This study is the first to explore the interplay between haptic interaction and avatar representation in Shared Virtual Environments (SVEs). Specifically, how these factors shape users' sense of social presence during dyadic collaborations, while assessing potential effects on task performance. In a series of experiments, participants performed the collaborative task with haptic interaction under four avatar representation conditions: avatars of both participant and partner were displayed, only the participant's avatar was displayed, only the partner's avatar was displayed, and no avatars were displayed. The study finds that avatar representation, especially of the partner, significantly enhances the perception of social presence, which haptic interaction alone does not fully achieve. However, neither the presence nor the type of avatar representation impacts the task performance or participants' force effort of the task, suggesting that haptic interaction provides sufficient interaction cues for the execution of the task. These results underscore the significance of integrating both visual and haptic modalities to optimize remote collaboration experiences in virtual environments, ensuring effective communication and a strong sense of social presence.

This study aims to clarify the characteristics of human learning processes and to examine the requirements for a trainee model that can reproduce them. The results of a Human– Human teaching experiment suggest that human learning processes consist of two phases: an initial phase where performance improves significantly with large variability, and a later phase where performance improvements slow down and variability converges. However, the proposed trainee model exhibited issues such as instability in early learning and performance degradation, despite being able to reproduce the learning progress. To address these problems, a learning rate scheduler that dynamically adjusts the learning rate according to the teaching force was introduced. This suppressed the instability in early learning and mitigated the performance degradation. These results suggest that the characteristics of human learning processes can be reproduced to some extent. In the future, it will be necessary to introduce probabilistic mechanisms such as Bayesian networks and develop dynamic parameter determination methods.

This paper presents the Environment-Adaptive Vibrotactile Transformation (EAVT), a novel method to enhance the realism of vibrotactile feedback in haptic rendering. Traditional texture modeling often faces discrepancies due to differences in device dynamics. EAVT addresses this by estimating and eliminating the transfer function of the recording device, thereby adapting vibrations to the rendering environment. This ensures that perceived vibrations match actual interaction forces. The method involves identifying the transfer function using test vibrations and estimating interaction forces from recorded accelerations. Simulations demonstrate high fidelity in reproducing these forces, ensuring consistent vibrotactile feedback across different devices. EAVT significantly improves texture realism and applicability, providing users with an accurate haptic experience regardless of the recording tools used.

Co-presence is the index that indicates the degree to which the observer believes he/she is not alone and secluded. The representation of avatars in virtual environments (VE) having an effect on co-presence was reported. However, these reports are only limited to visual interaction. This paper aims to clarify the influence of the avatar representation on cooperative haptic interaction. Haptic interaction in the VE is essential for teleoperation. A cooperative task with multilateral control was conducted, and the haptic interaction was analyzed by conducting a survey. In the experiment, the avatar representation was changed by three patterns of the avatar representation as follows; all of subjects’ avatars are displayed, only the subject’s avatar is displayed, and all of subjects’ avatar is not displayed. The experimental results showed that the avatar representation did not affect the haptic interaction. On the other hand, the subject’s impression was changed by the avatar representation. Therefore, it was concluded that the avatars are not always necessary in a cooperative task.

This paper proposes a real time objective Human-Human teamwork assist method with physical interaction. In this research, in order to evaluate and assist teamwork objectively in real time, we focused on changes in the input command that occur when cooperate with others and defined changes in the amount of input command as CFO. CFO can be estimated by input command in real time. CFO is estimated by learning the input command during the solo task using Neural Network and comparing with the input command during the cooperative task. This paper proposed an assist system with force feedback for teamwork assist using CFO. Then, in order to verify the proposed assist is effective, experiment performed to higher teamwork pairs and lower teamwork pairs on w/o assist and w/ assist. The experimental results suggested on higher teamwork pairs marginally significant difference to mean square error by the assist. The results suggest that real time teamwork assistance by CFO is possible.

