In this paper, we propose a method for investigating and analyzing activity information on each tourist on the basis of information technology in order to realize effective tourism promotion which is suitable for current tourism environment. GPS is employed as a device to collect activity information on each tourist, and then activity is analyzed on the basis of the position data collected by GPS. The detailed information such as the facility name in which tourist stays is required to realize effective tourism promotion. We also discuss a method for acquiring such information through the internet.

View full abstract

We are implementing an indoor evacuation guidance system on the assumption that indoor positioning and congestion sensing techniques are established. Because effective evacuation strategy is greatly transformed according to the congestion of evacuation route, approach to present an automatically acquired strategy suitable for the situation is desirable for such system. In this paper, effectiveness of the acquired strategy compared with a strategy of escaping from the nearest emergency exit and stability of the system performance when the ratio of users changes are verified on an evacuation simulator.

View full abstract

Art-Tourism is a new tourism related to art contents these are field sculpture, music, photograph, and so on. Especially, we think a field sculpture is one of the most important cultural properties, and we might be feeling a sense of closeness to it in our life. In this paper, we explain a Hokkaido Sculpture Web that is a community web site to support art-tourism based on more two thousands photographs taken by volunteer photographers, and it is developed by open source program to be used by other people in other area in future. In addition, we propose an application of Hokkaido Sculpture Web in emergency.

View full abstract

As the number of individual tourists increases, Web browsing is becoming one of the methods to gather tourism information such as restaurants and accommodations. We can distinguish two types of tourism information. First, there is objective information such as address and telephone number. Then, we can talk about subjective information, such as evaluation given by an individual. While, browsing tourism information on major sightseeing sites, these sites usually have a lot of advertisements and there are only a few candid opinions of the users. Further, the individuals post various articles on the Weblog, each one in their own way by subjective viewpoint. However, it is difficult to search between a lot of subjective data for necessary and valuable tourism information. The purpose of our work was to gather objective information and subjective evaluation about tourism contents, and to develop a user adaptive tourism information website using a recommender system. Moreover, an exclusive weblog was developed as an interface for tourism information gathering. In this paper, we present a prototype of the system applying restaurant information.

View full abstract

The objective of this paper is to investigate a search method in a linked space such as Web pages. Most of search methods in a linked space are a brute force one essentially, because we cannot apply an effective heuristic search method in the reason that most of heuristic algorithms are designed to works in Euclid space (or a metric space). For applying effective algorithms, in this paper, the following method is investigated: a linked space is arranged into Euclid space by multi-dimensional scaling (MDS), and a heuristic algorithm such as particle swarm optimization (PSO) is applied to in the Euclid space.

View full abstract

Some the algorithms for dividing the large-scale complex network into the community only by the topology are proposed. These existing algorithms are applied to static network, and adaptability to a network that changes depending on time and a different type of networks is not considered. In my study, I propose a stable community detecting algorithm for evolving complex networks. First, I bring in a stability to evaluate some existing algorithms, then, I developed a method to detect robust communities. For as numerical experiment, I applied a proposal method and some existing methods to three evolving network models and Weblog network based on real data on World Wide Web, availability of a proposal method is shown.

View full abstract

This study proposes a new system, analyzing a relationship among Internet Auction users. As a relationship among the users organizes a complex network, the system treats users and dealing relations between users as a set of vertices and a set of edges. Also, the system employs the network-analyzing tool, "Pajek". The proposed system analyzed the real dealing records extracted from "Yahoo! Auctions". As a result, the clustering coefficient explains that a collector user's degree has a strong relation with making a group formation and the closeness centrality explains that some pattern reveals in choosing partners for dealing.

View full abstract

In recent years, the use of the Internet is indispensable to a life and various services are offered on the website. Therefore, the search engine which looking for the target website have been attracted attention. Many methods are developed in order to search efficiently. Now, retrieval by keyword is the mainstream of web search. However, use of retrieval by keyword is difficult for a beginner. Thus, a beginner cannot obtain the result of the purpose. In order to search efficiently, the knowledge about the usage and a keyword is required. Since those who get interested in a car are decreasing in number, Automobile search is raised as an example of search to which use becomes difficult. Then, the search method of assisting a beginner is proposed. We treated only the car selection process using emotional factors.

View full abstract

In this paper, we report the construction method of multi-agent system by using genetic algorithms. We construct the multi-agent system applying for the problem of capturing moving objects by genetic algorithms. Then we consider the following two themes: 1) the determination of rewards for agents, 2) the construction of simple multi-agent system using two kinds of agents. From the result of simulations, we discuss the conditions to construct multi-agent system and the method to construct a simpler multi-agent system.

