Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Vibgyor ePress  

The safety of complex Socio-Technical Systems highly depends on the non-linear interaction of variabilities between task performances in human, machines and working environments and proper management of those is essential to ensure the safety. However, the practical means to ensure the safety of those systems have not been established yet, and this is mainly due to the gap existing between the theory and the reality as well as to the lack of means for visualizing the potential risks. In this paper, we propose an extended method of FRAM: Functional Resonance Analysis Method and construct a simulator that enables to visualize the potential risks. Then, an example of interactive analysis using the simulator is presented, which enables us to visualize the degrees of resilience of an existing work procedure in operation in facing with potential variabilities. In the end, how to utilize the proposed method for the establishment of the resilient systems from the point of view of the design of operation procedures is presented.

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

In recent automated production lines, it is important for industrial robots to recover from unexpected errors autonomously without intervention by human operators. In this paper, a hierarchical planning system that realizes autonomous error recovery of robots is proposed and demonstrated, and its effectiveness in cooperative work is verified. In order to realize a variety of flexible error recovery functions, a partial order planning scheme as well as a re-planning capability utilizing semantic information on the production environment are proposed.

DOI： 10.9746/sicetr.53.80

CiNii Books

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

The present paper investigates eye-gaze data of professional train drivers with different years of experience using Markov Cluster Algorithm (MCL) in order to extract characteristic eye-gaze patterns fostering a better understanding of their visual perceptual skills. MCL distilled a basic eye-gaze pattern in their visual behavior indicating that all the drivers would repetitively move their gaze ahead soon after looking at another area of interest, but they were found different in the "strength" of the pattern depending on their level of expertise. It was also clarified that inexperienced drivers made frequent deviations from the basic eye-gaze pattern in particular segments of the route where they had to deal with multiple tasks in parallel imposing higher cognitive loads of them.

DOI： 10.3156/jsoft.28.598

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

In this study, a novel method of safety analysis for safety management in train maneuvering is proposed based upon a systemic accident model, in which a train maneuvering activity by a train driver is regarded as a system of interrelated functions to be attained. Our method is based upon an original idea of Functional Resonance Analysis Method (FRAM) that was proposed by Hollnagel (2004), and our proposal attempts to identify how an unsafe event did happen as a result of interactions of deviated functions. For this purpose, at first a representation scheme for structuring interrelationships among train drivers' task components is provided. These interrelationships are derived from sources of drivers' instruction manuals, rules, regulations, etc. and are represented as interrelated networks of functions. Then, for a particular unsafe event that was actually happened in the past, our scheme is applied to represent the underlying maneuvering task and it is shown how the normal task procedures turned out to be an unsafe event through the propagations of primitive and local deviations towards the other part of the task. In this way, our method can replay a scenario of how the unsafe event did happen, and in addition to that, is able to detect latent hazards that did not actually happen but might happen when some conditions underlying at the event were to change from the original case.

DOI： 10.9746/sicetr.52.68

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JLC/20020705983?from=CiNii

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

Train driver's attentive behavior can be considered to reflect differences in his/her level of expertise. In this paper, the authors investigate train drivers' eye-gaze movements to identify characteristic attentive patterns that can distinguish different levels of driver expertise. Eye-gaze data were collected from measurement experiments in which six active train drivers participated. Classified into experienced and younger drivers groups, the data sets are analyzed using "sequence alignment", which helps us to extract partial temporal sequences of areas of interest that are common in each group of drivers. Resultant sequences are compared with each other to find features that could distinguish the groups in their visual behavior.

DOI： 10.9746/jcmsi.8.114

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.9746/sicetr.51.171

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：JAPAN SOC MECHANICAL ENGINEERS  

Like all man-made artifacts, Advanced Driver Assistance Systems have certain limitations beyond which they cannot function properly or work at all. The users in the driving domain have diverse backgrounds and it is likely that many of them start interacting with the technology without having gained detailed knowledge, such as by reading user manuals. This study explores whether drivers with only minimal knowledge of the Lane Departure Warning system can recognize the system's speed threshold and update themselves about this in a dynamic driving environment. Since drivers are not always single-mindedly focused on driving so participants were divided into two groups. Group 1 subjects performed only the driving task, while Group 2 was prompted to carry out a secondary task in addition to the primary driving task. Our investigation allowed us to estimate the effects of drivers' minimal mental model of the system working due to lack of sufficient knowledge concerning its capacities upon their 1) learning from experience and 2) situation awareness. It was assumed that the demands of multitasking would impair the driver's ability to observe the system's state and delay the mental model improvement process, compared to a setting in which driving was the only task. Experiment results using a driving simulator have been presented. The results revealed that the number of the drivers, who did not become aware of the operating condition of the LDW system, was quite high for both groups regardless of the nature of their tasks. It had been deduced that the users' preconceptions and expectations due to their limited knowledge about the capacities of the technology, could make them rely on the system to alert them in different situations even when the system was not operating. Hence, these factors can not only adversely affect the social benefits associated with ADAS, but also the acceptance and usability of these systems.

DOI： 10.1299/jamdsm.2014jamdsm0032

Web of Science

researchmap

Language：Japanese  

CiNii Books

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese  

CiNii Books

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.9746/sicetr.47.656

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press  

In this article, we focus on the coordinative structure of human behavior, which contributes to specifying dynamics from time-series kinematic data. We propose a method for the extraction of dynamical interaction from time-series data of human behavior using Singular Spectrum Transformation. Using the proposed method, human behavior can be described as a letter string whose letters indicate where the motion segmentation is detected. We also discuss a method of extracting coordinative structures by constructing multiple alignments from the timing structure of extracted motion change points. To confirm the effectivity of the proposed method, the results of motion analysis are shown.

DOI： 10.20965/jaciii.2011.p1019

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IGI Publishing  

DOI： 10.4018/jkss.2010100104

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

DOI： 10.9746/jcmsi.3.299

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

Automated systems are often structured into modes for providing enhanced capabilities to perform complex tasks, but they also increase the need for and difficulty of maintaining mode awareness, which may induce mode errors and eventually lead to automation surprises. Salient feedback on the mode status is necessary for the user to maintain mode awareness, and it should be carefully designed not to interfere in the user's primary task to be achieved in collaboration with the systems. For the purpose of developing a practical theory for effective and robust supports for maintaining mode awareness, this study focuses on <i>action-effect relations</i> as an intuitive human-machine communication channel. An action-effect relation is defined as an association of a user action with some observable effects on the system behavior, and it is expected to provide additional valuable information on the internal state of the system with no focal visual attention required. From this perspective, one hypothesis is made that different modes with a same or similar configuration of action-effect relations should be easily mixed up with one another. This paper tests this hypothesis through a driving simulator experiment where subject drivers were examined on their awareness of the operating mode status of ACC (Adaptive Cruise Control) during driving with the mode-rich automated system. Another experiment examines the contribution of differentiated configuration of action-effect relations to enhancing mode awareness.

DOI： 10.9746/sicetr.45.546

CiNii Books

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

CiNii Books

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SPRINGER-VERLAG  

The main difficulties of the skill transfer from human to machine exist in the fact that human skills are based upon two types of knowledge: one is the fundamental content knowledge needed to perform complex tasks, and the other one is knowledge of the process, by which those tasks should be executed. Distinguishing between those two types of knowledge, we present a comparative analysis between a fuzzy controller and a human expert. Regarding a human proficient expert as an ecological expert after Kirlik, we demonstrate that skillful control lies not only inside of the skill-performer's brain, but in the actor-environment system. In order to investigate into the relations between the human judgments and the environmental information, we adopt Brunswik's Lens Model to quantify both types of knowledge from the performance data. By analyzing how the ways of an operator's interacting with the task environment change and how the cues in the environment utilized by him/her alter, we formalize his/her control-skill improving process. We investigate these in comparison with the conventional fuzzy controller. We conclude in the aspects in which the human expert is superior to the fuzzy controller.

DOI： 10.1007/s00500-002-0201-y

Web of Science

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

DOI： 10.9746/sicetr1965.38.1097

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Systems, Control and Information Engineers  

This paper presents a method for detecting invariant task structures that are commonly shared by the human operator and the autonomous robot in a VR-based (i.e., virtual reality based) mobile robot teleoperation system. We call such a property of having common structures as <I>task-morphology</I>, and we discuss how this property functions well to establish the naturalistic task sharing between the two who have different perception-acting capabilities. For this purpose, we analyze their continuous task performance and segment them into some qualitatively different phases using Adaptive Resonance Theory (ART).

DOI： 10.5687/iscie.13.12_537

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Association for Computing Machinery  

DOI： 10.1145/350752.350761

researchmap

Language：English  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

In polysomnography, sleep technologists visually analyze a time series of respiratory motions, i.e., a respiratory curve, to detect abnormal respiratory events such as apneas and hypopneas. We investigate time-series processing to automate this analysis process. The present study examines the performance of abnormal sleep epoch detection models that utilize information on the occurrence frequencies of respiratory motion patterns. Respiratory motion patterns are obtained by dividing the respiratory curve at change points and then clustering the resulting partial curves. We compare three clustering methods—k-means(Euclidean), k-means(DTW), and k-Shape—by an experimental study evaluating the performance of sleep epoch classification models created using logistic regression. The experimental results suggest that the epoch classification model using the k-Shape performs the best in detecting abnormal epochs, indicating the extraction of effective respiratory motion patterns.

researchmap

Authorship：Last author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

In this paper, we hypothesize that, if lying accompanies some psychological disturbance, the pupil diameter at that time would be larger than otherwise. We conducted an empirical study measuring the pupil response of respondents in an experimental task that simulated the Concealed Information Test. As a result of the experiment, although statistical significance was not recognized, it was suggested that the amount of pupil dilation in response to critical stimulus questions, where there is a need to intentionally answer falsely, was larger than that for non-critical stimulus questions.

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

In polysomnography, a sleep technologist visually analyzes a time series of respiratory movements, i.e., the respiratory curve, measured with a RIP belt or other equipment to record abnormal respiratory events such as apneas and hypopneas. The present study aims to develop a method for detecting peculiar patterns in the respiratory curve to automate this analysis process. Our approach here is to divide the respiratory curve into partial time series using Singular Spectrum Transformation, then classify them based on the similarity of the time series patterns to identify essential features that characterize abnormal respiratory motions. This paper presents our investigation into applying the respiratory motion patterns obtained by a time series clustering method to detect abnormal respirations.

researchmap

Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

Locomotive syndrome is the decline in mobility function due to aging-related deterioration of the locomotor system, which increases the risk of falling. This study aims to analyze and quantify behavioral components associated with falling risk by measuring daily walking motions. We propose a walking motion analysis method based on the Gaussian Process Dynamics Model (GPDM) to detect changes in body usage. This method compresses gait motion data into a low-dimensional latent variable model using GPDM and visualizes the differences in body movements caused by behavioral conditions through the model's output.

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

In polysomnography, sleep lab technicians visually analyze respiratory motion time series, i.e., respiratory curves, measured with a RIP belt or other equipment to record abnormal respiratory events such as apneas and hypopneas. This study aims to develop a method to detect peculiar patterns in respiratory curves to automate this analysis process. The approach of the proposed method is to divide a respiratory curve into partial time series using Singular Spectrum Transformation, then classify them based on the similarity of the time-series patterns and identify the parts that represent abnormal respiratory motions. In this paper, we compare four time-series clustering methods of k-means(Euclidean), k-means(DTW), k-means(Soft-DTW), and k-Shape for the classification process of respiratory curve segments. The experimental results suggest that classifying respiratory curve segments requires time-series distance measures thatallow for differences in period and phase but not in scale.