“Concern for Others: CFO” which is an element of evaluating teamwork in real time. In most of the studies, the teamwork evaluation was analyzed by using questionnaire after an experiment. When teamwork skill requires assisting, those methods cannot apply in real time and objectively. Therefore, we had been suggested a method to real-time analyzing objective teamwork support by the CFO. In previous research, we revealed the sum and subtract of CFO were affected by force interaction when the cooperative task. Then, in this research, we analyzed temporal changes in an operator for the cooperative task using the sum and subtract of CFO, for considering a method for assisting the teamwork in real time by force feedback, visual assist, etc. As a result, a positive correlation was seen between the cooperative task time and the sum and subtract of CFO. And each section shows to tend of increase one of the operator's input change with time by an intervention of the others input. These tendencies suggests the teamwork can be assisted by increasing the sum of CFO referring to the subtract of CFO.

Nowadays, studies on human-robot cooperation have been focused on because robots are expected to work among human society. Although one of important topics of human-robot cooperation is how to assist humans, contrived assist motion may make the human, in front of the robot, be bothering. Human beings often consider attention of other one for smooth communication in daily life. Estimating attention is therefore one of important factors for adequate communication. If robots detected human attention, a novel communication scheme using the artificial misdirection would be established. This paper focuses on the misdirection, which is a famous technique for magicians to distract its attention to achieve tricks. The conventional methods to estimate the human visual attention were mainly by Point of Gaze: PoG. However, few studies attempted to consider what the human did NOT make attention. The purpose of this research is to establish the Visual Attention Distribution Map: VADM considering depth of attention. As a first step of the artificial misdirection, validity of proposed VADM is verified by visual memory and dynamic spot the difference tasks. From these results, the VADM can be utilized to control spot the difference tasks.

Nowadays, Human-Machine Systems: HMS has been focused on. In the HMS, a human skill is one of key factors for human machine cooperation. In this paper, a novel human skill evaluation method with the tracking error attractor analysis in a tracking operation task via a joystick is investigated. There are several approaches to evaluate human skills, for example, tracking error measurement or human input characteristics. The proposed skill evaluation method belongs to the human input characteristics. The tracking error attractor consists of proportional and differential tracking error. Then, an estimation index on which proportional or differential tracking error is concentrated by the operator. In the tasks, subliminal calibration, modifying bar dynamics without awareness, and subliminal input filtering, mixing the input command with a human model by neural networks, are applied. The subliminal calibration employed tracking evaluation value, consists of proportional and differential tracking error, to estimate human internal model error. Here, the weight parameters of these two errors are utilize the human skill characteristics. Thus we investigates the attractor by proportional and differential tracking error. Finally, the effectiveness of the subliminal calibration and the subliminal input filtering is discussed by command input characteristics focused on the error attractor.

Lots of researches on Human-Machine Systems (HMS) have been investigated to improve the task performance with human-machine cooperation. Most of the works have focused on a single operator as enhancing individual skill. Thus, few studies on teamwork assist for cooperative tasks by multiple human-beings. In our previous works, to realize a teamwork assist system, quantification technique of "Concern For Others: CFO," which was a key factor of cooperative performance of such cooperative tasks, was proposed. The CFO is defined as difference between the command input in a cooperative target task and predicted input by the human maneuver model in the solo task. The human maneuver model in the solo task is learned by prior solo experiment as a calibration. Since the maneuver model is very important to calculate the CFO, precise and efficient learning method would be required. Therefore, in this paper, a efficient learning technique, with recurrent neural networks and prediction based filtering, for the solo task model is proposed. Finally, the prediction performance are experimented, and validity of proposed method is discussed.

This paper addresses a quantification technique of cooperative characteristics on Concern For Others: CFO in a cooperative work by multiple operators. The cooperative works are generally required suitable cooperation to accomplish the task adequately. Although the “cooperation” is one of keywords for social psychological field, few study has been investigated for engineering application, that is, quantification of the cooperative characteristics is not focused on. Thus, most of researches on human-machine systems have been considered to assist a single operator. In that case, human intention concentrates upon only faced robot operation. Measured input command by the operator includes all of operator's intention. However, in cooperative works, measured input probably contains the effect from other operators, that is CFO. A challenge of this study is to quantify the cooperative characteristics as the CFO in cooperative tasks. Then CFO utilizes to realize a novel assist system to enhance their cooperation. To obtain CFO in such tasks, the human input pattern during a solo work was learned by human model before the cooperative task. Then, the difference between measured input command in cooperative work and the predicted command in the human model is defined as the CFO. Finally, this paper addresses the correlation analysis between the CFO and cooperative task performance.

