産業界やスポーツ・芸能・芸術分野等，様々な分野においてエキスパートや熟練者と呼ばれる人には，長年の経験や練習で得られた '勘' や 'コツ' と呼ばれるものを持っており，それらのなかには人間の体が覚えているが，定量的に人に伝えるということがむずかしい技能も含まれている。

In this paper, efforts of transfer skills in traditional crafts are introduced. A case example of a skill analysis of a paper making craftsman among traditional crafts is shown. Motion capture system measures the difference in motion between a trainee and a craftsman. The influence on paper quality from motion data will be analyzed.

In 2011, the Ministry of Education, Culture, Sports, Science and Technology showed a comprehensive promotion policy "Information vision of education" on information education for fiscal 2020. It is necessary to research and develop such as the evaluation indicators and the evaluation methods for the objectively evaluating the transformation of children by "Information vision of education". In this paper, the research results related to "visualization and support of child behavior data" that integrates the education field and the engineering field are introduced. Specifically, "support teaching material for nurturing cooperativeness" and "assist education of swing riding by using a teaching material" are introduced.

PID control is widely applicable in industrial field because of few control parameters and comprehensibility of physical meaning of control input. Moreover, experience and intuition of veteran engineers are needed to aim for good control performance, and the keyword of cultivation of their abilities is data. Therefore, this paper introduces recent results on utilization and measurement of human skill data, and mutual buildup between data-driven control and model- based control in the control technical committee of the Institute of Electrical Engineers of Japan. Finally, the relation and the role among the previous research fields and smart control technology are studied.

Since a decrease in muscle mass of elderly people, the physical burden of them in standing-sitting motion of a seat are increasing than young people. For this problem, several standing-sitting support systems have been developed, and evaluations of these systems focusing on a reduction of physical load have been discussed. However, these evaluations are of support systems with an active actuators or of support systems when the use a handrail and a standing aid. Thus, an evaluations of a support system with only passive actuators has not been reported. In this paper, in standing motion with using a standing-sitting support system with passive rotational seat mechanism which our group propose, we objectively evaluate an effectiveness of the passive rotational seat mechanism by statistical analysis.

In the control education, students are able to understand the control theory intuitively through the education using experiments. In this study, the authors develop an experimental device for the control education. By using the device, it is possible for students to comprehend the control through experience and to enhance the students comprehension in the control. In the device, a controlled object is floated by rotating a propeller, and the floating level is adjusted by controlling the angular velocity of the propeller. Further, students recognize the control effect resulted from the control actuation since the propeller is rotating while the object is floated. The floating device is developed of commercial components, and hence, an experimental setup can be prepared at a low price. Students are educated on the control theory using the device, and its effectiveness is validated. In the control education, the device operation method and the control theory are explained alternately. From the questionnaire before and after the experiment, the effectiveness of the control education using the device is demonstrated.

In process Industries, there is a growing trend toward low carbon society. Especially in oil refinement and petrochemical industries, furnaces consume a lot of energy. In order to reduce the energy consumption, excess O₂ has to be reduce safely. However, high gain O₂ control cannot be done due to L/T>1. Therefore, conventional O₂ controllers are operated with low gains. That is, high stable control performance for disturbances has not been able to obtain, then minimum excess O₂ control cannot be achieved. The proposed method which utilizes the estimated O₂ variable can derive much better control performance. Moreover, since the proposed estimation technique consists of a physical model part and a stochastic soft-sensor, both estimation accuracy and reliability for extrapolation can be obtained. The validity of the proposed scheme is verified through actual application results.

This paper proposes a design method of strong stability self-tuning controller based on on-demand type feedback control. For safety in industrial applications, although it is important to consider on-demand type feedback control system, the previous papers about on-demand type feedback control did not consider the influence of noise and fixed the design parameter to constant value. Therefore, this paper extends the design parameter of on-demand type feedback control as stable rational function through the design method of strong stability system using coprime factorization. Moreover the self-tuning controller of the proposed method is given and the control result with noise is shown by numerical example.

This paper deals with a design problem of parallel feedforward compensator (PFC) without controlled system model. PFC has been proposed to overcome the problem that controlled system does not satisfy the conditions of almost strictly positive real (ASPR) based adaptive control system. In most conventional methods, the design parameters are decided by the designers knowledge or trial and error of designer by using controlled system model. In the proposed method, PFC parameters will be optimized via differential evolution (DE) by using the given input/output data set of controlled system.

