This study illustrates a system for blind persons to enjoy playing maze games. Age of a totally blind person lost his/her sight (vision) is different from each other. He/She was a totally blind when he/she was born. On the other hand, He/She might lose his/her sight (vision) around twenty years old, and similarly in old age. This paper presents that the authors develops a system for blind persons to enjoy playing maze games by using a haptic device.

View full abstract

In this article we describe a method of sharing information between a tactile map (for navigation purpose) on a paper and an online-map on the information terminals (say a smartphone used by the sighted) via "electronic pen" technology (which is commonly used as an input method of hand-written characters). It is based on a synchronizing of a touched point on the tactile (paper) map with the corresponding point on the online-map achieved through the following steps: 1) Create a road-only map in SVG (scalable vector graphics) form from the given (target) online-map and copy it onto a tactile (i.e. touch sensible) paper, 2) point on the tactile map by touching electronic pen in it, then the coordinates (x, y) of the touched point is synchronized by the electronic pen synchronization mechanism and mapped onto the corresponding point in the online-map (we assume that a proper calibration is done beforehand between the road-only map and the original-online map). Thus, various information of the online-map along the roads can be provided to the visually impaired user in the text form from the online-map in the similar way as to the sighted user.

View full abstract

In this paper, we propose a macrophage simulator for analyzing the behavior of macrophages. Our simulator is based on the evolutionary computation methods. The moving image data extracted from rats by using two-photon microscopy are available for the training data. Several rules imitated macrophage’s behavior are automatically generated by a evolutionary computation method using such moving image data.

View full abstract

Total knee arthroplasty is a surgery to replace original injured knee to a prosthesis. In this surgery, it is important to select a proper size of the prosthesis as fitting to the patients. This selection strongly depends on the experience and feeling of surgeon. This study aims to develop a system which can support to select the proper size. Therefore, this study examines an effect of varus and valgus motion to the tibial component of the prosthesis by using the developed system. In general, when the force concentration happens on the tibial component, it results in wear of the prosthesis after surgery and can shorten lifetime of the prosthesis. This system can contribute to avoid these problems.

View full abstract

Currently, papers have reported measurement systems for joint motion in diagnosing. Although measurement systems with optical device such as camera is used widely, it has disadvantage that the precision decreases in case of presence of obstacle between a marker and a camera. Therefore, it is not applied to a clinical scene that hands and arms of an examiner interfere the vision of the camera. When medical doctors performe manual test to diagnose an injured joint, the hands of the doctors easily hide the makers. This study proposes an analysis method of knee motion by extracting only the region of the femur and tibia from images taken by an optical device. Then it can be used as measurement system for the manual test.

View full abstract

Tactile graphics are images that use raised surfaces so that a visually impaired person can feel them. Tactile maps are used by blind and partially sighted people when navigating around an environment, and they are also used prior to a visit for orientation purposes. Since the ability for reading tactile graphics deeply depends on individuals, providing tactile graphics individually is needed. This implies that producing tactile graphics should be as simple as possible. Based on this background, we are developing a system for automating production of tactile maps from hand-drawn figures. In this paper, we first present a pattern recognition method for hand-drawn maps. The usability for our system is then evaluated by comparing with the two different methods to produce tactile graphics.

View full abstract

Visually impaired individual, when he walks on a pathway alone, recognizes path conditions through using a white cane. The purpose of our study here is to reproduce haptic sensing of his hand during the investigation in which white cane is grasped. First about our virtual reality (VR) system for this purpose. In our VR system we prepare "studded paving blocks to aid the visually impaired" as pavement, a "white cane" and a "hand" to grasp and manipulate the cane on the block. An actual hand worn a data glove on it is superposed on the virtual hand by the system. Thus the touch and investigation on the paving blocks by the white cane in the VR space is conveyed to the actual hand via data glove. We also show some experimental vibration value and haptic value data and some skin surface voltage of EMG (electromyogram) data.

View full abstract

This study presents support system for low vision persons to move autonomously. The support system which the authors developed can let low vision persons know information of pictogram around them by voice and expanded picture, by using tablet with camera and an object-recognition-server through the Internet.

View full abstract

Developing an elderly monitoring information system is described in this paper. The developed system is expected to solve the problem of caregiver shortage. The system assists the single aged person by using the information network. There are three features in the system. Non-invasive sensors are used. A third-person can monitor the bahavior of the single aged person on a web page in real time. The outline of the elderly monitoring information systems and the state are described.

