In this paper, we implement a new map generation tool for RoboCup Rescue Simulation (RRS). RRS is a multi-agent test bed and simulates large-scale natural urban disasters. RRS needs many maps of various regions for disaster risk reduction. However, there are only a few maps due to the difficulty of map generation. The reason is that the current map generation tools produce maps which need a large number of manual modifications in order to be used in RRS. Therefore, we design and implement the new tool that generates RRS maps from OpenStreetMap (OSM). We confirm that the tool can generate maps that needs a few manual modifications.

This paper analyzes factors which affect the J-REIT investment unit price. Multivariate analyses such as multiple regression analysis, principal component analysis and cluster analysis were conducted. Reviewing past researches, there are few papers made on this. There is a big volatility in investment unit prices of J-REIT, therefore prediction about them becomes difficult.Systematic analysis methods were adopted to the investment unit prices of all J-REIT. The results were interesting and instructive and the examined method may be utilized for making investment decisions of J-REIT in a simple manner.

Among many dimensional and dimensionless amplitude parameters, Kurtosis (4-th normalized moment of probability density function) is generally regarded as a sensitive good parameter for machine diagnosis. However, higher order moment may be supposed to be much more sensitive. Bicoherence is an absolute deterioration factor whose range is 1 to 0. The theoretical value of n-th moment divided by n-th moment calculated by measured data would behave in the same way. We propose a simplified calculation method for an absolute index of n-th moment and name this as simplified absolute index of n-th moment. The case in which the rolling elements number is nine is examined. Some favorable results are obtained.

Recently in Japan, Low Cost Carriers (LCC) have entered into the market and the fierce competition is unfolded. It is our great concern that how become the selection of airline. In order to clarify the preference shift of airline, we have made the questionnaire investigation. In the general case of purchasing, it is often observed that consumers select/purchase upper class brand when they purchase next time. Suppose that former purchasing data and current purchasing data are gathered. Also suppose that upper brand is located upper in the variable array. Then transition matrix becomes upper triangle matrix under the supposition that former purchasing variables are set input and current purchasing variables are set output. If the top brand were selected from lower brand in jumping way, corresponding part in upper triangle matrix would be 0. Then in the airline selection case, how does it become. The questionnaire investigation to airline ticket purchasing case is executed and above structure is confirmed. If transition matrix is identified, S-step forecasting can be executed. Generalized forecasting matrix components’ equations are introduced. Thus, this proposed approach enables to make effective marketing plan.

Currently, there is a problem that leakage is due to damage water pipes. In this research, we propose position estimation method using sound wave and OFDM in order to detect the location of water leakage accurately. And we aim to propose a method that enables position estimation more simply and for longer distances than before.

For autonomous control of the robot, it is necessary to decide motions and recognize the scene. In this paper, we report the image processing algorithm for robot control using the stereo vision system. The three-dimensional position can be estimated by the stereo camera using to USB cameras. This system measures the distance, detects the object, and calculating the center of gravity. We will try to control the robot based on the coordinates obtained by image processing. This paper shows some results using real image processing system that we developed.

Our previous automatic sailing machine used only positional information and it controlled its sail and rudder to maximize the boat speed. So, our previous sailing boat could not head upwind. In this study, we propose a new set of control laws using goal direction component of the boat velocity vector. We implemented the control laws in a small sailing boat equipped with Raspberry Pi and GPS module, and verified them in the real world experiments. We set a triangle course with two target positions and indicated to return to the start point. As a result of the experiments, the use of the goal direction component worked effectively, and the sailing boat came back to the start point successfully in the case where sufficient wind blows.

In RoboCup Soccer Simulation 2D League, almost all teams use an evaluation function in action selection. Team performance highly depends on the evaluation function. The aim of this paper is to propose a method that improves the performance of a team by imitating a target team. For this purpose, a neural network is employed to model the target team’s decision making. The neural network is trained by using positive and negative episodes of a target team’s action sequences that are extracted from game logs. Though our computational experiments, it is shown that the performance (e.g., win rate, scored goal, and so on) is improved by mimicking the winner of RoboCup 2017 soccer simulation 2D league.