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：The Iron and Steel Institute of Japan  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

Recently, the Cabinet Office is advocating a concept of "Super City" that realize new lifestyles and businesses for fundamentally changing the state of society by AI. The authors have established a research unit of "Health Smart Network" at Research Institute for Socionetwork Strategies (RISS), Kansai University as a research base to meet these social demands. Specically, the purpose of this research unit is to make health promotion services smarted with "eHealth + AI", create new social communication, and contribute to extending healthy life expectancy and sustaining healthy life. In this presentation, we will outline the concept of the research unit of "Health Smart Network" and introduce various research topics on the theme of smart health promotion for the progress of future research.

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

As motor function improves or declines, the way to control the body changes even for the same class of actions. To investigate what parts of the body movements characterize such changes, it is practical to summarize a high-dimensional time series of body movements into a low-dimensional time series by using the information on the coordinated behavior across different body parts. Significant differences in the summarized motion series should likely originate from characteristic differences in body control. In this study, we apply a conditional Gaussian Process Dynamic model to walking motions and propose a method to analyze the characteristics of body motions based on the dynamics of latent variables derived from a high-dimensional body movement time series.

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：Human Interface Society  

In order to gain insight into the cognitive processes that characterize the proficiency level of English as a Second Language (L2), this study experimented with measuring the eye-gaze point of L2 learners at different proficiency levels while reading English text. We applied the Singular Spectral Transformation, a change-point detection algorithm, and clustering method to the eye-gaze point time series to extract the characteristic patterns of eye-gaze movements. The analysis results suggest that the occurrence rate of particular eye-gaze patterns depends on the learners’ proficiency level.

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

Recently, due to the widespread effects of Covid-19 infection, social communication, such as person-to-person, person-to-society, or social connection, is being diluted. As one solution to the problems, it is expected to create new social communication utilizing advanced technologies such as ICT and AI. On the other hand, the Cabinet Office is advocating a concept of "Super City" that realize new lifestyles and businesses for fundamentally changing the state of society by AI. The authors have established a research unit of "Health Smart Network" at Research Institute for Socionetwork Strategies (RISS), Kansai University as a research base to meet these social demands. Speci cally, the purpose of this research unit is to make health promotion services smarted with "eHealth + AI", create new social communication, and contribute to extending healthy life expectancy and sustaining healthy life. In this presentation, we will outline the concept of the research unit of "Health Smart Network" and introduce various researchs on the theme of smart health promotion for the progress of future research.

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

This paper proposes a data analysis method for detecting abnormal breathing events during sleep from respiratory curves. The proposed method segments a respiration time series using Singular Spectral Transformations and classifies resulting partial respiratory curves based on their features consisting of autoregressive coefficients and respiratory amplitude ratios. Applying it to a polysomnography dataset confirmed that the proposed method could extract temporal patterns characteristic of respiratory abnormalities such as apnea and hypopnea.

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：The Society of Instrument and Control Engineers  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：The Society of Instrument and Control Engineers  

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：The Society of Instrument and Control Engineers  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：The Society of Instrument and Control Engineers  

Structural Equation Modeling (SEM) is a statistical analysis technique for fitting networks of constructs to data. The strength of extracting latent factors and estimating causal relations among them makes it accessible to analyze the structure of social problems, such as those around the economy, healthcare, and ecology. The present paper introduces SEM to the machine learning field for reducing the number of data dimensions. Too large dimensionality of the input data increases the computational load to build machine learning models. The proposed method utilizes SEM to sift variables and sort out relations between factors. The results show that the proposed method is able to identify the least number of data dimensions for machine learning-based prediction models to achieve the same or even higher accuracy as the prediction models trained with the full dimensions do.

researchmap

Authorship：Lead author,　Last author,　Corresponding author   Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)   Publisher：The Iron and Steel Institute of Japan  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Elsevier B.V.  

The realization of the 'super-smart society' of "Society 5.0"is being promulgated as the 5th Science and Technology Basic Plan by the Japanese Government. This article summarizes what is needed to promote new science and technology for achieving a super-smart society. Especially, to accelerate the development of "control for societal issues,"three aspects of "Feedback,""Ring"and "Harmony,"each of which corresponds to the means, the object and the goal of the future control-related science and technology, respectively, are to be stressed. That is, the object of science and technology is to target society or community, not single persons and individuals, and we have to understand complex feedback structures made up of many interactions as essential means to attain the goal of harmonizing technology, human and environment. For this purpose, innovations will necessitate elaborating the SoS (System of Systems), mutually connected systems, including human-in-the-loop systems. It would be significant to model and understand the dynamical complexity of such an SoS and to develop a technique for guaranteeing their resilience. This article presents overviews of some of the author group's works that are related to the above three aspects.

DOI： 10.1016/j.ifacol.2020.12.2734

Scopus

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：English  

DOI： 10.1016/j.ifacol.2019.12.083

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

DOI： 10.14864/fss.34.0_217

researchmap

Language：Japanese  

DOI： 10.14864/fss.34.0_265

researchmap

Language：Japanese  

DOI： 10.14864/fss.34.0_429

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：English  

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：Japanese  

DOI： 10.14864/fss.33.0_553

researchmap

Language：Japanese  

DOI： 10.14864/fss.33.0_181

researchmap

Language：Japanese  

DOI： 10.14864/fss.33.0_349

researchmap

Language：English  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：American Society of Mechanical Engineers (ASME)  

Sleep apnea and other sleeping disorders impair health and quality of life. Polysomnography is the primary method for diagnosis, but involves cost and utilization of medical resources, which limit access for potential patients. The clinical environment and sensors of polysomnography hinder typical sleep patterns in many individuals, thus degrading the analysis. Sensors suitable for at-home monitoring of sleep have recently become available. At-home monitoring of sleep may improve diagnosis due to increased familiarity for sleeping and ability for multiple sleep sessions, as well as lowering the cost. However, more robust algorithms would be needed to partially compensate for the less controlled conditions and sensor systems. A mat with a grid of force sensors has become available. This study was developing a state machine algorithm to analyze the activity at multiple force sensors of a mat while the subject was lying in supine position on the mat and undertaking natural, rhythmic respiration. The algorithm monitored the subset of active sensors to detect potential respiratory cycles. The similarity of the timing of the detected cycles between different sensors was used to determine the overall pattern of respiratory activity for the subject. Reliable detection of timing for respiratory cycles would be useful for detection of sleep apnea events.

DOI： 10.1115/IMECE2017-72240

Scopus

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ELSEVIER SCIENCE BV  

In this study, analysis of team situation awareness using a serious game of supply chain management for steel production is provided. By tracing the development of the performances of the game made by two different teams, a comparative analysis is done with respect to team situation awareness shared within the team. Especially, we focus into the individual players mental models that may evolve and grow as they accumulate a number of game sessions. From the acquired protocols of players during the sessions, players temporal mental models of situation awareness are traced, then the progress of team situation awareness among the players are derived. Through a comparative analysis of the two different teams with respect to their shared team situation awareness, we discuss about the relationships between the maturity of the teams and performance scores attained by the teams. Finally, a simulation model for generating such multiple players behaviors within a variety of teams is proposed. This is based upon a descriptive decision model of Garbage Can Model, and findings discovered during the gaming sessions are verified and generalized through a model-based constructive approach. (C) 2016, IFAC (International Federation of Automatic Control) Hosting Elsevier Ltd. All rights reseirved.

DOI： 10.1016/j.ifacol.2016.10.617

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ELSEVIER SCIENCE BV  

Analyzing the feasibility of procedures is important for ensuring safety when using systems that feature human operations and highly developed automation. This is certainly the case with flight-deck procedures due to their complexity, which sometimes leads to deviation from standard operation procedures (SOPs) and other serious outcomes (e.g., air crash accidents). To analyze the feasibility of procedures, we adopt the functional resonance analysis method (FRAM) (Hollganel, 2004) to examine the safety management of flight-deck procedures. However, FRAM is essentially a theoretic method, and there are currently no specific approaches or supportive tools to bridge the gap between theory and practice. In this paper, we propose an adaptation of the cognitive reliability and error analysis method (CREAM) (Hollnagel (1998)) that we call Fuzzy CREAM for systematic and quantitative FRAM analysis. We applied the proposed method to an actual air crash accident that occurred near Cali Airport, Colombia in 1995 and conclude that the accident was due to deviation from SOPs. On the basis of our analysis, we show that FRAM can identify potential hazardous paths that may lead to an accident. We also propose a new method using FRAM for pre-analysis of the safety of designed procedures. (C) 2016, IFAC (International Federation Control) Hosting By Elsevier Ltd. All rights reserverd.

DOI： 10.1016/j.ifaco1.2016.10.455

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In the present study, dominant eye-gaze patterns in professional train drivers' visual behavior are investigated using the Markov Cluster (MCL) algorithm. Applying the MCL algorithm results in a common gaze pattern showing a sort of perception tactic the drivers usually follow. The drivers repetitively move their gaze ahead soon after looking at somewhere else, independently of their years of experience. They are, however, found different in that experienced drivers can consistently follow the tactic while inexperienced drivers cannot. Time variation in the number of attentive pattern deviations demonstrates that, as well as the inexperienced drivers made higher frequency and larger fluctuations of pattern deviation, there were several particular segments in the route in which intensive pattern deviations arose in common. Inexperienced drivers would make intensive pattern deviations in such route segments that may have higher requirements of their cognitive resources.

DOI： 10.1109/SCIS&ISIS.2016.116

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER INT PUBLISHING AG  

In the present paper, additive syntheses of two tones are examined in order to look into what data-to-sound mapping design is better to acoustically communicate varying quantities of a multivariate system. The research focus is placed on the effectiveness of mappings for listeners to discriminate one variable component of sonified data from another. Dyads of pitches are tested through a couple of experiments where a special attention is paid to the "dissonance" of two tones. The first experiment investigates its effects on the perception of two static tones presented simultaneously. The second experiment investigates the effectiveness of choosing dissonant pitch combinations to sonify two quantities of a system being controlled manually. Experimental results demonstrate that the dissonance of tones can improve the perception of changes in individual variates out of a synthesized sound but that the frequency interval between two tones has stronger effects in dynamic situations.

DOI： 10.1007/978-3-319-40349-6_3

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In recent automated production lines, industrial robots are required to recover from unexpected errors autonomously. In this paper, a hierarchical planning system that realizes autonomous error recovery of robots is proposed with some virtually demonstrated results. In order to realize a variety of flexible error recovery functions, a partial order planning scheme as well as a "deliberate" re-planning capability based on semantic information on the production equipment and/or workpieces are proposed. Its effectiveness in cooperative work simulations of multiple robotic manipulators is verified.

DOI： 10.1109/SMC.2016.7844433

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ELSEVIER SCIENCE BV  

The present paper investigates train drivers' eye-gaze data to find important features to explain behavioral differences between experienced and inexperienced drivers. The obtained eye-gaze data. contain too complex transition structure to find any meaningful patterns that might be common across or differentiate. drivers. The Markov Cluster Algorithm, which is an unsupervised algorithm for graph clustering, is therefore utilized to divide such a structure into clusters that represent constituent eye-gaze patterns of frequent occurrence. As a result, a common eye-gaze pattern was identified to represent a perception tactic that drivers would repetitively move their gaze ahead soon after looking at other specific areas. Comparing cluster Structures extracted with different clustering parameter settings clarified that all of the drivers implement this tactic more or less, but that they arc different in that the experienced drivers can consistently follow it while the inexperienced drivers can not. (C) 2016, IFAC (International Federation Control) Hosting By Elsevier Ltd. All rights reserverd.

DOI： 10.1016/j.ifacol.2016.10.633

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ELSEVIER SCIENCE BV  

This study proposes a novel method for safety management in train operation and maneuvering based upon Hollnagels functional resonance analysis method (FRAM), which assumes that components of a system interact mutually and resonantly. According to the original FRAM, the general functions of train drivers and conductors tasks are defined, and the normal procedure of a common case is represented. Our method first attempts to replay how fluctuations of some of the tasks propagate through the entire system. Then we examine whether a countermeasure found by replaying is valid under various situations. In this way, we demonstrate that our method is able to detect latent hazards of the system as well as to identify how the system is influenced by various external and internal changes. (C) 2016, IFAC (International Federation of Automatic Control) Hosting Elsevier Ltd. All rights reseirved.