This paper addresses a quantification technique of cooperative characteristics focusing on Concern For Others: CFO in a cooperative work by multiple participants. The CFO is utilized for objective evaluation of human altruism. We have two goals of this research are to realize a mediate assist system for cooperative tasks and to develop a training tool of social skill for persons with autism. Most of studies on human-machine systems have been considered to assist a single operator. Here, we focused on a cooperative task by multiple participants. In this case, the command input of each operator includes CFO to achieve suitable cooperation. The CFO, however, may bring about misreading of the player's intention to give a suitable assist for the system. Thus the COF is expected to work as a filter to obtain operators characteristics accurately in such tasks. On the other hands, if the CFO can be obtained by the system, a new assist scheme to help their cooperation as a mediator will be invented. As a first step of such assist systems, a cooperative work platform employing haptic interface without force sensors is developed. The cooperative task is very simple and intuitive as controlling a ball on the plate whose edges are held by multiple humans. To obtain CFO in this task, the human input pattern during a solo work was learned by CMAC before the cooperative task. Then, the difference between an input command in cooperative work and the predicted command in the solo one is defined as the CFO. Finally, this paper addresses the technique to obtain CFO and the applications with it is discussed.

Analyzer of conversation is effective to estimate a person's intention and to predict the person's actions, and such function of the analyzer can be utilized for watch-over system. It is, however, difficult to make such perfect analyzer with a simple computer system. Hence in this study, we focused on change in utilization of corpus by limiting the situation of human interaction. We investigated construction of common grounds and sharing ontology based on action words using utterances occurred in the skill acquisition of virtual collaborative conveyer task. Specifically, we constructed corpus and analyzed conversation between three adults that occurred while completing 10 task trials. We extracted action words from the conversation and classified them into three categories, Move, Stop, and Turn. Action words were defined as the word directly related to transportation task. The results were that the number of action word types decreased in all action word categories and in all experimental groups, showing that the ontology of action concepts converged into a few words in each category. The results also showed that the construction of common grounds among the participants seems to contribute to the convergence of action ontology.

Based on a new concept of Kinder-GuaRdian which aims to realize a support system for kindergarten's staffs and parents by watching over children's activity and their growing process, activity recognition (AR) for children is treated in this paper. In case of children, smaller amount of sensors is better since easy fitting of the sensor device to child's body is strongly preferable. In this study, using one acceleration sensor attached to the upper arm, AR in cases of adults and children were evaluated using several orthodox classifiers and self-organizing map (SOM). Through the several types of validation, effectiveness, issues, and benefits of their methods are discussed.

To realize interactive robots that watch for humans, it is necessary for robots to keep distance adequately from humans. The robot must keep away from the person to monitor for avoiding unnecessary interference, while the robot must approach to the person for active communication or supportive actions. Thus, in addition to that distance between the robot and the person must be properly controlled, the human intention of either “I act, without your help” or “You act, help me” must be specified. Demonstrative words such as “this” and “that” may be used as one way to control this distance information and understand human desire for support. We investigated the relationship between distance and demonstrative words when a speaker utters demonstrative words to a hearer in cooperative carrying task using virtual space. We used 3 corpus of speech to extract demonstrative words, then we calculated the distance between the speaker's and the hearer's position to the object to be moved. Results indicated 1) Demonstrative “Kore (This)” was used when the speaker's position was near to the objects. 2) Demonstrative “Sore (That-proximal)” was used when the hearer's position was near to the objects. 3) Demonstrative “Are (That-distal)” was not observed. It was also suggested that the aspect of distance (proximal/distal) was not the key factor of the use of Japanese demonstratives in virtual space. “Kore” was often used for objects that the speaker oneself wanted to act on, while “Sore” was often used for objects that the speaker wanted the other person (who was close to the object) act on. Whether the speaker himself wanted to act or he wanted the hearer act was the key factor of the use of these demonstratives

To accomplish the objective to make human-collaborative robots, we need to clarify how humans actually interact each other when they do collaborative work. In this study, we transcribed all utterances produced while participants' completing human-human collaborative conveyer task, and computed and categorized all morphemes (minimal unit of language meaning) using 4 categories based on the morpheme's role in the task. The role categories were Robot Action, User Action, Modifier, Object. We analyzed the utterances produced by 4 groups, 3 participants in each group. Results were that frequency of each category per minute decreased over ten trials. However, the variety of words in each category tended to show an inverted U-shaped pattern. Based on these results, we proposed three stages of language skill acquisition in a collaborative work.