View full abstract

In this paper, we propose a human condition monitoring system using a small lightweight MEMS device. While MEMS device can measure several kinds of data, we use data from three-axis acceleration and ECG. Some of the measured data are used for training of fuzzified neural networks(FNN). Then long-term monitoring data are evaluated by the trained FNN. From the outputs of the trained FNN we estimate subject's physical condition.

View full abstract

A pelvic belt is a tool of first aid for the pelvic fracture patients. This belt can stabilize the unstable pelvis and control bleeding from the arteries and organs inside the pelvis. The purpose of this study is to develop an evaluation system for the pelvic belt. Especially we evaluate the pressure around greater trochanter under pelvic belt. It is reported that a pressure by tightening the belt increases the risk of tissue damage with pelvic belt. A mechanism which decreases the pressure by the tightening is required. We performed experiments that recorded pressure around greater trochanter of right and left when subjects turned body. As a result, we confirmed that the pressure of the contralateral greater trochanter of the direction tends to increase with the same direction of subject turned.

View full abstract

This paper describes fuzzy evaluation method for criteria dependency in human health conditions by changing medical checkup reference. In our previous paper, we proposed an evaluation method based on fuzzy logic for health checkup data. This method converted health checkup data into fuzzy normal degree to evaluate multivariate data. We defined the fuzzy normal degree as an attribute value in closed interval [0, 1] by using fuzzy membership function which was defined by diagnostic criteria. Furthermore, total health indices which are defined by operation of the fuzzy normal degrees are treated as the same scale. These indices can express human health conditions in health checkup result using fuzzy set. In this paper, to investigate fuzzy normal degrees by changing to new reference interval, we employ new reference interval and form new diagnostic criteria, and visualized health changes by calculating the total health indices using our proposed method. As the result, we succeeded in visualizing health change by fuzzy normal degree and fuzzy health indices, and confirmed quantitatively that disease patients decrease by changing to new reference interval.

View full abstract

We propose a dynamic feedback linearization of a car-like robot as a non-holonomic system with a piecewise bilinear (PB) model. We also apply a method for a tracking control based on PB models to the car-like robot. Although the controller is simpler than the conventional I/O feedback linearization controller, the control performance based on PB model is the same as the conventional one. Examples are shown to confirm the feasibility of our proposals by computer simulations.

View full abstract

Colors give various psychological effects. The effects are "stress" or "healing" for people. As psychological effects of color, "blue has the effect of stabilizing the spirit" or "red has the effect of exciting the nerve" have been reported. In this paper, we created a LED CUBE was controlled by the FPGA. Our LED CUBE aims to realize "healing" as a pre-stage by a variety of light illumination. We will do demonstration of light illumination.

View full abstract

This research investigates the stabilizability of polynomial fuzzy systems by searching control Lyapunov functions (CLF) for the systems numerically using sum-of-squares (SOS) optimization technique. At first, we express a nonlinear plant with a polynomial fuzzy system. Next, a global stabilizability criterion represented in SOS conditions is proposed to search for a CLF for the system. In our approach, path-following method is employed due to the existence of bilinear terms in the CLF conditions. Finally, a simulation example is given to show the effectiveness of our proposed method.

View full abstract

A pursuit problem is a multi-agents' benchmark problem, where four hunters pursue and capture the prey in a cell environment. We have proposed a staged view that reflects a human's view where we can see an object easily in the neighborhood but more difficult in the longer distance and easily in the center direction but more difficult in the righter and lefter directions in both cell environment and real number environment. We showed that hunters can capture the prey even with the staged view. In the previous research, if hunters lose the prey, they select no special actions. However, human will try to surmise the prey's position. In this paper, we apply a fuzzy Q-learning to hunter's learning. Hunters surmise prey’s lost position from the previous prey's position and their own action by fuzzy Q-learining. We simulate this problem, and have a result better than the previous one.

View full abstract

In fuzzy Q-learning, it is very difficult to design a state space for a given problem. We have, therefore, proposed a dynamic fuzzy Q-learning with generating, tuning and removing fuzzy sets and a pairs of state and actions using forgetting facility. We have also applied a conventional fuzzy Q-learning to a car agent of IEEE CEC 2007 Car Racing Competition, which is very difficult to define fuzzy sets of Q-table. In this paper, we apply the dynamic fuzzy Q-learning to Car Racing Game in order to acquire a appropriate state space. We have a good result and compare it to the conventional fuzzy Q-learning.