DOI： 10.1016/j.ifacol.2016.10.619

Web of Science

researchmap

Language：English  

File： 012Full.pdf

researchmap

Language：English  

File： 020Full.pdf

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

The present paper investigates train drivers' eye-gaze movements to find characteristic attentive patterns that can distinguish different levels of driver expertise. Eye-gaze data were collected from early-morning and nighttime observation studies in which six active train drivers participated. The acquired eye-gaze data were processed with Markov Cluster Algorithm which generates composites of Area-of-Interest (AoI) clusters, whose structure characterizes individual drivers' attentive behavior. Resultant structures are compared with one another to specify characteristic transition patterns that have a small number of hubs with different strengths within AoI networks according to the driver's level of expertise and the driving situation.

DOI： 10.14864/fss.31.0_574

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ELSEVIER SCIENCE BV  

This paper presents a novel human-system collaborative robot-programming platform, where case-based reasoning (CBR) and explanation-based learning (EBL) are integrated together. CBR takes advantage of its unique CBR cycle to achieve the knowledge acquisition and reuse in the form of case, realizing efficient robot programming with the support of experienced human experts. EBL optimizes the rule structure in the knowledge base through learning in retrieving speedup rules in order to accelerate the case adaptation process. Feasibility of this proposal is verified via a number of experiments that allow the system to output both schemata for generalized robot programming tasks whose calculation rules are adaptive enough so that it can be applied to novel task inputs. Moreover, it is shown that our system is adaptive to the increase of the cases processed and be able to tackle with the learning utility problem.

Web of Science

researchmap

Language：Japanese   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

Train driver's attentive behavior can be considered to reflect differences in his/her level of expertise. In this paper, we investigate train drivers' eye-gaze movements to identify characteristic attentive patterns that can distinguish different levels of driver expertise. Eye-gaze data were collected from observation studies in which six active train drivers, three younger and three experienced, participated. The acquired eye-gaze data were processed with Sequence Alignment, resulting in partial time sequences of gazed objects that are common in each group of drivers. We compare those sequences with each other to find characteristic patterns differentiating the driver groups in their attentive behavior.

DOI： 10.14864/fss.30.0_674

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this study, we applied the methodology of Interpretive Structural Modeling (ISM) in order to extract eye-gaze movement patterns characteristic of experienced train drivers in their visual attentive behavior. Application of ISM to eye-gaze data generates multilevel structural models that represent dominant transition patterns hidden in complex eye-gaze movements. The extracted structural models demonstrate that, as a basic perception strategy, the drivers repetitively moved their gaze ahead soon after looking at some other specific area. What differentiates the experienced and younger drivers is that the former group of drivers can consistently follow this basic perception strategy during driving while the latter can not. The experienced drivers can balance their limited attentional resources between vigilance to the external environment and other operational requirements, distributing their attention effectively and efficiently among various areas of interest.

DOI： 10.1109/SCIS-ISIS.2014.7044709

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

The present paper proposes a procedural design method which makes use of dependency structures underlying menu item descriptions in order to generate well-structured and easily-learned menu hierarchies. A dependency structure captures syntactic relations among conceptual units that constitute a natural language description of what function the corresponding menu item stands for. The proposed method classifies computerized system functions after dependency structure prototypes and then serializes variable elements to be specified in each function class after phrase structure analysis. Its clear and consistent policy provides generated menu systems with high communicability to users. The effectiveness of the method is empirically investigated.

DOI： 10.1007/978-3-319-07731-4_16

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IFAC Secretariat  

This paper presents a novel human-system collaborative robot-programming platform, where case-based reasoning (CBR) and explanation-based learning (EBL) are integrated together. CBR takes advantage of its unique CBR cycle to achieve the knowledge acquisition and reuse in the form of case, realizing efficient robot programming with the support of experienced human experts. EBL optimizes the rule structure in the knowledge base through learning in retrieving speedup rules in order to accelerate the case adaptation process. Feasibility of this proposal is verified via a number of experiments that allow the system to output both schemata for generalized robot programming tasks whose calculation rules are adaptive enough so that it can be applied to novel task inputs. Moreover, it is shown that our system is adaptive to the increase of the cases processed and be able to tackle with the learning utility problem.

DOI： 10.3182/20140824-6-za-1003.01492

Scopus

researchmap

Language：English  

researchmap

Language：Japanese  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This study empirically examines parameter-mapping sonifications which generate non-speech audio to represent the gap between the current and the desired state of the system being controlled. Focusing on manipulations of sound timbre and intensity, several data-to-sound mappings to transform the gap or system error into audible sound are compared in order to induce general knowledge of sonification design. Two experiments, using a one-dimensional tracking task, confirmed that timbre variation is useful in conveying categorical information about the target direction, and that scaling functions to produce no audible sound at the goal state with a relatively large dead zone are advantageous.

Web of Science

researchmap

Language：Japanese  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Human operators who supervise advanced automation systems need information not only about the systems being controlled but also about the automation itself engaged in the system control. They are required to have correct mental models on mechanisms and constraints underlying in the system functions to make decisions on monitoring and adjusting complex interaction among subsystems or components. In the present paper, a new method is proposed to investigate complex control structures consisting of human and technological agents based on the framework of Cognitive Work Analysis (CWA). After the first two stages of CWA, their separate results will be combined into a unified picture which gives visualization of connections between different controllers' tasks and the system being controlled. Finishing mill, a highly automated manufacturing process, is employed as an example domain to demonstrate how this method make visible the complex control structure within the system. Copyright © 2013 IFAC.

DOI： 10.3182/20130811-5-US-2037.00075

Scopus

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

Because visual attention is so important for train drivers to operate trains safely, analysis of their attentive action may lead to knowledge to improve their vigilance and information retrieval skills. In this paper, by means of Interpretive Structural Modeling, we try to extract characteristic eye movement patterns to distinguish train drivers with different levels of expertise. The result shows that the method can extract characteristic patterns in attentive action from train driver's eye-gaze time series.

DOI： 10.14864/fss.29.0_125

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Like all man-made artefacts, Advanced Driver Assistance Systems have certain limitations beyond which they cannot function properly or work at all. The users in the driving domain have diverse backgrounds and it is likely that many of them start interacting with the technology without having gained detailed knowledge, such as by reading user manuals. This study explores whether drivers with only minimal knowledge of the Lane Departure Warning system can recognize the system's speed threshold and update themselves about this in a dynamic driving environment. Since drivers are not always single-mindedly focused on driving, a group of participants was also prompted to perform a secondary task while driving. Our investigation allowed us to estimate the effects of 1) drivers' minimal mental model of the system's working and 2) insufficient knowledge concerning the system capacities, upon their situation awareness and their capabilities of learning from experience. It was assumed that the demands of multitasking would impair the driver's ability to observe the system's state and delay the mental model improvement process, compared to a setting in which driving was the only task. Experiment results using a driving simulator have been presented. Copyright © 2013 IFAC.

DOI： 10.3182/20130811-5-US-2037.00022

Scopus

researchmap

Language：English  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Segmentation of human's behavior plays an important role in behavior analysis. As a method based on Singular Spectrum Transformation (SST) can extract change-points in human's behavior caused not only by muscle forces but also by external forces, it has been proven that SST is an effective tool for behavior analysis. However, attitude representation appropriate for change point extraction has not been considered. In this paper, a method to extend SST which can be applied to detect change points of multi-dimensional time series is described. Because coefficients of AR models of the time series was compared with in extended SST, it is proven that attitude representation using unit quaternion is appropriate due to its linear kinematic equation. On the other hand, extended SST using Euler angles can not extract motion change points properly due to its nonlinear kinematic equation. Moreover, it is also proven that SST of individual elements of unit quaternion fails to extract motion change points and that extended SST of unit quaternion is an essential tool to extract motion change points appropriately.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Singular Value Decomposition (SVD) plays an important role in analyzing human behavior. Human behavior can be represented using attitudes of human body parts. Unit quaternion is widely used to represent attitudes because of lacking singularity. In the previous study, unit quaternion was used to represent attitudes of human body parts. However, a constraint exists in the representation using unit quaternion, and it has significant effect on the feature extracted by SVD. The purpose of this paper is to establish a proper representation of attitudes of human body parts for behavior analysis using SVD. Behavior analysis results using unit quaternion were compared with the results using exponential map, and the contribution ratio of the first mode in using unit quaternion is too large to analyze human behaviors properly. Finally, we concluded the exponential map is suitable attitude representation.

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This study applies information scent model to the domain of menu-based interfaces and demonstrates its ability to predict user behavior within a menu hierarchy. The information scent model addresses how people make use of perceptual cues like snippets of text in order to make their navigation decisions in choosing a particular option and gives a formula to measure the relevance of each option to a particular goal concept. The model was used to analyze the data collected in a set of experiments to study characteristics of users' menu selection behaviors. Users have to base their navigation decisions on the lexical similarity found on menu labels with their intended use when they have not learned the structure of the menu system yet. Our analysis confirmed that the information scent distribution can give a useful indicator to predict which item in a menu such users may select.

DOI： 10.1109/SCIS-ISIS.2012.6505334

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This paper proposes a help system that would simplify the use of search functions on audio visual (AV) appliances. The proposed help system is equipped with an interface for user interaction that employs a frame-driven dialog model. The frames are designed using common structures extracted by analyzing device functions based on interaction primitives. In addition, the proposed help system estimates the types of common structures (function schema) on the basis of the operation history recorded prior to the user pressing the help button to reduce user effort. To ascertain the effectiveness of the function schema estimations, we constructed a help system for a DVD recorder and evaluated the decrease in user effort by using the operation history data of the DVD recorder derived from 30 tasks performed by 14 users. The experimental results showed an 8.6% reduction in user effort when compared to a system that does not apply function schema estimation. © 2012 IEEE.

DOI： 10.1109/GCCE.2012.6379590

Scopus

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Human behavior is affected by some dynamic parameters, which represent the interaction between the body and the environment. In the previous research, a method to extract adaptive components in the behavior has been proposed. This method provides precise reconstruction of behavior using the adaptive components. However, it is not proven that these reconstructed behaviors affect people to perceive the dynamic parameters properly. In this paper, a method to estimate the cognitive scale of the dynamic parameter and a method to extract the characteristic components necessary for perception are proposed. The proposed methods have been applied to loading motion. Those results have suggested that the extracted characteristic components, that is the cognitive index, are not identical with the adaptive components.

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Humans are able to acquire information on dynamics that cannot be directly obtained through observation of kinematics. Since the coordinative structure of the human behavior is self-organizing in interactions with the environment, a change in the environment induces a change in the coordination structure. It is posited that humans identify dynamic interactions between the body of the actor and the external environment through the interpolation of coordinated structures for observed behaviors. Coordinative structures consist of invariant part and covariant part. Invariant part is a set of motion change points which are found even if the environment or dynamic interaction has changed. Covariant part is a set of motion change points caused by the environmental changes. In this paper, we examine more closely the changes in coordinative structure induced by changes in dynamic interactions with the environment, and describe a method to extract invariant and covariant part of coordinative structures. Moreover The importance of the change in the coordinated structure in recognition of the dynamic interaction by observers is also described. The results of dynamics recognition experiments revealed that information necessary for the estimation of relative weight is contained in the covariant part of the coordinated structures but not in their invariant regions, and furthermore that timing is also an essential for estimation. © 2011 IEEE.