This paper presents how to make steady-state error zero on designing strongly stable Generalized Predictive Control (GPC). Recently, the design method based on placing poles of the controller was proposed on strongly stable GPC, and it enabled to stabilize both the closed loop system and the controller. However, there is a problem that the steady-state error remains after the feedback loop is accidentally cut. In this paper, a conditional equation for controlled output to follow the reference value is derived. It determines the design parameters of strongly stable GPC using Computer Algebra System. Not only when feedback is cut, but also when output feedback is a fixed value, that is, output sensor is in a sticked error, it is derived that the output steady error is zero by using the obtained conditions. Controlling simulation is performed and it is confirmed that no steady-state error remains. As the result, this method makes steady-state error zero and enhances the safety on strongly stable GPC.

Some industrial equipments are required by the human operation such as bulldozers, excavators and cranes. Construction machines with better operability are necessary to improve productivity. Evaluations of those operational equipments are performed to obtain the better operability by subjective evaluations of the specific evaluator at the design stage. However, subjective evaluations require the skill of evaluator or a lot of sample data. In this paper, relationships between approximated parameters expressed as a low order system and subjective evaluations are verified by using the Weber-Fechner law. The evaluated motion is the boom up motion for an excavator. The simulator for the boom motion has been made because the real excavator has the limited condition. Therefore, the evaluation is performed numerically and experimentally. Approximated system parameters for the real excavator are calculated by the genetic algorithm.

PID controllers have been widely used for process systems. However, a good control result is not always obtained with fixed PID gains when a controlled object has nonlinearity. This paper proposes a design method for a nonlinear PID controller that utilizes a neural network to overcome the problem. In the proposed controller, PID gains are tuned online by a neural network and a controlled object is manipulated by the PID controller with the tuned PID gains. The neural network is learned by an offline learning algorithm based on the Extended Fictitious Reference Iterative Tuning (E-FRIT) and the backpropagation. E-FRIT is a method that tunes control parameters directly by using operating data and evaluates not only a controlled output but also the difference of manipulated variable. Simulation examples are provided to show the effectiveness of the proposed method. Moreover, the experimental result of a level control of a tank system is also given to demonstrate the performance of the proposed method.

In this paper, a reproduction of a swing-up and a giant swing motion of underactuated robots based on technique of the horizontal bar gymnast is discussed with focusing on an Equivalent Center of Mass(ECM) of underactuated robots and the gymnast. At first, the behaviors of the Equivalent Center of Mass of the gymnast(ECMG) are analyzed by using a motion capturing technique and an efficient motion of the ECMG for the swing-up and for the giant swing motion are identified from analysis results, respectively. Next, a partial linearization method, which can realize that the Equivalent Center of the Mass of the Acrobot(ECMA) replicates this efficient motion, is designed and reproduces the underactuated robots the swing-up and giant swing motion like the gymnast. Finally, an effectiveness of the proposed controller is shown by numerical simulations.

To deal with the local dissatisfaction with air-conditioner environment in office rooms, we propose a novel method for adjusting setting parameters of multiple indoor equipments together. In this paper, we demonstrate the framework of the method which sequentially tries various settings of multiple indoor equipments based on neighborhood search. Then, we show the experimental results in a room that has two indoor equipments. As a result, we can control the temperature in 9 out of 21 measuring points in the sense that we can cool the respective point most among all the 21 points.

In this paper, for an adaptive model predictive controller(AMPC) which has new internal model, tuning method of control parameters is discussed. Internal model in proposed AMPC has a control parameter to compensate a performance degradation caused by modeling errer between the internal model and an real model, and a parameter tuning method is proposed based on a steady-state characteristic of the control system. In order to verify an effectiveness of the proposed tuning method, numerical simulations with several condition are performed for Hammerstein system.

This paper proposes a method to design the weighting factor of control inputs in performance index of model predictive controllers. The weighting factor decides the closed-loop poles of predictive controllers and in this paper, the factor is determined to suppress overshoot peak value in transient response. For this purpose, a procedure to estimate the peak value of the overshoot is proposed. The procedure is derived based on the fact that the output transient response is obtained by a combination of powers of the closed-loop poles. Although the procedure does not give the exact overshoot peak values because the combination will change according the initial values and zeros of the plant, simulation examples give a good transient output response by choosing the weighting factor to minimize the estimated overshoot peak values.

