criteria. As results, it is clear that these criteria are useful to analyze changes of human motion patterns

The purpose of our research is to build a system for mutual adaptation between a user and assistive devices for restoration of motor function. To build such system, it is necessary to know human's motion patterns. In this paper, as the first step, we investigated human motion characteristic on human-machine system like EMG (electromyogram) prosthetic hand and EMG to motion classifier system. In the experiment, we measured the EMG signals and investigated a difference between motion patterns of teaching motion, i.e. user's intended motion, and that of actual motion using the proposed criteria. As results, it is clear that these criteria are useful to analyze changes of human motion patterns.