DOI： 10.1109/SII.2011.6147459

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This study empirically examines parameter-mapping sonifications to represent the difference or error between the desired state and the current state of the system being controlled using the non-speech audio. Eight different types of data-to-sound mappings were prepared by which the error was mapped onto either or both of two acoustic parameters of intensity and frequency. The mappings were tested through an experiment using a one-dimensional tracking task. The experimental results shows the following points: 1) as for the tracking task, the mappings are preferable in which the sound intensity falls as the error decreases, 2) no audible sound when no error remains is effective because listeners can easily identify the achievement to the goal from the sound itself, and 3) it seems more effective to map the error to the intensity and frequency redundantly than only to either of them, if perceptual interaction between changes in loudness and pitch is considered enough. Copyright 2011 by Human Factors and Ergonomics Society, Inc. All rights reserved.

DOI： 10.1177/1071181311551257

Scopus

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Iron and Steel Institute of Japan  

Nowadays steel manufacturing pays a critical attention to on-time delivery and inventory reduction. In our study, we propose a heuristic algorithm to solve the hybrid flow shop scheduling problem (HFSP) considering reducing total tardiness and inventory time. In order to achieve a feasible solution in limited computational time, the algorithm is based on Three-zone Buffer Management (TZBM) approach. TZBM approach combines the Drum Buffer Rope (DBR) method proposed in theory of constraints (TOC), with a contract net protocol in multi-agent system (MAS). The effectiveness of the proposed algorithm is shown by simulation results comparing with the results obtained by a Tabu search (TS) algorithm. In the production field, a shipment buffer is set between the due date and the shipping time in order to avoid possible late shipment. The size of shipment buffer is always decided according to the experience of a system operator who has responsibility to draw up a production schedule. Wherein, an operator needs to take a decision considering trade-off between on-time delivery and inventory reduction. Therefore, a new interactive decision support system is proposed which is based on the two-stage TZBM approach. The operator needs to make his/her judgment whether an optimal solution shown by the system is preferable or not, then to change the trade-off between the two conflicting objectives. The system needs to compute another optimal solution according to judgment given by the operator. Under the result of numerical experiments, to integrate the experience of operators with the computational intelligence of systems proves to be an effective scheduling method.

DOI： 10.2355/tetsutohagane.97.352

Scopus

researchmap

Other Link： http://repository.kulib.kyoto-u.ac.jp/dspace/handle/2433/171951

Language：English   Publishing type：Research paper (international conference proceedings)  

To avoid breakdowns in human-system interaction on the way to the target item, menu systems must be designed as compatible with his/her way of thinking. Concerning this issue, this study focuses on conceptual dependencies between elements involved in device functions. Dependency structures are examined to design interaction processes in which the user can narrow down options reasonably and effectively. A small set of common structures to classify device functions were extracted from the analysis of a wide variety of appliances' functions. The proposed method makes use of those structures to generate a menu hierarchy, facilitating the users applying the mental model they have already developed from their past experience to another new search context. © 2011 IEEE.

DOI： 10.1109/SII.2011.6147457

Scopus

researchmap

Language：English  

researchmap

Language：English  

DOI： 10.3182/20100831-4-FR-2021.00078

researchmap

Language：English  

DOI： 10.3182/20100831-4-FR-2021.00077

researchmap

Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

For the purpose of developing a practical support tool for human workers engaged in industrial robot teaching, the activity of teaching operation positions to a robot is analyzed by use of a mechanical explanation model to capture the force-displacement relationship inherent in the work system. The analysis clarifies a basic, normative operation strategy making use of a characteristic frame of reference in the position search space. Based on this finding, a prototype GUI is developed that can provide effective information supports for different granularities of the activity in position teaching. In terms of the distributed information resources model, each element of the GUI is qualitatively investigated to confirm that the proposed GUI represents the activity-related information in a way that affords the robot operators' strategic operations.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

We proposed a method to extract similarities and differences of human behavior using singular value decomposition. It was proved that the proposed method is able to extract some characteristic motion patterns from periodic behavior. This paper reports similarities and differences of non-periodic human behavior, such as loading motion, which are extracted by the proposed method. Instead of using Euler angles, unit quaternion is adopted for attitude representation to avoid singularity. However, it is difficult to analyze the characteristic motion patterns represented by unit quaternion. The method to analyze the extracted characteristic motion is reported in this paper. Discussions on these results are also reported.

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

DOI： 10.11509/sci.SCI10.0.268.0

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Measurement of the heading angle which is free from magnetic declination using several GPS antennas is widely applied, but it is not suitable for small unmanned aerial vehicles because of their limit of the payload and body length. In this paper, we propose a method to estimate the magnetic declination angle using an inertial measurement unit(IMU) and only one GPS antenna. The method consists of 1) estimation of the acceleration of the motion from GPS measurements and 2) estimation of the magnetic declination angle using the estimated accelerations and IMU measurements. We show that the estimation of the magneticdeclination angle results in reduced QUESTflO], that is an eigenvalue problem of 2 x 2 real matrix. Flight experimental results using an unmanned helicopter demonstrate the effectiveness of the proposed methods. ©2010 IEEE.

DOI： 10.1109/ICIT.2010.5472530

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：INSTICC-INST SYST TECHNOLOGIES INFORMATION CONTROL & COMMUNICATION  

The problem in applications of case based reasoning (CBR) is its utility problem, that is, the cost of retrieving the most appropriate case from the case library for a new given problem and the cost of adapting the retrieved case for solving the new given problem. This paper proposes an approach to solve the utility problem of CBR by integrating CBR and explanation based learning (EBL) from a perspective that emphasizes the function of learning in CBR. In this paper, CBR and EBL are integrated in an apprentice agent, and the application of this apprentice agent in the robotic assembly domain is given as an example.

DOI： 10.5220/0002763306670670

Web of Science

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Adaptation to the environmental changes, such as wind, has very important role to improve the reliability of an autonomous unmanned helicopter. Adaptive yaw (heading) control for an autonomous helicopter is discussed in this paper. The control structure is based on a hierarchal scheme utilizing an inner yaw feedback control loop plus an outer feedback loop, which estimates the direction of the airspeed using roll angle and roll angular rate. Stable coupling in yaw and roll motion is induced by the proposed controller to improve the stability of the helicopter's flight. Turning utilizing the proposed adaptive control system is discussed in particular. Results of flight experiments showed bifurcation of the helicopter's maneuvering in turning occurs depending on the airspeed. The results indicate that the autonomous unmanned helicopter can select a turning maneuver that is suitable for the environment and thus stabilizes its flight.

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

DOI： 10.11509/sci.SCI10.0.185.0

researchmap

Language：Japanese  

DOI： 10.11509/sci.SCI10.0.331.0

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publisher：一般社団法人日本ロボット工業会  

CiNii Books

researchmap

Language：English  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Conventional methods of online robot teaching have difficulity for the teaching operator to detect accurate positional errors between the hand and a workpiece. For the purpose of developing practical supports for online robot teaching, this study investigates information requirements of the operator's work and then discusses their effective visualization of data from a camera and a force sensor attached to the robot's wrist. Hierarchical Task Analysis is applied to the operator's task in defining positions and postures of the hand against actual workpieces, to extract specific pieces of information or relationship that should be incorporated into the visual assistance. Considering these information requirements, we propose a GUI to visualize the force information in order to find accurate motion points effectively. The proposed GUI can help the operators to figure out motion points more precisely, and can reduce the amount of time for robot teaching.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

To develop flying rescue robots using autonomous unmanned helicopters, it is necessary to improve performance and reliability of flight control systems. Adaptation against the environmental changes, such as wind, has very important role. In this paper, adaptive heading (yaw) control for an autonomous helicopter is proposed. Roll angle and roll rate are used to determine desired yaw angle. Therefore, roll dynamics and yaw dynamics are coupled and stable dutch roll is induced to change the yaw angle corresponding to wind direction or the direction of the helicopter's motion. Results of flight experiments show the effectiveness of the proposed method.

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

DOI： 10.11511/jacc.52.0.39.0

researchmap

Language：Japanese  

DOI： 10.11511/jacc.52.0.58.0

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

This research focuses on supporting an analyst&apos;s activity of interpreting the contents of existing incident reports. During this activity, analysts are always predicting expected scenarios of the incidents at hand in comparing that with the actual development of the incidents reported therein. In order to learn lessons from a particular prior experience, analysts should be aware of the latent similarities among the incidents and should experience a breakdown called "expectation-failure" to let that incident be surely printed in their memory. To let the human analysts experience this breakdown, our system introduces a theory of Memory Organization Packets (MOPs) as a framework for explaining the dynamic memory structure of the human. By utilizing this idea as a basis for scenario-based expectation of human analysts and by integrating this idea with a text-mining method, a system for supporting an incident analysis is developed for a domain of medical incidents. Results of the experiments using our proposing system are presented, where the subjects are nurses working for a hospital. Based on those results, effectiveness of the system is discussed from various viewpoints by investigating into the protocols gathered from the subjects of the experiments.

DOI： 10.1007/978-3-642-02559-4_65

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

High communicability of the menu-based system is on the basis of consistent vision and clear policy in designing the system of menus, and then they should be perceivable to the users. In this light,failures in menu-based interactions can be explained that they might emerge from lack of information in the users&apos; available cues to identify the design vision. This study focuses on communicative break-downs in menu-based human-computer interactions from this perspective, and investigates their causes in ill-organized structures of menu hierarchy in terms of the user&apos;s interpretation of the menu items. Pirolli&apos;s information scent model is extended and utilized as an analytical tool for describing the meaning system of menus from the users&apos; point of view, and their decision making in search of particular menu items is analyzed by use of information scent.

DOI： 10.1007/978-3-642-02574-7_49

Web of Science

researchmap

Language：Japanese  

DOI： 10.11511/jacc.52.0.297.0

researchmap

Language：English   Publishing type：Part of collection (book)   Publisher：Springer London  

In this chapter, we summarize the findings and knowhow in the individual developments of rescue robots and attempt to establish design guidelines for the human interface for rescue robots. The guidelines for image display were established on the basis of some case studies in the DDT project and RoboCupRescue. Since the established guidelines are general and independent of the purpose of a robot, we decided to build a standardized image display prototype. After introducing some implementation examples based on our guidelines, the standardized image display prototype is presented. The established guidelines would be useful for the future development of rescue robots. Further, the prototype image display is a good example of a standardized interface testbed. © 2009 Springer London.

DOI： 10.1007/978-1-84882-474-4_7

Scopus

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

researchmap

Language：Japanese  

DOI： 10.14864/fss.24.0.48.0

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Measurement of the heading which is free from magnetic declination by use of several GPS antennas is widely applied, but it is not suitable for small unmanned aerial vehicles because of its limit of the payload and body length. In this paper, we propose a method to estimate the magnetic declination using an IMU and one GPS antenna. The method consists of 1) estimation of the acceleration of the motion from GPS measurements and 2) estimation of the magnetic declination angle using the estimated accelerations and IMU measurements. We show that the estimation of the magnetic declination angle results in reduced QUEST, that is an eigenvalue problem of 2 x 2 real matrix. Experimental results of flight control of an unmanned helicopter demonstrate the effectiveness of the proposed method.

CiNii Books

researchmap

Language：English  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper discusses usability issues in the design of hierarchical menu systems in terms of the user&apos;s interpretation of the interface signs, or the menus. Focusing on the exploration of a particular functional item in the menu hierarchy, novice users&apos; interactions with a computerized system are examined from the perspective of communicative breakdowns at first, and then the menu structure is evaluated through the information scent analysis. The former analysis investigates differences between the designer and the users in the basic assumptions for signifying or interpreting the interface signs, (i.e., menu texts and their organization for listing), especially on the &apos;peripheral&apos; functions of the product. The latter analysis investigates the distribution of information scent in a menu list in order to obtain new quantitative clues for the assessment of whether the menu design is compatible with the users&apos; naturalistic decision strategy or not.