Because of the spread of microblogs and SNS (Social Networking Service), life logs submitted by their users are becoming various. With analyzing these life logs, attributes of those users such as gender, age, occupation, and living region can be acquired. So it is being expected that these attributes can be utilized in corporate sales strategies, etc. In particular, it can be said that the living region is very useful in area marketing. In existing researches for estimating living regions, methods using the region’s special words selected from topic models or analyzing differences in preference for each user's living region have been proposed. However, in the first method, the appearance frequency of topics concerned with a season such as snow is different from another season. Therefore, here comes a problem that the appearance frequency of region’s special words extracted by this method in different seasons are different. And then the region which special words stands for is different from living region. On the other hand, the second method also has a problem that it is difficult to estimate living region because of the low correlation between it and preference of a user. Therefore, in this research, we consider a method focused on synonyms which are used in different regions have the same meaning.

In recent years, it is expected to improve the efficiency of criminal investigation and flow line analysis in facilities using surveillance cameras. Furthermore, it will be the safety measures that workers are identified when they are in dangerous areas of construction sites. For these reasons, automatic person identification technology is desired. As for the existing researches using deep learning, there are many approaches like face authentication, gait identification and person identification. However, in these methods, there have some problems to identify the person. Therefore, in this research, we propose a method using deep learning to identify person wearing a helmet.

As smartphones and IoT become widespread, opportunities to utilize services provided by companies and individuals are increasing. In using these services, terms of use or standard forms are displayed, with which users often agree without confirming their contents. This may cause troubles in some cases. For this problem, a method for extracting important sentences based on common important concepts registered in advance is proposed; however, this cannot deal with unknown provisions. In this study, we propose a method for extracting the unknown provisions that are different from general matters by dividing the terms of use into individual items and calculating the similarity and feature quantity.

Long-term care insurance system is operated for overall society to take elderly care. The self-perceived health for welfare is estimated to realize sustainable service of long-term care in the study. The survey database of the attitude of citizens for elderly welfare is utilized to find the factors of the self-perceived health for welfare. The deep learning technique is applied for estimation of self-perceived health for welfare with the survey database. Finally, it can be summarized future effective health care policies for elderly people referring to the self-perceived health for welfare produced by mentality of citizens.

A different approach to the index selection problem is proposed, with partial indexes as the focal point.　FP-Growth, as well as a dynamic programming algorithm for optimization, is adopted from a previous research project, which also serves as a basis for the proposed solution.　The aim of this research is to minimize wastage of memory and, ultimately, allow for more indexes to expedite the processing of incoming queries. Instead of suggesting potential indexes based solely on which attributes appear most frequently in the workload of queries, the density of filtered queries are also accounted for, i.e. columns of corresponding attributes are only indexed in smaller portions as long as these portions satisfies the majority of all queries for said attributes. MAFIA, a subspace clustering technique based on CLIQUE, has been implemented in order to discover these dense areas. To evaluate the effectiveness of the solution, a comparison to the former solution was made through numerous experiments, and the performance brought about by the generated indexes was benchmarked and compared between different test-cases.

Recently, many researchers have proposed recommendation systems which based on user's interesting.On the other hand, it needs to purchase items, such as clothes for ceremonial occasions, separately from user’s interesting. in the case, it is necessary to dress along with situations.To purchase these clothes, we need to consider good manners. Though, it is hard to choice clothes that satisfied good manners. In this article, we propose recommendation method that recommended items is satisfied with good manners.

In the field of histopathology, characteristics analysis for pathology images was made a study for devising precision medicine and patient-specific diagnosis. Our research group is currently addressing the problem of recognizing and extracting quantitative features from Glioma images by defining, counting, and measuring morphological characteristics that can to serve as fundamental data elements capable of distinguishing key disease types and subtypes. Actually, subtypes of Glioma is not specified so far. Moreover, disease stages of Glioma are determined by a combination of genes. Therefore, even if the given tissue images belong to the same grade, the patterns of gene expression are different among patients. The final goal of this study, we specify subtypes of Glioma from histopathological images for precision medicine by genes expressions level.