PID controllers have been widely employed in real process systems. Since PID gains strongly affect the control performance, lots of schemes for tuning PID gains have been proposed. In processes industries, most of the PID controllers have fixed PID gains. However, it is impossible to obtain good control performance for time-variant systems by the fixed PID controllers. Therefore, it is important to tune PID gains in an on-line manner. In addition, especially in process systems, it is difficult to practice an open-loop experiment. Since, it is important to determine PID gains directly from the closed-loop operating data. In this paper, in order to overcome these problems, an implicit self-tuning PID control scheme is proposed. In the proposed method, a new PID gain tuning law based on an augmented output generated by closed-loop data is introduced, and recursive least squares method is applied to tune the parameters in an on-line manner. The effectiveness of the controller design scheme is evaluated by simulation and experimental examples.

In this paper, two-degree of freedom adaptive control system design scheme via an adaptive output feedback control with a simple adaptive feedforward input is proposed based on ASPR property of the system for MIMO discrete-time systems with the relative degree of zero. Since the ASPR discrete-time system must have the direct feedthrough term of the input, the control system is not able to design directly due to causality problem. To solve this causality problem, we will consider introducing a pre-compensator and introducing a parallel feedforward compensator for making the system ASPR again. We will also propose a simple adaptive feedforward control design scheme for improving the control performance. The effectiveness of the proposed method will be confirmed through some numerical simulations.

In April 2011, “information technology vision of education” which is a measure for the information of the education to 2020 year were summarized. For this reason, the Ministry of Internal Affairs and Communications, and the Ministry of Education, Culture, Sports, Science and Technology were the business in cooperation. In this report, improvement of teaching using ICT are mentioned. In junior high school technical education, a mechanism for processing the structure and material of tools and equipment, it is necessary to education that is based on scientific evidence. Therefore, the thrust force and torque by a gimlet are measured using an instrument intended to be quantitatively evaluated, including a process.

In recent years, the development of high resolution imagery has also covered digital terrestrial broadcasting, as well as mobile devices like smart phones and tablets. People have easier access to a huge selection of digital images on an increasingly large variety of devices. In order to provide attractive content, it is important to design and manage services based on users' Quality of Experience (QoE). The most effective method of evaluating QoE so far is subjective evaluation. In recent years, researchers in the field have been increasingly focusing on the biological information approach. In this paper we examine the relationship between the change in the amount of blood in frontal lobe of the brain and the preference of images by using Near-Infrared Spectroscopy (NIRS). We first measured the change in the amount of blood in the frontal lobe by NIRS while images were shown to subjects. Then performed a subjective assessment test through questionnaires, in order to get the level of preference for each image. We found that channel 6, 7, 8 and channel 10 belong to prefrontal cortex is highly relevant to human preferences of images and would allow us to indicate these preferences, and the amount of blood in prefrontal cortex changed more when subjects were observing image they liked more.

In the diagnosis of peripheral nerve disorder such as carpal tunnel syndrome, the doctor and the scientific occupational therapist mainly narrow down the position of the lesion by function evaluation. The diagnosis has been decided by the evaluation of the electrophysiological nerve conduction characteristics. In addition to this, in recent years it has become possible to observe the median nerve as a tomogram using an ultrasound diagnostic apparatus, so it is possible to check the thickness and shape of the median nerve without cutting the wrist joint part. However, in order to objectively observe a tissue having a three-dimensional structure such as a median nerve, it is required to obtain not only tomographic images of the median nerve but also three-dimensional shape information. Therefore, in order to visualize the median nerve as the three-dimensional structure, we develop a device that mechanically manipulates an ultrasound probe, while operating an ultrasound probe, it takes photographs the nerve tomographic images as well as it extracts the nerve from the images. In this paper, we describe the development of a device that operates an ultrasound probe and extract the median nerve in an ultrasound images.

A nail tip sensor which can continuously measure photoplethysmogram in daily life environment is proposed for remote monitoring device in home medical care. The nail tip sensor can be used without uncomfortable feeling, because finger nails have no sensory nerve nor sweat gland. The sensor also realizes rigid fixation over a long period of time like artificial nails. As a result, typical pulse wave is obtained in sleeping condition, but not well in working condition due to motion artifact. From twenty-four hours continuous photoplethysmogram data in daily life environment, the circadian variation of heart rate is successfully obtained.

FRIT (Fictitious Reference Iterative Tuning) is a model free designing method and a practically useful method. In the method, the optimal controller's parameters are searched to minimize a performance index. If the index is expressed in a form of a quadratic function of the parameters, then the optimal parameter is derived by a least square method straightforwardly. This paper gives a quadratic form of the index for the case of the general form of controllers.

Current, automation of the manufacture of machine parts is processing with the improvement of automation technology and robot technology. However, while demand for precision parts (mold, gear, and so on) is increasing, finishing work requires the expert skills. On the other hands, force control system by a force sensor has problem of frequency band and installation cost.