DOI： 10.1109/SICE.2008.4654686

Web of Science

researchmap

DOI： 10.11509/sci.SCI08.0.156.0

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

DOI： 10.11499/sicejl1962.46.945

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00304666735?from=CiNii

Language：English  

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

This paper presents a case study on developing a smart user interface for supporting the adjustment works on automated weighing machines of computer-controlled multihead weighers. Based upon the theoretical and practical framework of Vicente and Rasmussen's Ecological Interface Design (EID), we clarified the functional structure of the work domain in terms of the means-end relations, and visualized it on the screen displays to encourage the human operator's "direct perception" of the meanings or values of his practicable actions to those automated processes. Comparative experiments using test subjects with a variety of skill levels confirmed the effectiveness of the redesigned user interface that can facilitate unskilled operators appropriately evaluating and effectively responding to their immediate work situations, and that will take the place of the experts' knowledge and insights on the works as one of the distributed resources for cognition.

DOI： 10.1007/978-3-540-73354-6_96

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This study investigates a practical guideline on human-machine interaction design to facilitate human operator's correct awareness of the actual operating state of an automated control system in their joint activity. Based on the comprehension that a human being is a "proactive" agent who would affect the external world to obtain new cues to foster its understandings, the mapping structures between the human operator's input operations and the machine's response behaviors, i.e., the input-output relations, may be one of the dominant characters that can explain the operator's awareness of the machine's operating state. From this perspective, we speculate that different control modes with similar input-output relations should be easily mixed up with one another since interactions mediated by those relations would intimate a same mode. This hypothesis was verified through an experiment using a driving simulator that examined drivers on their recognitions of the operating mode of ACC (Adaptive Cruise Control) while driving with it. Another experiment confirmed that a human-machine interaction design to differentiate modes in their input-output relations could successfully reduce mode confusion errors.

DOI： 10.1109/ICSMC.2007.4413787

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Transdisciplinary Federation of Science and Technology  

A new method to analyze transformation processes of work procedures for on-site workersengaged in organizational activities, based on the concept of the four levels of contradictions, is proposed.Various contradictions, i.e., misfits between components of the activity system, might arise outof and propagate throughout their activities, then induce some sorts of changes in the procedures forthe better or for the worse. The proposed method, focusing on the negative aspects of such proceduretransformation, represents every phase of a changing activity in terms of a diagrammatic triangle ofEngestr¨om's activity theory, thereby visualizing the process of its changing with latent contradictionscan contribute to the in-depth analysis of organizational accidents. Within organizations, a humanplays variable roles as an agent that is an actor, an observer, a cognizer, and an interpreter, whichproduces a complex organizational behavior. In order to model this, we introduce the subject of C.S. Peirce's semiosis, which is any form of activity, conduct, or process that involves signs, includingthe production of meanings. The criticality accident occurred at the nuclear fuel conversion facilityof JCO (1999) is employed for an illustrative case to explain this method's capabilities.

DOI： 10.11487/trafst.1.2_106

researchmap

researchmap

researchmap

researchmap

Language：Japanese  

DOI： 10.14864/fss.22.0.135.0

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Although autopilots are contributing much to reducing human workloads in aviation, they may induce a new type of aircraft accidents, i.e., "automation induced surprises". This paper proposes a systematic approach to this issue from a perspective of safety management. We employ Hollnagel's FRAM (Functional Resonance Accident Model) to figure out possible collapses in a series of operational procedures due to variable cockpit conditions. It is analyzed how the performance deviation in a particular operation does affect on the others, and when and where the expected scenario breaks down as a result of nonlinear interactions of multiple performance shaping factors.

DOI： 10.1109/SICE.2006.315635

Web of Science

researchmap

Language：English  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we summarize the findings and know-hows in the individual developments and try to establish guidelines of human-interface design of rescue robots. These guidelines would be useful for the future development of rescue robots. Preliminary guidelines were set from some case studies of the DDT project and RoboCup Rescue. A trial towards the standardized interface is also shown.

DOI： 10.1109/SICE.2006.315163

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IOS PRESS  

The authors have proposed a formal approach to designing effective human-machine interaction channels, i.e., shared communicational modalities, between a human operator and a machine autonomy. In this framework, Kirlik's classification scheme of information types defined as Generalized Lens Model is utilized for distinguishing information resources in the target system from the perspective of their proximity to both human and machine agents' perception and action. The shared communicational modalities are then embodied virtually by imposing some mutual constraints among their respective [PP,PA] (i.e., proximal for both perception and action) variables. This paper evaluates the proposed interaction design after an actual implementation of such modalities into a shared-control environment, especially focusing on the mutual relationship of those agents' cue-utilization strategies as well as their joint task performances.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：I O S PRESS  

This paper presents a motion pattern analysis to examine the effects of the robotic power assisting behaviors to the user&apos;s motion structures. The basis of the analysis is placed on how much the user&apos;s original motion patterns are deformed by the intervention of the robotic system in terms of motion trajectory shapes. As to the one-handed lift-weight motion the developed robotic orthosis is intended for, spontaneous human motions are confirmed as specified by the bell curves of the angular velocity at the elbow joint, analogous to the arm trajectories in point-to-point movements. Using this frame of reference, those deformed motion structures that would fluctuate due to the purely passive, impedance controlled orthosis behaviors, are demonstrated by the comparative experiments. In addition, based on a hypothesis that the power assisting behaviors at the appropriate timing would keep the user&apos;s spontaneous motion structure, mother experiment is conducted to reveal that attuned orthosis motions led their collaborative motion structure to the same with the user&apos;s original and unforced motion patterns in terms of their shapes.

Web of Science

researchmap

Language：English  

DOI： 10.3182/20060522-3-FR-2904.00028

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper proposes anew method to estimate possible confusions human users may make on the current mode of an automated control system that they are required to work with. The proposed method takes its stand on a philosophy that the users depend on the system's response behaviors to some extent as valuable cues to sense its state, and thus that different modes in which the system exhibits the same response behaviors to a particular user operation are hard to be distinguished by them. All the modes are encoded into vector forms from the perspective of their "input-output relations," and then the distances among the vectors are utilized as indexes to estimate the likelihoods of mode confusions. The idea is validated through the experiments using a driving simulator with Adaptive Cruise Control (ACC) implemented, in which drivers are queried as to the current mode of the ACC system to collect the probability distributions of mode confusion to be compared with what has been estimated. In addition, it is also examined whether purposefully designed input-output relations can contribute to the improvements of the drivers' awareness of modes.

DOI： 10.1109/SICE.2006.315649

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Japanese manufacturing companies are now facing a real shortage of skilled workers who have expertise on production facilities and instruments. Automating their tasks by machine may be one promising solution to this problem. However, it is not enough. The future production facilities should be designed so that it can develop and nurture workers' qualities and skills to overcome the "deskilling" problem caused by automation. This paper deals with the printed circuit board inspection system utilized in manufacturing sites, and we develop an automated teaching system for tuning the parameters of the inspection system so that it can work being customized to particular site conditions. This tuning task has been left to the expertise of the skilled system engineers. However, our developing teaching system enables non-expert users to achieve accurate tuning equivalent to an expert's level. We show that our teaching system can improve the accuracy of the tuning task drastically. Moreover, we present our system can encourage non-experts to enhance their skills for the tuning, which is verified by investigating into the verbal expressions on their expertise that is analyzed by usage of the fuzzy structural modelling technique. The result shows that a non-expert user's knowledge cultivated by usage of our teaching system can change to the one similar to what an expert has.

DOI： 10.1109/IECON.2006.347641

Web of Science

researchmap

Publisher：The Institute of Systems, Control and Information Engineers  

DOI： 10.11509/sci.SCI06.0.18.0

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a research work to develop a power assist orthosis for supporting care-givers' hard physical tasks such as moving bedridden patients and changing their postures for nursing purposes. A prototype robotic orthosis, which was intended for assisting flexion and extension of the elbow joint during lift-weight motions, has been developed as a testbed for those discussions. Realizing natural and unforced physical assists requires directability of the orthosis for user intention as one of the most important dimensions for their collaboration. Concerning this issue, basic experiment and analysis were performed to explore plausible configurations of load sensors in the developed orthosis, which determine the communication channels from a user to the orthosis.

CiNii Books

researchmap

Language：Japanese  

researchmap

Elsevier Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper addresses the adaptability in the human-machine collaboration as a necessary element for the adequate coordination between human and mechanized control. Currently, all efforts to resolve any mismatch between humans and machines are charged only to the human operators as machines have no ability to deal with and adapt to variable behaviors of their human partners. Concerning this issue, inspired by our human proactive actions to the external world, the authors advocate introducing proactive agency into machine autonomies so that they can probe or sound out the partner's covert judgments for their adaptation. This behavioral model is implemented into a simulated shared control environment, in which both a human operator and a machine autonomy can control the behavior of a mobile robot. This testbed environment is utilized to investigate their joint activity and then discuss feasibility of the co-adaptive approach towards the well-coordinated collaboration by the human operator and the machine autonomy.

Web of Science

J-GLOBAL

researchmap

Language：Japanese   Publisher：Japan Society for Fuzzy Theory and Intelligent Informatics  

This paper examines the operational skills for the search activity via a remote-controlled robot to find out some hidden objects under debris, which simulates a robotic surrogate system for human rescuers. Our skill analysis revealed that the accurate and effective navigation of the robot requires the adequate view control of the camera. This operational strategy enables the parallel execution of the robotic ambulatory movement and the accurate environment recognition, ensuring the necessary cognitive resources available or externalized for human epistemic actions. Based upon the analyses, we discuss the necessary considerations to mitigate the cognitive restrictions imposed by the mechanical systems utilized and what encourages the naturalistic skill performance by the human operators, in terms of their ecological perspectives.

DOI： 10.3156/jsoft.17.441

CiNii Books

researchmap

Language：English  

DOI： 10.1109/IROS.2004.1389540

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/ROBOT.2004.1307155

researchmap

Other Link： https://dblp.uni-trier.de/db/conf/icra/icra2004-1.html#ZhengTSOTHA04

Pergamon

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents a new concept of the knowledge management framework for sharing technical know-how in an engineer community, in which latent connections among technical keywords are utilized to search some work reports for reference. The framework is composed of two different semantic networks with spreading activation mechanisms: the Query Extension Network and the Query Suggestion Network. Each of them represents a different aspect of connections among words. The former network is intended to extend the search for a user's query through its periphery based upon the associations of words in their work domain ontology. The latter is leveraged to present the users with other possible keywords for their next queries, which might help their exploration advance to what they want. The interaction of both functions through search activities by individual users is expected to provide effective navigations for their data explorations in the forest of expertise.

Web of Science

J-GLOBAL

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：WORLD SCIENTIFIC PUBL CO PTE LTD  

Shared control is a challenge to combine and capitalize on both advantages of human and mechanized automatic controls. But any interventions by a machine-autonomy may introduce some kind of collapses in the isomorphism between the two different interaction fields of the human operation and of the system performance, which is necessary for the naturalistic involvement of a human operator with the system control. On this issue, we focus on the adaptability for the machine-autonomy to coordinate the mapping between the two fields to facilitate the human feeling of direct involvement with the system control, as well as the human adaptation (i.e., co-adaptation). In this chapter, we investigate their joint activity in a shared control of a virtual robot teleoperation environment, and then discuss the effect of the machine intervention in terms of the system operationality for human operators. The feasibility of the co-adaptive approach towards the well-coordinated relationship between a manipulator and a manipulatee is also examined based upon the experimental result with a simple adaptation algorithm implemented into the machine-autonomy..