Importance of developing human adaptive robots or systems is increasing these days. To accomplish it, we have to clarify features of human-human communication. In this study, we analyzed human speech while completing computerized collaborate task to clarify how humans speak in collaborative work. We extracted questioning speeches from conversations using CLAN, and classified them into four categories according to their content. We investigated whether the number and ratio of each type of questions changed over ten trials. The results indicate that leader role person is sensitive to task planning

This paper addresses evaluation methods for social cooperation characteristics in cooperative tasks. Using these we found that the ratio of four typical indexes relate to task performance. Most of conventional human-machine systems assumed to assist only a single operator. In such cases, the system should only pay attention to his/her intention or operation characteristics. If there were multiple operators they also included social behaviours affected from others. Such behaviours could not be regarded by the one-on-one assist systems. Since intelligent assist systems are expected to apply in human society, the system requires work among humans. A challenge of our research is to quantify the human social characteristics, especially when relating to cooperation in the task involving multiple operators. In this paper, human social characteristics are quantified in order to evaluate cooperative performance. For the evaluations, ratio indexes by four kinds of behaviours are focused on. One is observation behaviour for other operator's work, which is typical social behaviour in a cooperative task. Second is the ratio of approaching to other robots for physical supporting. Third, total activity by summation of robot motion distance. Then, an index of visibility of robots each other. Finally, correlation between expansion of the index distribution and a task completion time is analyzed. Furthermore advanced assist using the evaluation is discussed.

A group of participants performed a collaborative conveyer task that needs to transfer objects to designated positions by vehicles in a virtual space. All utterances were transcribed and questioning utterances were specified based on rising intonation patterns and content of the speech. Then we classified these utterances into four types: planning (ask whole transporting strategy in task), partner's (asking the movement and status of the partners), self (asking own movement and status), and other, and analyzed the changes of the frequency of these question types during 10 trials. We also constructed a collaboration levels and evaluated collaboration status for each completion of a box conveying. This collaboration level was determined by the number of people who collaborated to move one box during one trial. We investigated the relationship between strategy changes, collaboration levels, and question utterances. The results were that in a group that showed a strategy change, collaboration level was generally higher and more varied than the collaboration level observed in the group that showed no strategy change. In addition, in a group that showed a strategy change, question utterances about whole plan decreased but question utterances about the self status increased. However, in a group who showed no strategy change, all types of question utterances decreased. It was found that when a strategy change occurred, people tend to make more questions about self status to confirm the details of movement to complete the task. Collaboration level changes can be a good index to infer occurrences of strategy changes.

This paper addresses a quantification method for role sharing in cooperative tasks. By the method, we found that the ratio of three typical indexes relate to task performance. Most of conventional human-machine systems assumed to assist single operator. In such case, the systems should only pay attention his/her operation characteristics. If there were multiple operators, they also include altruistic behaviours during the work. Such behaviours could not be regarded by such one-on-one assist system. Since intelligent assist systems are expected to apply in human society, the systems are required work among humans. A challenge of our research is to quantify such humans' altruistic behaviours, especially what relating cooperation in the task involving multiple participants. In this paper, role sharing characteristics are quantified in order to evaluate the cooperative performance. For this quantification, ratio indexes of three kinds of behaviours is focused on. One is observation behaviours for other operator's work, which is typical altruistic behaviour in a cooperative task. Second is ratio of egoistic behaviour as active conveyance. Then, total activity by motion distance of the robot could be indexes of the role sharing. Finally, correlation between these ratio of indexes and a task performance is analyzed. Furthermore an advanced assist using the evaluation is discussed.