By combining fuzzy c-means clustering, the Tsallis entropy maximization method, and deterministic annealing, we have developed the single-q clustering algorithm. Then, the algorithm has been extended to the multi-q clustering algorithm. In this method, the qs are assigned individually to each cluster. Each q value is determined so that the membership function fits the corresponding cluster distribution. This is done by introducing a new parameter. However, the accuracy of clustering depends on the setting of a value of the parameter. Accordingly, in this study we propose a new clustering method that does not require an additional parameter to determine the q values. Experiments are performed on randomly generated numerical data and "SDSS quasar spectra" dataset, and it is confirmed that the proposed method works correctly and improves the accuracy of clustering and is superior to the conventional multi-q method.

The authors have measured electroencephalograms (EEGs) from subjects observing images of part of fruit and recalling them silently. The equivalent current dipole source localization (ECDL) method has been applied to those event related potentials (ERPs): averaged EEGs. ECDs were localized to the primary visual area V1 around 100msec, to the ventral pathway (TE) around 270msec, to the parahippocampal gyrus (paraHip) around 380msec. Then ECDs were localized to the Broca’s area around 450msec, to the fusiform gyrus (FuG) around 600msec, and again to the Broca’s area around 760msec. Process of search and preservation in the memory has been done from the result of some ECDs to the paraHip. With or without activities on the angular gyrus (AnG) are important to discriminate of the presented images.

This study assumes a group of the individuals and aims to obtain each individual inner evaluation from the intuitive judgments by a group of experts. The intuitive judgment is given as a comparison matrix of the multiple criteria and the inner evaluation is obtained as a fuzzy weight vector. Two kinds of the groups are considered: the judgments on the target individual and those by each expert. The inner evaluation of the target individual is simply obtained as the core of the former group. Furthermore, we compare the judgment on the target by an expert with those on all the other individuals by the expert to reflect his/her judgment characteristic. The relation among the judgments in each group is constrained by the inclusion relation and it is relaxed for a fuzzy weight vector.

We propose a robotic model to search a robot's standing position considering a change of human's personal spaces in a human-robot community. Recently, there has been an increasing effort to develop personal robots in human communities. In the environment, they are required not only to avoid a collision with a person but also to keep a suitable distance between a person in its community. However, their personal space in a community depends on their closeness. When a robot ignoring the change of their personal space is too close to or too far from the people, they do not feel an affinity with the robot. Therefore, a robot needs to search its standing position consider the change of their personal space. Thus, we propose the robotic model and examined whether a robot using the model can avoid to encroaching people's personal space in simulations. As a result, the rate of keeping a suitable distance accounts for about 77%.

We propose a system that simulatively experiences the act of eating without actually eating food. The experience of eating is reproduced by giving three stimuli at the same time that haptic stimulus representing food texture, visual stimulus representing food appearance and taste stimulus by using liquid flavored similar to food taste. At this time, only a single taste stimulus is used, and only a combination of a haptic stimulus and a visual stimulus is changed to express a plurality of foods having a similar taste.

In this paper, we propose an optimization model of emotional expression with a focus on action.Many communication robots bring healing to people through emotional communication by words and actions. However, compared to humanoid partner robots, pet robots, which are one of the representative communication robots, have few ways to express various emotions and it is possible to express emotions only by actions. Therefore, we propose an optimization model of emotional expressions by using an interactive genetic algorithm. This model enables robots to express their emotions considering user sensitivity, so it is possible for robots to communicate more emotionally. Moreover, we inspect the optimization capability using interactive genetic algorithm by simulations. As a result, the proposed system is able to optimize emotional expressions appropriately.

On the target instruction by pointing gesture, the indicated positions are different between the subjective position by user and the objective position by measurement system. Pointing gesture is expressed under unconscious using body position sense.The measurement system should construct the relationships between objective position and subjective gesture according to the position sense.In this paper, we focused on position sense for pointing gesture expression to understand subjective gestures.

This paper proposes a human-robot interaction system that not only acquires robot sensory information but also human personal information such as residential address, age, and occupation. We also extend the human-robot interaction system that can classify human personas and various interaction styles according to classified user information. The proposed method has been validated by series of experiment and real robot implementation.