Web of Science

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper proposes a framework for a machine agent to constantly adapt its behavior toward the naturalistic collaboration with a human operator, that is inspired by the proactive aspect of human cognition. We introduce the proactive agency into the adaptation process of the machine agent to get more information about the partner's internal state that is difficult to understand under the impoverished conditions where only passive information is available, and to establish their mutual understandings. As an experimental testbed for our proposing model, some experiments using a simulated robot teleoperation environment were prepared to consider the collaboration between a human operator and an autonomous tele-operated robot.

Web of Science

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, instead of taking conventional artificial intelligence approaches for an interface agent, we propose a new designing method using the ecological approach, since we would like to stress the importance of the agent's ability to create reciprocal relationships with the human user (i.e., a human user and an interface agent). Based on this analysis, then, we designed an agent that can infer the coordination level with a particular human, identify his/her skill levels, and adjust the level of shared situation awareness with its human partner.

Web of Science

researchmap

Language：English  

DOI： 10.1007/3-540-45336-9_7

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：I O S PRESS  

This paper presents a new style of human-robot collaboration in a teleoperating environment where the robot has the autonomy to control its behavior. Our model provides the shared communicational modality between the human operator and the robot autonomy to promote their mixed-initiative interactions, This paper describes the results of experiments using our developing system to evaluate our model. and discusses the interactions between the two autonomies based upon the Lens model framework known as a method for-judgment analysis.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This work presents how the expert human controller differs from fuzzy controller. Definition of ecological experts is to investigate into the relations between what cues in the task environment the human expert is actually utilizing in their judgments and how those cues are valid for the task (i.e., cue validity and ecological validity). We introduce a Brunswik's lens model for analyzing the interactions between the human expert controller and the control task environment. The task is to steer a large ship having complex dynamics to let it pass through a number of gates, whose simulator is constructed on PC. We gather a collection of interaction data from a particular operator, whose skill levels are initially very poor but are later gradually improve as he experiences trials.

DOI： 10.1109/FUZZ.2001.1008871

Web of Science

researchmap

Language：English  

IEEE

J-GLOBAL

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents a method for a design of a robot teleoperation system based upon their mixed-initiative interactions to realize the naturalistic collaboration between a human operator and a mobile robot with its autonomy At first we analyze a robot-teleoperating environment and classify its parameters into the four different classes of information after the Kirlik's Generalized Lens Model; framework: whether the operator and/or the robot can directly perceive the information or not, and whether they can directly manipulate the information or not. Based on this analysis, we develop an ecological interface system adopting a force-feedback-joystick as a device providing with a shared modality between a human operator and a robot's autonomy.

Web of Science

J-GLOBAL

researchmap

Language：English  

DOI： 10.1109/ICSMC.1999.814186

researchmap

Language：English  

DOI： 10.1109/ENABL.1999.805202

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/wetice/wetice1999.html#conf/wetice/SawaragiH99

Language：Japanese  

CiNii Books

researchmap

SOCIETY OF INSTRUMENT AND CONTROL ENGINEERS (SICE)

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：INSTITUTE SYSTEMS, CONTROL & INFORMATION ENGINEERS  

This paper discusses about character-preserving and diversity-maintaining strategies for an evolutionary behavior-based robot that is instructed to behave according to two different training environments: a whole learning method and a partial learning method. We implement a behavior-based robot by genetic programming (GP), whose evolution is varied by installing Angeline and Pollack's module acquisition (MA) processes. We discuss about how the different training environments affect a robot's experiencing diversity of an environment and what the appropriate character-preserving and diversity-maintaining strategy is.

Web of Science

researchmap

Event date： 2017.11

Venue：Tampa, Florida, USA  

Sleep apnea and other sleeping disorders impair health and quality of life. Polysomnography is the primary method for diagnosis, but involves cost and utilization of medical resources, which limit access for potential patients. The clinical environment and sensors of polysomnography hinder typical sleep patterns in many individuals, thus degrading the analysis. Sensors suitable for at-home monitoring of sleep have recently become available. At-home monitoring of sleep may improve diagnosis due to increased familiarity for sleeping and ability for multiple sleep sessions, as well as lowering the cost. However, more robust algorithms would be needed to partially compensate for the less controlled conditions and sensor systems. A mat with a grid of force sensors has become available. This study was developing a state machine algorithm to analyze the activity at multiple force sensors of a mat while the subject was lying in supine position on the mat and undertaking natural, rhythmic respiration. The algorithm monitored the subset of active sensors to detect potential respiratory cycles. The similarity of the timing of the detected cycles between different sensors was used to determine the overall pattern of respiratory activity for the subject. Reliable detection of timing for respiratory cycles would be useful for detection of sleep apnea events.

researchmap

Event date： 2017.8

Venue：東京都  

researchmap

In this paper, we hypothesize that, if lying accompanies some psychological disturbance, the pupil diameter at that time would be larger than otherwise. We conducted an empirical study measuring the pupil response of respondents in an experimental task that simulated the Concealed Information Test. As a result of the experiment, although statistical significance was not recognized, it was suggested that the amount of pupil dilation in response to critical stimulus questions, where there is a need to intentionally answer falsely, was larger than that for non-critical stimulus questions.

researchmap

Venue：Villas des Mariages Karuizawa  

Locomotive syndrome is the decline in mobility function due to aging-related deterioration of the locomotor system, which increases the risk of falling. This study aims to analyze and quantify behavioral components associated with falling risk by measuring daily walking motions. We propose a walking motion analysis method based on the Gaussian Process Dynamics Model (GPDM) to detect changes in body usage. This method compresses gait motion data into a low-dimensional latent variable model using GPDM and visualizes the differences in body movements caused by behavioral conditions through the model's output.

researchmap

Venue：Villas des Mariages Karuizawa  

Recently, the Cabinet Office is advocating the concept of ”Super City” that realize new lifestyles and businesses for fundamentally changing the state of society through AI. The authors have established a research unit of ”Health Smart Network” at Research Institute for Socionetwork Strategies (RISS), Kansai University as a research base to meet these social demands. Specifically, the purpose of this research unit is to make health promotion services smarted with ”eHealth + AI”, create new social communication, and contribute to extending healthy life expectancy and sustaining a healthy life. In this presentation, we outline the concept of Kansai University’s ”Health Smart Network” and discuss the progress and results of various research on the theme of smart health promotion.

researchmap

Venue：Villas des Mariages Karuizawa  

In polysomnography, a sleep technologist visually analyzes a time series of respiratory movements, i.e., the respiratory curve, measured with a RIP belt or other equipment to record abnormal respiratory events such as apneas and hypopneas. The present study aims to develop a method for detecting peculiar patterns in the respiratory curve to automate this analysis process. Our approach here is to divide the respiratory curve into partial time series using Singular Spectrum Transformation, then classify them based on the similarity of the time series patterns to identify essential features that characterize abnormal respiratory motions. This paper presents our investigation into applying the respiratory motion patterns obtained by a time series clustering method to detect abnormal respirations.

researchmap

In polysomnography, sleep lab technicians visually analyze respiratory motion time series, i.e., respiratory curves, measured with a RIP belt or other equipment to record abnormal respiratory events such as apneas and hypopneas. This study aims to develop a method to detect peculiar patterns in respiratory curves to automate this analysis process. The approach of the proposed method is to divide a respiratory curve into partial time series using Singular Spectrum Transformation, then classify them based on the similarity of the time-series patterns and identify the parts that represent abnormal respiratory motions. In this paper, we compare four time-series clustering methods of k-means(Euclidean), k-means(DTW), k-means(Soft-DTW), and k-Shape for the classification process of respiratory curve segments. The experimental results suggest that classifying respiratory curve segments requires time-series distance measures thatallow for differences in period and phase but not in scale.

researchmap

Language：English   Presentation type：Oral presentation (general)  

Country：United Kingdom  

researchmap

Venue：Fukuoka  

researchmap

Venue：Online  

Recently, the Cabinet Office is advocating a concept of "Super City" that realize new lifestyles and businesses for fundamentally changing the state of society by AI. The authors have established a research unit of "Health Smart Network" at Research Institute for Socionetwork Strategies (RISS), Kansai University as a research base to meet these social demands. Specically, the purpose of this research unit is to make health promotion services smarted with "eHealth + AI", create new social communication, and contribute to extending healthy life expectancy and sustaining healthy life. In this presentation, we will outline the concept of the research unit of "Health Smart Network" and introduce various research topics on the theme of smart health promotion for the progress of future research.

researchmap

Venue：Online  

As motor function improves or declines, the way to control the body changes even for the same class of actions. To investigate what parts of the body movements characterize such changes, it is practical to summarize a high-dimensional time series of body movements into a low-dimensional time series by using the information on the coordinated behavior across different body parts. Significant differences in the summarized motion series should likely originate from characteristic differences in body control. In this study, we apply a conditional Gaussian Process Dynamic model to walking motions and propose a method to analyze the characteristics of body motions based on the dynamics of latent variables derived from a high-dimensional body movement time series.

researchmap

Venue：Osaka  

In order to gain insight into the cognitive processes that characterize the proficiency level of English as a Second Language (L2), this study experimented with measuring the eye-gaze point of L2 learners at different proficiency levels while reading English text. We applied the Singular Spectral Transformation, a change-point detection algorithm, and clustering method to the eye-gaze point time series to extract the characteristic patterns of eye-gaze movements. The analysis results suggest that the occurrence rate of particular eye-gaze patterns depends on the learners’ proficiency level.

researchmap

Venue：online  

Recently, due to the widespread effects of Covid-19 infection, social communication, such as person-to-person, person-to-society, or social connection, is being diluted. As one solution to the problems, it is expected to create new social communication utilizing advanced technologies such as ICT and AI. On the other hand, the Cabinet Office is advocating a concept of "Super City" that realize new lifestyles and businesses for fundamentally changing the state of society by AI. The authors have established a research unit of "Health Smart Network" at Research Institute for Socionetwork Strategies (RISS), Kansai University as a research base to meet these social demands. Speci cally, the purpose of this research unit is to make health promotion services smarted with "eHealth + AI", create new social communication, and contribute to extending healthy life expectancy and sustaining healthy life. In this presentation, we will outline the concept of the research unit of "Health Smart Network" and introduce various researchs on the theme of smart health promotion for the progress of future research.

researchmap

Venue：online  

This paper proposes a data analysis method for detecting abnormal breathing events during sleep from respiratory curves. The proposed method segments a respiration time series using Singular Spectral Transformations and classifies resulting partial respiratory curves based on their features consisting of autoregressive coefficients and respiratory amplitude ratios. Applying it to a polysomnography dataset confirmed that the proposed method could extract temporal patterns characteristic of respiratory abnormalities such as apnea and hypopnea.

researchmap

researchmap

researchmap

researchmap

Venue：Online  

Structural Equation Modeling (SEM) is a statistical analysis technique for fitting networks of constructs to data. The strength of extracting latent factors and estimating causal relations among them makes it accessible to analyze the structure of social problems, such as those around the economy, healthcare, and ecology. The present paper introduces SEM to the machine learning field for reducing the number of data dimensions. Too large dimensionality of the input data increases the computational load to build machine learning models. The proposed method utilizes SEM to sift variables and sort out relations between factors. The results show that the proposed method is able to identify the least number of data dimensions for machine learning-based prediction models to achieve the same or even higher accuracy as the prediction models trained with the full dimensions do.

researchmap

Venue：Online  

researchmap

Venue：Online  

researchmap

Venue：Online  

Pupil diameter, which is modulated by the automatic nervous system, has been known as a useful index of the cognitive load while it also responds to changes in other factors like illumination and viewing distance at the same time. Aiming to develop a computational method for estimating the amount of devoted mental effort from pupil variation, we propose to introduce locality preserving projection (LPP) to extract cognitive-load dependent components from pupil diameter time series across different lighting conditions. The proposed method applies the class-distance weighted LPP to decomposed time-frequency representations of pupil diameter signals that are obtained by multiresolution analysis and then singular value decomposition. The experimental results show that the proposed method can find time-frequency components of pupil variation that depend on the cognitive workload level. Reconstructed time series from the extracted components indicated pupil dilations and contractions amplified with increasing the workload of experimental mental-arithmetic tasks.