View full abstract

Car racing game is a game of computer programs in IEEE CEC 2007 Car Racing Competition, where two car agents compete with each other for passing through waypoints in a two-dimensional real-number plane. Generally it is important for the car agent to pass through the current waypoint with smaller steps, while it is more important to pass through the current one so as to pass easily through the next one when the car agent can much more easily pass through the current one than the opponent. To learn such two different actions, we have applied a fuzzy Q-learning to a car racing game in a single trial, where the Q-table is updated by two kinds of rewards, and car agent's actions are strongly affected by frequently given reward. We, therefore, propose a fuzzy Q-learning with two Q-tables according to two kinds of rewards. This selects an action with two Q-tables and updates two Q-tables according to two kinds of rewards. We show a simulation result in this method.

View full abstract

Game logs have fruitful information on the game situation. It is no doubt that important information on the team strategies for winning the games are embedded into the game logs. In this paper, a method is proposed that extract team strategies from game logs. The extracted information is used to train non-linear functions.

View full abstract

In this paper, we propose a variable neighborhood model for reflex agent control, which enables reflex agents to make planning for attaining their goal. The model consists of three layers: (1) topological space (2) agent space and (3) linear temporal logic. In order to make reflex agents which have an ability of goal-attaining superficially, we focus on local properties of the binary relation between agents. As a result such a sequence of low-level behaviors, that is, keeping or abandoning of local properties of binary relations, generates higher level actions as if they could attain a goal with such actions. By simulation, such low-level behaviors are shown to generate attaining some goal like target chasing and cooperation.

View full abstract

In this paper, we study kernel functions for clustering with nominal data sets, which are consisting of objects described by nominal attributes. For such data sets, it is important that clusters have simple representations of the attributes, as well as clusters of similar objects are obtained. To obtain such clusters, we propose a kernel function defined through Boolean formulas which reflect discernibility of clusters. We apply the proposed kernel function to clustering under cannot-link constraints.

View full abstract

It is already known that a BCK-algebra has the partially-ordered structure equipped with the least element, also it has a lattice structure when it is bounded and commutative, and when it satisfies the condition (S) and the commutativity. In this report, the authors create some concrete examples of BCK-algebras with different features on a representative simple lattice, and we investigate the characteristics of BCK-algebras from the viewpoint of the construction of lattice operations.

View full abstract

In this report, first of all, the authors revisit the concepts of a t-norm and t-conorm in a bounded partially-ordered set, and newly define a quasi-t-conorm in a lower bounded partially-ordered set. Next, we show that some kinds of BCK-algebras are equipped with a t-conorm (or a quasi-t-conorm), and a t-norm. As the main results of this work, it is clarified that a union operation of multi sets is realized by a quasi-t-conorm in a BCK^S-algebra, and a union/intersection operation of fuzzy sets is realized by a t-conorm/a t-norm in a bounded BCK^S-algebra, respectively. Finally, some concrete examples of set operations for multi sets and fuzzy sets are presented.

View full abstract

Rough approximations, which consist of lower and upper approximations, are described under objects characterized by possibilistic information that is expressed by a normal possibility distribution. Concepts of not only possibility but also certainty are used to construct an indiscernibility relation. First, rough approximations are shown for a set of discernible objects are shown by using the indiscernibility relation. Next, a set of objects characterized by possibilistic information is approximated. Consequently, rough approximations consist of objects with a degree expressed by an interval value where lower and upper degrees mean the lower and the upper bounds of the actual degree. This leads to that the complementary property linked with lower and upper approximations holds, as is valid under complete information.

View full abstract

In rough set approaches, decision rules are induced from a given data table showing the relation between attribute values and classes of objects. The induced decision rules are used for the classification of new objects by their attribute values. However, because the given data table does not always include all possible patterns, some of new objects do not match any decision rule conditions. In those cases, no estimated classes are obtained. Classes of such new objects are estimated by using partially matched decision rules. In this paper, to enhance the classification accuracy, we propose to add supplementary rules which can work well for the unmatched new objects in the class estimation. We define the supplementary rules and propose a method for inducing them. We examine the performance of the classifier with supplementary rules by comparisons with the classifier without supplementary rules.

