In road maintenance, it is necessary to construct an environment that manages three-dimensional data and maintenance information. The primary objective of this research project was to support road maintenance work using three-dimensional data by combining terrestrial laser scanning (TLS) and unmanned aerial vehicles (UAV) with photogrammetry. TLS and photogrammetry technologies have been used to survey road structures. Three-dimensional data for the pavement, landform, and bridges have been merged using point cloud data measured and generated by TLS and UAV photogrammetry. Point cloud data have been combined using coordinate points. Coordinates of the three-dimensional data have been created on the basis of the structure from motion range-imaging technique of photogrammetry using video camera data. In addition, the data have been used for road maintenance, taking into consideration data size and accuracy. Road maintenance information can be referenced at any three-dimensional point in the three-dimensional data. This paper has evaluated the accuracy of the three-dimensional data of laser scanning and photogrammetry for road maintenance.

Understanding the actual condition of tourist’ travel behavior is important for mitigating tourist congestion and marketing purposes. Typical travel survey methods have practical difficulties (e.g. necessity of staying at a certain period of investigation) for fully uncovering tourists’ travel behavior. In this study, we extracted unique identifiers of Wi-Fi packets which presumably be from tourists in Okinawa Main Island by focusing on the entrance information about the entry into the islands (e.g. cruise ship terminal users, domestic air terminal users, international air terminal users and, low-cost career terminal users). After appropriately conducting the data cleaning, we analyzed tourists’ travel behavior mainly in terms of travel flow patterns across different entry points into the island. Through the data analysis, it became clear that tourist’ travel behavior is significantly different by means of transportation into Okinawa Main Island. Furthermore, we have confirmed the possibility of fusing Wi-Fi based tourist excursion data and probe-based vehicle trajectory data that can reveal the whole travel patterns of each tourist, though its matching rate is not enough large at present.

In the recent super aging society in Japan, it has been observed that the railway traffic demand for urban area tends to decrease. Basically, the urban traffic demand is regarded as derivative demand for urban activities corresponding to the increased economic activities. In particular, the passengers in railway station should be strongly connected to the activities in surrounding area. Therefore, the integration between urban development and railway convenience in the station has been proposed as a recent urban planning direction. The estimation model of railway passengers in the railway stations is created by fuzzy reasoning with the knowledge derived from the previous analysis for railway service level and activities. Finally, the proposed fuzzy reasoning model may provide the indices of area activities relating with the railway service of the stations.

As the car oriented city is extended in local areas, reduction of carbon dioxide emission should be required. The development of environmental smart city is discussed in the study to promote the low carbon society as global environment issues. The multi-agent model is developed to describe the decision process concerning with modal choice as well as vehicle type choice of individual trip makers. The proposed model is applied to evaluate the future public transport policies to reduce the carbon dioxide emission such as service of ART (Advanced Rapid Transit) as well as the environmental charging for high emission vehicles. Finally, the impacts of public transport policies are properly evaluated in contrast to the no policy introduction for low carbon transport systems.

Trapezoidal fuzzy inference is a general form of triangular fuzzy reasoning, that has proven to be effective at solving various types of inference problems. When used as a clustering method, fuzzy inference allows for adjusting cluster boundaries with each new datapoint. In trapezoidal fuzzy inference, both the membership function for the antecedent part and the singleton real value of the consequent part need to be learned. However, in typical applications of fuzzy inference, there is not much discussion of the appropriate way to construct fuzzy rules as a design problem. For example, when coding program, it is not clear how to determine the learning coefficients of the membership function and singleton, how to set their initial values, or how to schedule the learning sequence of the various parts of fuzzy rules. In this paper, we frame the problem of parameter adjustment for fuzzy rules not as a tuning problem but rather, as a design problem. We focus in particular on how to learn coefficients both for the membership functions of the antecedent part and the singletons of the consequent part, how to set initial values, and how to determine the learning sequence of consequent and antecedent parts. We start with a definition of trapezoidal fuzzy inference and introduce the steepest descent method for adjusting fuzzy rules. Next, we propose six new methods for initializing parameters, and five ways to schedule the learning sequence. We discuss the accuracy of the proposed design methods in light of quantitative evaluations performed on sample datasets.

In this study, an experiment was conducted in Arakawa, Tokyo wherein three groups of people, namely Japanese residents, Japanese visitors, and non-Japanese visitors, wore an eye tracker and walked along a conventional Japanese shopping mall. Visual fixation points were measured with respect to a two -dimensional (2D) orthogonal relative coordinate system of the eye tracker screen. A method was developed for mapping these 2D data into a three-dimensional (3D) polar absolute coordinate system, and the fixation point distributions in 3D space were thereby obtained. The fixation time distributions and the associated visual objects were also determined. Gaze behaviors of the three groups were subsequently analyzed and compared. Furthermore, subconscious visual attention while walking in the town was quantitatively characterized according to the social backgrounds of the people.

Based on the fuzzy c-means clustering method maximized with the Tsallis entropy, we have achieved its extension that assigns the q parameter of Tsallis entropy to each cluster as q_i. In this method, however, there remains three problems. That is, (1) determination q_i reflecting the data distribution, (2) occurrence of abnormal bias of q_i, and (3) calculation termination condition of q_i. In this article, we propose a new calculation method and a termination condition of q_i, and show that all problems can be solved.