researchmap

Venue：Jeju Island, Republic of Korea  

researchmap

Language：English   Presentation type：Oral presentation (general)  

Venue：Yokohama  

researchmap

Venue：千葉市  

順向抑制の働きを抑制し，新技能の習得を支援する新旧対照法は，技能をある程度有する人に有効である．しかし，新旧対照法による技能学習は初心者には適用されていない．新旧対照法が初心者に対しても有効かを検証するために，テンキータイピング技能の習得実験を実施した．練習法として反復練習法を用いたグループと比較して新旧対照法を用いたグループのほうが成績の向上が顕著であり，順向抑制の働きを抑制していた．新旧対照法は初心者の技能学習に対しても効果があることが確認された．

researchmap

Venue：Chiba, Japan  

researchmap

Venue：Chiba, Japan  

researchmap

Language：English   Presentation type：Oral presentation (general)  

Venue：Hanoi, Vietnam  

researchmap

Venue：Tallinn, Estonia  

Autonomous driving has the potential to be the next critical technology that changes the lifestyle of a human. It has, however, not been verified whether the autonomous technology at different lower levels decreases or increases the mental workload of drivers. This paper verifies the relationship between a driver’s mental workload and level 0, 1 and 2 of autonomous driving using NASA-TLX and adjusted response time in a secondary task simulation study. Experiment results show that level 1 causes the lowest mental workload of the driver, followed by level 2 and level 0. We discuss this nonlinear relationship between the levels of autonomous driving support and the mental workloads incurred to the human driver.

researchmap

Venue：Toyama, Japan  

Combining generative adversarial imitation learning with reinforcement learning can encourage learning agents to obtain an appropriate behavior policy even in a sparse reward environment. However, if the given demonstration is imperfect, the imitation learning component may hinder the agent from acquiring the optimal policy. In this paper, we propose a learning method for overcoming this problem by shifting the learning mode from imitation learning to reinforcement learning with self-imitation gradually. The key idea behind the method is to replace the demonstration memory with the agent's past high-performance state-action trajectories after it rises to the demonstrated performance level. The proposed method was tested with a cart-pole swing-up task experiment, showing that it can enable the agent to learn a better policy than that of the demonstrations.

researchmap

Language：English   Presentation type：Oral presentation (general)  

Venue：Nagoya  

researchmap

Venue：Kyoto, Japan  

This study applied the network alignment method named IsoRank, which calculates the similarity between graph structures by optimization, to usability evaluation of user interfaces that allow users to use different operation screens depending on the progress of work. The proposed method estimates the usability of interface designs from two points of view. One is the compatibility of information resources distribution that examines whether individual screens provide just enough amount of knowledge and information necessary for the user task, and the other is the compatibility of transition structure that examines whether screen transitions are consistent with the flows of work to be performed by the user. The proposed method was tested by the experimental data that had been obtained from the user testing of operation terminals whose design was changed in their information resources distribution and screen transition structure, demonstrating that it can give a usability score close to the participant users’ feeling of use when the weight for the two compatibility aspects are adjusted appropriately.

researchmap

Venue：Kyoto, Japan  

The complexity of robot programming is an obstructive factor for the introduction of industrial robots to production sites. The human operator has to teach individual motion segments one-by-one to the robots to fit their behavior in the work environment, which is one of the most tedious and time-consuming tasks in robot programming. This study elicited user interface requirements to streamline the motion-point teaching by the hierarchical task analysis of off-line robot programming, and proposed UI functions that can meet these requirements. The proposed UI functions consist of the four-view display feature that generates orthographic views suitable for teaching motion points according to the operator's selection of the reference plane and the pointing-based robot operation feature. As a result of the verification experiment conducted using a Peg-in-Hole assembly task, the four-view display feature demonstrated its effectiveness in making the motion-point teaching more efficient.

researchmap

Venue：Toyonaka, Osaka, Japan  

In Specific Health Guidance, professional instructors provide periodic counseling and guidance to those who must change their lifestyle habits as being expected to have or having metabolic syndrome according to the results of checkups. For obtaining useful knowledge of practical guidance from written reports, this study introduces a labeled latent Dirichlet allocation method to develop a topic model of the Specific Health Guidance domain. Individual records of conducted health guidance have labels attached to indicate what kind of advice the instructor gave to his/her clients. By considering these labels, the proposed model estimates the content and distribution of guidance topics, including unlabeled ones. The estimated topic distribution that represents each guidance record in a numeric vector can provide another means to find correlations between guidance aspects and outcome measures.

researchmap

Venue：Toyonaka, Osaka, Japan  

For effectively and efficiently screening potential sleep-disordered patients, it is desired to establish simplified and easy-to-use techniques to measure the sleep state in a sufficient accuracy without elaborate facilities. Pressure sensor mats are a promising candidate but methods to estimate the vital signs of subjects from their sensing data are to be devised. For non-invasive sleep disorders screening, Generalized Morphological Component Analysis (GMCA), which is a kind of blind source separation methods, is applied to estimate the respiratory effort curve and its spatial distribution from time series of sleeping body pressure distribution. This study explores respiratory motion features that are effective for detecting and discriminating respiratory events such as apnea and hypopnea by examining the results of applying GMCA to the measurement data of the pressure sensor mat collected in polysomnography tests.

researchmap

Venue：Toyonaka, Osaka, Japan  

Moving object attractiveness map is a computational model that evaluates the possibility that a moving object in the field of view is gazed by the observer, which generates a view image in which each moving object is represented by shading according to the calculated gaze probability. Saliency map is, on the other hand, a computational model of visual attention that applies scientific knowledge of information processing in low-order visual cortexes to image processing, which generates an image that represents the saliency distribution based on low-level image features such as brightness and hue. This paper presents a method that combines outputs of the moving object attractiveness map and the saliency map to estimate the extent to which each area of the view image is likely to attract the visual attention of a human driver. The proposed model demonstrates that it can simulate a driver in different task engagement modes, such as watching out for surrounding moving objects or looking for a particular environmental object, by changing the weight of each of the two maps for linear combination.

researchmap

Venue：Osaka, Japan  

In this study, we developed an automatic sleep staging model that predicts the sleep stage of each sleep epoch from a time series of cardio-respiratory signals. The model has a Neural-CRF structure that incorporates conditional random fields (CRF) into convolutional recurrent neural networks (C-RNN). Given the sleep-stage score vector the C-RNN module output for each epoch, the CRF module generates a probable temporal sequence of sleep stages by considering the learned stage transition structure. The developed model was tested with the SHHS2 polysomnography dataset, successfully demonstrating a good performance of sleep staging (κ = 0.594±0.122, ACC = 0.749±0.079).

researchmap

Venue：Osaka, Japan  

Remote control of the unmanned helicopter becomes much easy due to the the stabilizer, but instead, the attitude motions of the unmanend helicopter becomes slow. Therefore, performance of attitude control of unmanned helicopter is not enough. In order to improve it, two methods including flap angle control by dynamic inversion method were analyzed in this paper. The stabilizer flap angle is ignored in the first method. The time scale separation between the rotor blade and stabilizer blade flapping was assumed in the second method. It was found that the second method was suitable for design the attitude controller for the unmmaned helicopter.

researchmap

Venue：Toyama, Japan  

Because Fuzzy C-Means (FCM) uses a sum-of-squared-errors objective function, it tends to equalize cluster populations, causing drifts of centers of smaller clusters to larger adjacent clusters. As a priori knowledge for clustering, the proportions of individual cluster sizes could well be available depending on application domains, on the other hand. The present paper proposes a method that can assimilate this information into the cluster structure by refining membership functions obtained by FCM. The proposed method, limited to one-dimensional fuzzy sets, modifies the position and shape of each cluster by the two-stage adjustment of its membership curve points towards the given proportions of cluster sizes. A numerical experiment was conducted to confirm the effectiveness of the proposed method, demonstrating that assimilating a priori knowledge of cluster sizes can contribute much to the accurate extraction of the original data structure.

researchmap

Venue：富山市  

経路選択に関する選好構造を特定できると，ユーザの好みを細やかに反映した推奨経路を提示可能な使い勝手のよいカーナビが実現する．本研究では，カーナビが選択肢として提示する複数個の経路をどのように構成すれば，それらに対する選択結果からユーザの選好構造を正確かつ効率的に推定できるかについて，シミュレーション実験により調査した．シミュレーションでは，一定の選好構造をもつユーザがカーナビから提示される推奨経路リストの中から最も好みに近いものを選択するのに対して，カーナビ側は選択結果に基づいてベイズ推定によりユーザの選好構造についてのモデルを逐次更新する．推奨経路リストには，目的関数の重みを固定して探索した推奨経路（距離優先や一般道優先に相当）と，選好構造の推定に基づいて重みが変化する目的関数を用いて探索した「ユーザ適応経路」が含まれる．推奨経路リストにおけるユーザ適応経路の数を操作して実験した結果，少ないやり取りの回数で精度よくユーザの選好構造を推定するためにはその数をより多くとることが有効であるが，二つのユーザ適応経路が提供されるだけでも十分な効果が得られることが確認された．

researchmap

Venue：富山市  

注視行動を形づくる因子に視覚的オブジェクトの目の引きやすさがあり，この度合いは誘目性とよばれる．平嶋らは，衝突可能性の認知に基づく注意制御の機能をモデリングし顕著性マップと統合することで，移動条件下にある人の注意を引きやすい移動体を視野画像において特定する「移動体誘目性マップ」を提案した．しかし，彼らのモデル構築と検証実験は観察者が前方に直進する場合に限定されていた．本研究では，横に曲がりながら移動する場合の観察者の注視についても実験し，観察視点の運動方向が移動体誘目性に及ぼす影響を調査した．調査結果からは，観測視点の運動に関する特徴量を用いることで，移動体注視率推定モデルの性能が向上することが確認されたまた，直線軌道にせよ曲線軌道にせよ，観察者は進行方向周辺に存在する移動体により注意を配る傾向があることがモデルの特徴量重要度から示唆された．

researchmap

Venue：Sendai, Miyagi, Japan  

researchmap

Venue：Tsukuba, Ibaraki, Japan  

Sonification is a type of auditory display techniques to perceptualize information using non-speech audio. This paper investigates the effective use of sonification in multi-task conditions, thereby to mitigate cognitive resource requirements for processing visualized task information. Experimental studies were conducted to see the effects of introducing different forms of sonification to various combinations of tracking and search tasks, comparing their efficiency with each other as well as with a visual display. The result of the studies shows the compatibility of the pitch-based sonification with the search task,suggesting that the encoding process demand to execute search tasks can be reduced much by this type of sonification.

researchmap

Venue：Tsukuba, Ibaraki, Japan  

In the present paper, we investigate the masking resistance of modulated sounds to clarify principles for designing data sonification that is tolerant to environmental noises. Modulating a tone can make the acoustic spectrum spread within the bandwidth of a human auditory filter, enforcing the masking resistance of the tone because of the auditory energy added by the modulation. Two sets of experiments were conducted to examine this effect of the sound modulation using a one-dimensional maximum search task under low signal-noise ratio conditions. As a result, Frequency Modulation (FM) synthesis was found more effective than pure tone and Amplitude Modulation synthesis. The study also confirmed that it is better to choose the modulation index for FM so that the bandwidth of the signaling tone is equal to or larger than the critical bandwidth.