View full abstract

The authors have proposed a rule induction method using discernibility clustering. In this method, we apply clustering to the set of the positive objects using a similarity measure reflecting discerni- bility from the set of negative objects. Comparing LEM2, which is a conventional rule induction method, the proposed method obtained simpler rule sets, but it tends to be worse with respect to classification accuracy. In this paper, we improve clustering accuracy of the proposed method by preventing overfitting of obtained rule sets.

View full abstract

We are investigating rough sets and rough set-based computing, and proposed a framework Rough Non-deterministic Information Analysis (RNIA). In RNIA, NIS-Apriori is the core algorithm, andthis article adds the parallel functionality to this algorithm. According to some experiments, we show thevalidity of this parallel functionality.

View full abstract

Rough set-based interrelationship mining, proposed by the authors, provides an approach to extract characteristics based on interrelationship between different attributes from decision tables. The main idea of interrelationship mining is to clearly represent the interrelationships between attributes with respect to various binary relations between attribute values by newly introduced attributes, however, the number of the new attributes tends to numerous. In this paper, we discuss a method to select binary relations for representing the interrelationships between attributes.

View full abstract

Robot technology is very useful for learning task at the engineering education in the field of electronics and mechatronics, because it is consolidation of various engineering factors. Moreover, encouraging and maintaining the motivation is important in engineering education, because it affects the effect of education. Thus, in a robot production subject at universities and high schools, robot competition is employed for aiming to motivate students and to encourage intelligibility of technology and knowledge. In this research, we aim at developing the educational tool and verifying the effect for educational tool using comic. We report the verifying result of effect of developed educational tool.

View full abstract

The report of Ministry of Economy, Trade and Industry shows that industry expects cultivation of embedded system developers. This study aims at propose of effective method to cultivate embedded system developers and develop it in National College of Technology (NCT). We have noted robot, one of the embedded system, as educational material. Then we propose education method using walking robots that carries I/O and controller, and can be implemented advanced information processing technology. In 2012, we had planned exercises for higher-grade students and conducted. In accord with the result, we have planned exercises about basic control of a biped robot to teach programming technique which is essential for implementation of embedded system. The targets of the exercises are lower-grade students in NCT. The effectiveness of proposed method can be confirmed from the result of questionnaires. And, it is found that clarification of problems and its matters is important for effective education. The exercises for higher-grade students also have conducted after reconstitution. In this report, the detail of conducted education method is shown. And the effect of proposed method is evaluated by subject's answers for given tasks and their feedbacks.

View full abstract

Various kinds of robots have been developed as computers and information processing technology advance. Operations in extreme environments such as disaster areas, space and ocean are getting one of the practical solutions for hazardous missions. The underwater robots are one of the extreme environment robots that are expected as one of solutions for underwater activities; maintenance of underwater structures, observations, scientific research. These activities require robots that can cover large area deep under ocean water. Their efficiencies have been investigated during recent decades and are proven by ocean experiments. However, the robotic system including the support vessels is still large in scale, and is not so easy to handle without number of researchers. In this paper, we describe the design of "DaryaBird" developed to be easy to handle, small-scaled underwater robots that can operate only with two researchers. In addition, experimental results and mission strategies for Robosub2014 are reported.

View full abstract

This paper describes control system for autonomous mobile soccer robot "Braves-1". Braves-1 is developed by Nippon Bunri University for using RoboCup middle size league. The features of the robot are omni-direction wheel, low center of gravity and extensible desing. In this paper we introduce the control system for "Brave-1". And we also report the difference of simulation result and real mobile robot result.

View full abstract

In the field of RoboCup Middle Size League, ball handling mechanism is very important for holding ball and dribbling ball smoothly. We made ball handling mechanism based on RoboCup MSL rules. We employed a silicon tire controled by motor to rotate the ball with natural direction and to control the ball with robot motion. Thus we must calculate the motor speed of handling system based on robot parameter. In this report, we introduce the hardware of handling system and calculation method of robot parameter.

View full abstract

In the field of RoboCup Middle Size League, self-localization is important to estimate own position including goal and robot positions and to decide strategy. Basically, we estimate the self-position with the image information, the environment information and the field information. However, the above method, cannot finished process in real time, so it is difficult to perform the strategic behavior and cooperative operation. In this paper, we construct the self-localization method in real time by applying genetic algorithm.