A purpose of our research is to realize teleoperation assistance to improve the performance with operator’s initiative. One of advantages of teleoperation system is utilized human abilities which are global recognition, planning, prediction and so on. However, the operator is required to get skills for suitable operation, and assistance from the system is important. Although effects of the operation assistance are generally contemplated by system designers, it is not always available because of depending on variable operator’s characteristics and environment condition. Solving this problem, adaptive assistance is required, however, autonomous robot motion may disturb to give an initiative to human operator, that is, human abilities cannot be brought out. Therefore, assistance by modification of robot dynamics parameters has been proposed. Because human operators tries to adapt and learn the robot dynamics instinctively, such assistance often brings about discomfort and less maneuverability. Consequently, human awareness characteristics are considered to improve this problem an assistance technique without human perception using a cognitive science approach is addressed. In this paper, new assist technique without discomfort feelings of operators is proposed.

To realize new kinds of human-machine system proposed as Human Adaptive Mechatronics(HAM), experimental analyses to elucidate human’s skill is reported in this paper. For skilled operation, we has been thinking that an adequate self-switching of reference targets on the manipulation is important. To prove this idea, we designed a special task by using virtual computer graphics of a hovercraft, measured gaze behavior of the operators, and analyzed their skill that seems to be related with the switching references. From the analyses of behavior of the skilled operator, it was confirmed that early switching of sub-controllers / reference signals plays a significant role in skill.

This paper addresses a real time human sensitivity estimation technique on GUI for human adaptive teleoperation system. Although human operators have valuable abilities as global recognition, prediction and planning, they also have possibility to make mistakes. Especially, an overlooking among lots of information, is serious. In order to solve this problem, adequate GUI display planning for emphasising important feedback information is required. We therefore attempt to consider human perceptional characteristics. In other words, robustness for human individuality, skills, fatigue an important alert information window is put with perceptive properties consisted of colors, size, frame line width and so on. At the same time, unimportant optional windows put without disturbing the alert notice. However, the sensitivity depends on operator's individuality and GUI circumstances, namely, it is variable. Unfortunately, human operators often make careless mistakes that is not essential for the sensitivity. To evaluate the variable and including unessential mistakes sensitivity, in this paper, a real-time human sensitivity identification technique with statistical method is proposed and experimented.

Most of studies on path planning assume a wheeled robot as the vehicle. However, the robot is allowed to only make a detour for obstacle avoidance that often includes inefficiency. This research focuses that a walking robot has a high adaptation on irregular terrain and ability to transform the robot posture during walk. Therefore, the walking robot follows an efficient path in a three-dimensional complex environment in which a wheeled robot cannot do. In order to obtain an efficient path, we notice that the appearance of obstacles depends on the robot's body height and lateral foot breadth. For example, the walking robot avoids under overhead obstacles by crouching and pass as if there are no low level obstacles. When crouching, the robot should react to low-level obstacles as if they are high-level obstacles because the maximum lift of robot's foot is restricted by the crouch position. At the same time, low obstacles should be taken up as though higher obstacle because maximum lift of foot is restricted by the crouching posture. By such property, virtual obstacle, which is defined depended on robot posture, is defined. This paper presents path and posture planning method with the virtual obstacle. Finally, the method is verified by some experiments.

A legged walking robot is able to not only move on irregular terrain but also change its posture. For example, the robot can pass under above head obstacles with crouching. The purpose of this research is to realize efficient path planning with a quadruped robot. Therefore, the path planning is expected to extended in three dimensions because of the mobility of the quadruped robot. Even tough these advantages, some issues of the quadruped robot that are instability and workspace limitation complicate realizing the application. Especially, a deadlock is serious problem because it causes the walk stop and less controllability. The deadlock is brought by concurrence of instability and workspace limitation. In order to improve the deadlock, we propose the new free gait pattern for a quadruped robot, called TFG (trajectory following gait). The TFG intends to obtain high controllability like a wheel robot without the deadlock. The TFG additionally allows changing posture, during the walk, which is one of merits of a walking robot. In this paper, experimental result shows that the TFG improves the problems of a quadruped robot and is valuable for three dimensional path planning.