In this study, we aim to construct a low cost and surviving skilled skills system with combining sensorless force control system. This paper proposes construction method of broadband high-order reaction force observer for the reproduction of the expert skills. Moreover, we confirm the estimate accuracy of the proposed method through simulations.

There exists interference in a frequency modulation (FM) radio wave in a motor vehicle. The interference is called as a beat noise which is an electromagnetic radiation caused from electronic devices equipped on the motor vehicle. The beat noise becomes a perceptual sinusoidal noise and/or breaks a spectral structure of a broadcasting sound. It is required to suppress the beat noise by using a low cost device such as a DSP with a low-clock frequency. This paper proposes a noise suppressor employing two adaptive line enhancers as a promising method to be implementable on a DSP with low-clock frequency. We apply the proposed noise suppressor to signals obtained in the middle of the demodulation process. Simulation results show that the proposed method effectively suppresses the beat noise and extracts the broadcasting sound with high accuracy.

The decentralized robust control problem is considered for a class of large scale interconnected dynamical systems with both matched and mismatched perturbations and interconnections. It is assumed that the bounds of perturbations and the strength of interconnections are unknown. For such an uncertain large scale system, a class of decentralized robust sliding mode controllers is proposed by designing the sliding surface in consideration of the influence of mismatched uncertainties. The proposed decentralized sliding mode controller can guarantee that the solution of a large scale system converges asymptotically to zero. Finally, a numerical example is given to demonstrate the validity of the results.

This paper describes a person-invariant method of classifying subtle facial expressions. The method uses keypoints detected by using a face tracking tool called “Face Tracker”. It describes features such as coded movements of keypoints and uses them for classification. Its classification accuracy was evaluated using the facial images of unlearned people. The results showed the average F-measure was 0.93 for neutral (expressionless) facial images, 0.78 for subtle smile images, and 0.93 for exaggerated smile images.

We investigate an approach to apply Genetic Programming for the evolution of optimal escaping strategies of a team of caribou agents in the wolf-caribou predator prey problem (WCPPP) where the WCPPP is comprised of a team of caribou agents attempting to escape from a single yet superior (in terms of sensory abilities, raw speed, and maximum energy) wolf agent in a simulated two-dimensional infinite toroidal world. We empirically verify our hypothesis that the incorporation of empathy in caribou agents significantly improves both the evolution efficiency of the escaping behavior and the effectiveness of such a behavior. This finding may be viewed as a verification of the survival value of empathy and the resulting compassionate behavior of the escaping caribou agents. Moreover, considering the fact that a single caribou cannot escape from the superior wolf, the ability of a team of empathic caribou agents to escape may also be viewed as an illustration of the emergent nature of a successful escaping behavior - in that the team-level properties are more than the mere sum of the properties of the individual entities. Within this context, we also present empirical results that verify the complex (nonlinear) nature of the relationship between the size of team of caribou agents and the efficiency of their escaping behavior.

Differential equations are considered to be a fundamental tool in many areas of engineering. The goal of our project is to develop an education system that teaches differential equations, in which exercise and assessment problems are automatically generated and delivered to the students. A student's comprehension is identified using his or her answer history record of past answers, and problems at the appropriate level of difficulty are selected and delivered to the student. The proposed education system is implemented by MATLAB.

Item characteristic parameters of each class of problem are equated using test result from 183 students in advance. The implemented proposed system has been examined by 13 examinees in 4 weeks. The experimental results show that the proposed system can deliver suitable problem for each student, and all students can improve their performance on solving differential equation through the experiment.

Measurement of bio-information and prevention of fall from bed are needed for elderly person during sleep. In this paper, the authors propose a novel system for these purposes using game machine. Respiration signal can be detected by fluctuation of load value from body motion. Possibility of fall from bed can be easily discriminated from movement and distribution of load value. From experimental results, proposed system was worked well for the purpose of watching elderly person.

This paper describes collection of responsive utterances of showing attentive hearing for narrative speech. In this research, a method of annotating an existing narrative corpus with responsive utterances was adopted. Actually, 16,213 responsive utterances were collected for a narrative corpus consisting of 13,381 utterances. It was confirmed that the method can collect a variety of natural responsive utterances.

The purpose of this research is to generate a character string from the DCT sign of the image and investigate whether it can be used as a password. The sign of the DCT coefficient of the low frequency component of the image is extracted by a zigzag scan, it is segmented every 6 bits and converted into a corresponding character to generate a password character string. Experimental results show that unique character strings are generated for each image. In conclusion, it was confirmed that the character string generated from the DCT sign of the low frequency component of the image can be used as a password.