View full abstract

This paper describes an effect of knowledge transfer rate for transfer learning of heterogeneous multiple robots. In recent years, various researches for a real-world application of multiple autonomous robots system have been discussed. In a dynamic environment such as RoboCup soccer, a Multi-Agent Robot Learning System (MARL) using reinforcement learning is useful to obtain adaptive behaviors autonomously. The transfer learning shows an effectiveness of reuse of knowledge to several issues of MARL such as a long learning time. However, in the transfer learning of heterogeneous multiple robots, the effect of the transfer rate, which means a ratio of knowledge transfer between robots, is not discussed. In this paper, we confirmed feature of change of transfer rate in the heterogeneous multiple robots' environment by conducting a computer simulation.

View full abstract

RoboCup is an international join project to promote Artificial Intelligent (AI), robotics, and related field. It is an attempt to foster AI and Intelligent robotics research by providing standard problems where a wide variety of technologies can be integrated and examined. In RoboCup, soccer game is selected as a main topic of research, aiming at innovations to be applied for socially significant problems and industries in the future. We also have organized RoboCup team "Hibikino-Musashi" Middle-Size League soccer competition. In this paper, we introduce the next generation soccer robot platform "Musashi150" with ball keeping ability and fast movement.

View full abstract

Picture books play an important role as a material that develops children's linguistic competence. Thus, English picture books can be considered to be indispensable in the children’s English study. In this paper, metrical characteristics of some English picture books were investigated, compared with English textbooks for Japanese junior high schools. In short, frequency characteristics of character- and word-appearance were investigated. These characteristics were approximated by an exponential function. Furthermore, the percentage of Japanese junior high school required vocabulary and American basic vocabulary was calculated to obtain the difficulty-level. As a result, it was clearly shown that the English picture books have a similar tendency to literary writings in the characteristics of character-appearance, and some books are more difficult than English textbooks.

View full abstract

The Nintendo DS is a dual-screen handheld game console developed and released by Nintendo. Because it introduced distinctive new features to handheld gaming such as a touchscreen, a built-in microphone, and so on, it has been used not only as a game machine, but also as a teaching material in educational site. In this study, in order to search for the possibility of using the DS at a university, we carried out an investigation concerning English learning DS software. As a result, it has been cleared that both the students who are good at English and those who are not so think learning English using the DS is enjoyable and effective.

View full abstract

Onomatopoeias play an important role in daily conversation in Japanese. Onomatopoeias convey the subtle nuances. Sinc e Japanese is a language rich in onomatopoeias compared with other languages, words corresponding to the Japanese onomatopoeias is small in other languages. For this reason, it is clearly essential for Japanese learners to learn Japanese onomatopoeias. There is a web site which Japanese learners can learn 72 onomatopoeia words through skits as an existing learning method. However, methods of learning Japanese onomatopoeias currently has not yet been established. Thus, the system which supports learning of onomatopoeias for Japanese learners is required. In this paper, in order to consider the best method for learning onomatopoeias for Japanese learners, we investigate how Japanese native speakers choose onomatopoeia words. In addition, we have developed an application which Japanese learners can learn and search onomatopoeias with for tablet devices.

View full abstract

An onomatopoeia is an useful linguistic expression to describe sounds, conditions, degrees and so on. Japanese has rich onomatopoeic expressions. They are frequently used in daily conversations. An onomatopoeia in a region may have a different meaning from one in different regions even if it is the same expression. Therefore, we attempt to investigate practical usage of onomatopoeias taken into account for the regional difference. However, general morphological analyzers cannot always recognize onomatopoeias. In this paper, we propose a method for automated extraction of onomatopoeias in regional assembly minutes. Although most previous work treats only four-letter onomatopoeias, our work treats not only four-letter onomatopoeias but also ones shorter than four letters.

View full abstract

In this paper, we aim to develop a supporting system for quizzes, which are used for grasping students' understandings, in class time. We developed an automatic keyword extraction method from descriptive answers of a quiz. The method refers to frequency of words in answers and general documents (corpus). We discuss an effective database of frequency in corpus for our extraction method. We conclude that (1) Japanese WikiPedia is a good for corpus, (2) the system can extract compound words by connecting morphemes.