researchmap

Venue：Nagoya, Aichi, Japan  

For effectively and efficiently screening potential sleep-disordered patients, it is desired to establish simplified and easy-to-use techniques to measure the sleep state in a sufficient accuracy without elaborate facilities. Pressure sensor mats are a promising candidate but methods to estimate the vital signs of subjects from their sensing data are to be devised. The present paper focuses on the estimation of the heart rate from the high-dimensional time series of sleeping body pressure and proposes to apply a blind source separation technique named Generalized Morphological Component Analysis (GMCA) to the problem. We tested GMCA by data sets with two different sleeping positions of prone and supine, demonstrating that the latter is much harder to separate the heartbeat signal from the body pressure distribution time series. A detailed discussion is made on how to improve the estimation accuracy for the supine position.

researchmap

Venue：Nagoya, Aichi, Japan  

Sentiment analysis is a technique that analyzes and extracts the evaluation information for products and services from text data. In the present paper, we propose variations of unsupervised sentiment analysis models that can estimate semantic orientations of comments (whether they are either positive or negative) together with their topics by processing text data without any ratings. The proposed models extend a standard topic model by introducing dictionaries of emotional words so that it can analyze in what point of view people are evaluating subjects positively or negatively.

researchmap

Venue：Nagoya, Aichi, Japan  

The present paper proposes a computational model of the human drivers' vision named "moving object attractiveness map", aiming at image processing technique to analyze unsafe traffic conditions that may provoke their inappropriate allocation of visual attention. Saliency map is a computational model of image processing to simulate human visual attention after scientific findings of bottom-up information processing at the low-level visual cortex, which extracts distributions of salience features from a given image and combines them to estimate which areas in the image can attract the observer's visual attention more. The proposed model adds a top-down attention filter to the saliency map that predicts the likelihood of visual objects to attract the observer's attention for possible collisions.

researchmap

Venue：Cardiff, Wales, UK  

researchmap

Venue：Kyoto, Japan  

Supply chains work with involving various factors, making it difficult to expect their dynamic behaviors. To make it clear, it is important to consider the systems from a systemic perspective; interactions of each components of those systems play important roles in their functioning rather than focusing on each of their characteristics. In this paper, we will present an extend model of FRAM: Functional Resonance Analysis Method, which is a practical way to analyze systems from the systemic perspective, based on an idea of cell automaton. Then, a case study with the proposed method to analyze behaviors of a steel production line will be carried out. In this analysis, we simulate how variabilities existing locally in the system propagate to the rest and influence a global behavior of it. In the end, we will discuss how such simulation will be utilized to design complex Socio-Technical Systems such as supply chains, airliners, and nuclear power plants.

researchmap

Venue：Sendai, Miyagi, Japan  

For the purpose to discuss design requirements for a new industrial robot programming environment, the present paper investigates two different, existing robot programming systems in terms of two cognitive perspectives: how effectively they support the human operator's decision-making by supplying externalized information resources for each phase in the robot programming, and how effectively they support the operator's transitions from one programming activity to another. The analyses can indicate what information resources are to be externalized for reducing the operator's knowledge and cognitve demand as well as what activity transitions are to be simplified for fostering the progression of designing robot programs.

researchmap

Venue：浜松市  

情動とは生体に入力された感覚刺激への評価に基づいて生ずる短期的で比較的強い反応と定義され，情動と身体反応の間には密接な関係がある．本研究では，不快情動喚起が身体技能にもたらす影響について調べることを目的としている．本稿では，身体技能として運筆技能を取り上げ，不快情動喚起刺激を負荷したときの書画結果への影響を調べた結果を示す．特に，視覚に基づく動作制御を要する課題と体性感覚のみに基づく動作制御による課題を対象として，その影響の違いを調べた．その結果として，不快情動喚起刺激は技能に影響を及ぼすことが分かった．また，技能への影響は動作に用いる感覚に依存することが示唆され，体性感覚により制御する動作では影響が現れないが，視覚に基づいて制御する動作では影響が現れることが示唆された．

researchmap

Venue：浜松市  

自動走行システムからドライバへの安全な運転権限の委譲には，ドライバの正しい状況認識の形成が不可欠である．本研究では，運転を引き継ぐドライバが正しい状況認識を形成するまでにどれほどの時間を必要とするのかを検討するために，SAGAT を応用してドライバの状況認識形成過程を調べる実験を実施した．実験では，自動走行中の運転席からの視界をシミュレーションした動画像を提示し，それを任意のタイミングで隠した後，質問票を使ってその時点で獲得されている観察者の状況認識の正確性を調査した．調査の結果，周囲の車両が加減速を行い自車との位置関係を時々刻々変化させる「動的」な交通状況では，3 秒ほどの短時間で正確な状況認識を確立することは困難で，引き継ぎ後の運転判断をドライバが誤る可能性が高いことが確認された．

researchmap

Venue：浜松市  

現在, 医療機関で実施されている精密な睡眠の検査方法としてPSG 検査（Polysomnography）があるが，被検者への負担が大きいことが指摘されており，より簡便な睡眠の計測方法が求められている．本研究では，800点の圧力測定点を持つマット式センサを用いて，睡眠評価のための基本的な生体情報である呼吸数と心拍数を推定する手法を提案する．手法として，体圧の時系列に対しフーリエ変換を施し，ピーク周波数を調べることで呼吸と心拍の周波数を求める．実験を行った結果，呼吸数はよく推定できるが，心拍数については信号が微弱であるため解析に工夫が必要であることが明らかになった．

researchmap

Venue：浜松市  

本研究では，加算合成を用いて2 次元データを可聴化するパラメータマッピングの好ましい設計について，1次元の位置決めタスクを用いて実験調査を行った．位置と速度の組で表される動的システムの状態を2音の加算合成によって可聴化する場合に，目標位置からの偏差を一つの音の大きさに対応づけるのに対して，目標速度からの偏差をもう一つの音のどのパラメータに対応づけるのが適当かを比較検討した．比較は音の大きさに対応づけるマッピングと音の高さに対応づけるマッピングの間で行われ，実験の結果，後者の優位が確認された．

researchmap

Venue：浜松市  

テキストデータから特定の製品やサービスなどに対する評判を自動的に抽出する技術は評判分析と呼ばれる．また，文が示す内容が肯定的か否定的かは感情極性と呼ばれ，単語や文書に潜在する感情極性の推定はテキストマイニングの一分野として研究されている．評判分析手法の一つに，文書の低次元化や要約を実現するトピックモデルと単語レベルの感情極性辞書を組み合わせた手法が存在する．この手法では，文書情報を抽出する際に単語の出現順序を省略しているため，否定表現の係り受けによる感情極性の反転を捉えるのは難しい．本稿では，トピックモデルを用いた評判分析において問題となる否定表現の処理法として二つの手法を提案し，それらの効果を検証した．一つは否定表現による感情極性反転の検出法であり，もう一つは逆接の接続助詞における文の分割である．商品レビューのデータセットを用いた検証実験の結果，レビューテキストに基づいて商品の評判を分類するモデルの分類精度が二つの手法によって改善することが確認された．

researchmap

Venue：Yonezawa, Yamagata, Japan  

Human drivers get most of the information about their world through vision. Therefore if we are able to specify regions to likely attract their attention, it will lead to advanced driver assistance systems, such as by detecting situations where their visual attention may get highly distracted. Saliency map is a computational model that can extract visually conspicuous points or regions based on low-level image features. In this study, we examined how well saliency maps generated from in-vehicle camera images correspond with regions of interest human observers have paid their visual attention to. As a result, we found that incorporating opticalow features in saliency maps can improve the model's performance while it still has some disagreement with actual visual behavior.

researchmap

Venue：Yonezawa, Yamagata, Japan  

Sleep stage pattern provides important clues in evaluating sleep quality. This study is performed to develop a classifier that estimates sleep stage from heart rate variability. The classifier is based on Long Short-Term Memory (LSTM) which is a kind of Recurrent Neural Network, and classifies sleep stages into wake, Non-REM sleep, and REM sleep from 13 features derived from heart rate variability. As a result of the performance evaluation of the classifier, it was confirmed that the recursive connections of LSTM successfully grasps the context-dependency of sleep stages, and especially improves the classification accuracy of REM sleep.

researchmap

Venue：Yonezawa, Yamagata, Japan  

Sleeping mat that can measure vital signs of a person sleeping on it is one of promising alternatives to PSG (Polysomnography) for simplified sleep assessment. In this paper, we investigate computational methods to predict the respiratory and heart rates of sleeping subjects from measurements by the pressure sensor array embedded in a sleeping mat. Two computational methods were compared by experiment. Whilst both methods utilize frequency domain analysis to extract those components of body movements that are induced by breathing and heartbeat, they are different in the way to choose sensor nodes to take into account for computation. As a result, it was found that the respiratory rate can be predicted in a higher accuracy by using the sensor node with the largest pressure value for each instant of time and that the heart rate prediction accuracy is not affected by the way to choose nodes.

researchmap

Venue：Otsu, Japan  

Information scent represents the semantic relevance of labeled options to destinations of navigation tasks and has a potential as a computational basis to simulate human users’ navigation decisions to search for items in a menu hierarchy. In the present paper, we propose a method to inspect the usability of menu hierarchies by using the information scent model. The method applies a software agent that makes scent-based decisions for simulating menu item selection by unlearned users. Every function that can be launched from the menus are thoroughly explored by the agent to identify ill-designed part of the menu hierarchy that makes it difficult for users to accomplish particular tasks.

researchmap

Venue：Kyoto, Japan  

In recent years, small-scale rotorcrafts are widely used for various purpose. In particular, inspections of infrastructures such as bridges or tunnels by aerial vehicles are highly demanded. During inspection activities, aerial vehicles must approach their walls. In such situation, walls affect on the airflow around the rotor and also aerodynamic characteristics of the rotor. We conducted experiments to measure thrust, rotational speed, and required power of the rotor near a vertical side wall or a ceiling wall. It was found that the thrust and the rotational speed of the rotor do not change when it approaches to the vertical side wall. On the contrary, we found that thrust and rotational speed of the rotor increases and the required power decreases as the rotor approaches the ceiling wall. The results suggests that the vertical side wall effect is considerably small but the ceiling wall effect is large. The ceiling wall effect may be utilized to extend the flight time for inspection without equipping large batteries.

researchmap

Venue：Kyoto, Japan  

Development of technologies (e.g. automation) makes the mechanical systems complex, building Socio-Technical Systems. For the safety of those systems, feasibility of procedures against a specific context plays an important role. E. Hollnagel proposed FRAM: Functional Resonance Analysis Method for this issue. In this paper, we show a practical methodology of FRAM which we proposed before: Fuzzy reasoning is applied to FRAM in this methodology, and we built a simulator of FRAM by implementing it. Also, we investigate how the methodology contributes to the safety of Socio-Technical systems.

researchmap

Venue：Kyoto, Japan  

The automation of route navigation led to the implementation of in-car navigation systems, widely used nowadays. However, their interactions with the drivers still cause some confusion, thus damaging road safety. In this paper, we first propose a categorization of the interaction issues found in the literature into two categories: “context awareness issues” and “expression issues”. We then focus on “context awareness issues” and propose that they are mainly due to a difference of perception means between the automated systems and the human drivers. Finally, we suggest that joint attention between these two agents can improve the shared amount of context information and give research directions to test this hypothesis.

researchmap

Language：English   Presentation type：Oral presentation (general)  

researchmap

Application no：特願2006-070772  Date applied：2006.3

Announcement no：特開2007-248199  Date announced：2007.9

researchmap

Application no：特願2004-268845  Date applied：2004.9

Announcement no：特開2006-85389  Date announced：2006.3

researchmap

Award type：Award from Japanese society, conference, symposium, etc.  Country：Japan

researchmap

Award type：Award from Japanese society, conference, symposium, etc.  Country：Japan

researchmap

Award type：Award from Japanese society, conference, symposium, etc.  Country：Japan

researchmap

Award type：Award from Japanese society, conference, symposium, etc.  Country：Japan

researchmap