View full abstract

In this paper, we propose a QA system that returns a suitable answer to a question submitted by a user. It divides a question sentence into units, each of which is a tuple of a noun, a particle, and a predicate. The QA system returns the answer whose assigned units match the units of the question. In order to absorb the variant of particle usage, we applied cluster analysis to questions in Yahoo! Answers to group particles. In order to maintain the relationships between units and answers, we utilized topic maps.

View full abstract

Increased demand for web advertising has resulted in a corresponding increase in the need to develop personalized advertisements targeted at individuals online. We propose an automated advertising slogan selection system that can satisfy this requirement. Many customer reviews and comments are available publicly on online shopping sites. The proposed system uses content mining to extract favorable reports from the web and arranges the data into a specific knowledge representation structure to improve the advantage of the target product. For a particular business, the proposed system first extracts tuples, composed of elements that express the knowledge representation from each user-written review. Then, these tuples are selected using a frequency-based approach and emotion corpus. Subsequently, for each tuple, advertising slogans are chosen from the advertising slogan corpora using a neural network. For verification, we used data from an electronic commerce website for hotels to evaluate two aspects of our system (namely, quality of selected tuples and advertising slogans). The results of the experiments confirm that the proposed system can extract suitable tuples when the given data are sufficient. It can also retrieve slogans even when their meanings are convoluted.

View full abstract

There are many regional dialects in Japan. And we often talk with such a dialect in casual conversation. On the other hand, there are some sentences with dialects in regional assembly minutes. It is difficult to recognize dialects correctly using a morphological analyzer with a common dictionary. If we can get correct results of word segmentation in various regional assembly minutes, then we can attempt to share problems or discussions between various local municipalities using their assembly minutes. In this paper, we describe analyses of dialects in regional assembly minutes and results of word segmentation of them.

View full abstract

It is well known that difference in memory strategy has an influence on the cognitive structure and learning performance. After learning concept group to make hierarchical structure by the repetition of the concept map completion method, the results of the test showed that the group which regenerated hierarchically were the most superior.

View full abstract

Imitation is commonly observed in various motor learning. Although it is a computationally difficult problem that involves solving self-other correspondence problem and calculating appropriate motor commands that reproduce the observed motion, human adults and children seem to be somehow able to effectively perform this process. In this study, we have found that imitative motor learning of a sequential finger movement, similar to piano playing, can be facilitated when the learner is under the body ownership transfer illusion. We presented a computer-generated animation of a hand performing the movement as the instructor, and induced the illusion by visuo-tactile stimulation using a moving computer-generated cone-shaped object and a vibration motor. We tested two conditions, synchronized and unsynchronized visuo-tactile stimulation, and found that the learning was significantly improved when synchronized stimulation was given. The result suggests existence of a common mechanism shared by both perception of body ownership and imitative motor learning.

View full abstract

In daily life, we often have been reading color digital comics. But, a lot of color digital comics don't support color blindness. Especially, in digital book, there is not evaluated color combination corresponding to color blindness. We measured distribution of gaze direction of when be reading color digital comics, and investigated for color digital comics corresponding to color blindness.

View full abstract

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.

View full abstract

Presently, elderly over the 65 years old is about 25 % of the total population, and increase of elderly people living alone has been a problem in Japan. A study of sensor system for monitoring elderly people living alone is attracted attention. In the previous studies, a system, which some sensors are attached in the room in order to monitor people, can monitor only in the room and costs expensive. Various wearable sensors system is able to monitor anywhere, however it is tiresome to wear. This study uses a pressure-sensitive conductive wear, which can be used as a sensor. This electrical resistance of the wear varies effecting on expansion and contraction; it can measure the wearer's movement. This study recognizes wearer's feeling and action by wearing it. This study measures the total six points, both elbows, both shoulders, two waists. This study recorded as the state of the wearer's emotion, the degree of concentration, location, and behavior at that time. This paper sho ws the result of the recognition by machine learning.

View full abstract

Imitating the human intelligent action is one of the best ways to develop the intelligent robot. However, it is difficult to obtain high-lever intelligence for the robot due to that the human misoperations cannot be get rid of by only using the imitation method. In this paper, in order to enable the robot have the human intelligence, a knowledge extraction method required by the development of high-level intelligent robot was developed considering the human misoperations. Finally, the effectiveness of the proposed method was verified by experiments.

View full abstract