View full abstract

Harary's structural balance theory based on the idea of Heider explains social processes and is used to account for social actor's attitudes toward another. We propose a new social value emergence model in the form of an agent-based simulation model. In this model, the structural balance theory is used to explain the feelings, attitudes and beliefs. Each agent is in effort to reach balanced states and communicates with each other. We analyze a social value emergence by using the proposed model.

View full abstract

Reinforcement learning is a promising approach to realize intelligent agent such as autonomous mobile robots. This paper presents hierarchical modular reinforcement learning method that Q-Learning is combined with modular structured Profit-Sharing reinforcement learning algorithm hierarchically based on task decomposition in multi-agent environment. By using the learning method, this research aims at learning the behavior of the agent in the pursuit problem, i.e. distributed AI problem in the multi-agent environment. Through numerical experiments, we found that the proposed method has good convergence property of learning compared with the conventional algorithms.

View full abstract

Boid is one of the major swarm behavior models, which expresses complex behavior by using three simple rules. The model needs the accurate data of all neighbors in order to control the swarm well. However, in general, it is difficult to obtain their data under practical environments. In this paper, we try to generate swarm behavior from the restricted local information and apply the proposed method to the obstacle avoidance as keeping a swarm form. The validity of the proposed method is confirmed by computer simulations.

View full abstract

This research deals with the exponential fuzzy numbers in Analytical Hierarchy Process (AHP), employing the fuzzy concept. The objective was determined a method of deriving the weights of criteria and their alternatives. Next, the influence of fuzzy is examined by using the membership function. Specifically we use the exponential fuzzy numbers for the pairwise comparison scale. Same experiments with AHP were conducted using the pairwise comparison scale and the new pairwise comparison scale of the method was examined.

View full abstract

Analytic Hierarchy Process (AHP) is proposed to give priority weights with respect to many alternatives with many criteria. In the conventional AHP, local weights of alternatives under each criterion are obtained from the pairwise comparisons given by a decision maker. By extending the weights from crisp to interval to reflect inconsistency of the given comparisons, Interval AHP has been proposed. The global weight of an alternative is assumed as the weighted sum of local weights. The weights represent importance of criteria in evaluation so that they can be determined for each alternative from various viewpoints. We propose the interval global weights whose bounds are from the optimistic and pessimistic evaluations. Although their widths represent the possibilities of global weights, too large width from too optimistic and/or pessimistic evaluations can not be a meaningful for a final decision. Then, the obtained interval global weight should be normalized relatively so as to reduce redundancy.

View full abstract

In the design of computer software system with various functions, the control of implementation is not easy. When we consider the control flow as a transition chart, we should make a graph only from the elements and their relations, In this paper, the best structure of graph is acquired by GNP. First, we assume an appropriate fitness function for a given problem. However, even the necessary element will be disappeared in the end of evolutionary process. This paper explains that the sparse structure among their relations are extracted by ISM. Then, the necessary node will be embedded in the evolutionary process.

View full abstract

In the previous paper, we proposed a method with dynamic construction facility of the state space, where we initially have no state and gradually add a new state of fuzzy set with removing unnecessary actions. We adjusted Q values for actions but not fuzzy sets for states. In this paper, therefore, we propose a method to adjust fuzzy sets, the central value and width of its membership functions, by TD (Temporal Difference) error. Then, we apply this method to the pursuit problem in real number environment.

View full abstract

This article presents an approach to find multilayer neural networks with better performance by using the Design of Experiments method. Multilayer neural networks can approximate nonlinear functions by employing back-propagation learning algorithm. However, it is difficult to design the neural network with high performance because of an unknown relationship between the learning capability of the neural network and the design parameters such as its structure and conditions at learning. Our approach is to introduce the Design of Experiments method, which is based on an orthogonal array and analysis of variance, into design of multilayer neural networks in order to efficiently design them with a quantitative index.

View full abstract

The development in information technology, especially computational performance, enables us to have the intelligent systems such as humanoid robot and entertainment robot in real world. These robots have to adapt their behavior depending on the dynamically changing environment. Some kinds of learning capabilities are implemented in such systems. Reinforcement learning (RL) is one of the most actively investigated algorithms, and Q-learning is one of the most popular ones. Q-learning is generally needs quite long time for learning behaviors, so many kinds of improvements for effective learning have been proposed. Considering that these robots live with us in our daily life, it is also important to act without making us keep waiting, since we are not so patient. Reducing computational cost while executing some behaviors is important to achieve these properties. Recent studies in brain science suggest that memory based predictions may be the keys to realize saving the activities in the brain. In this paper, we try to discuss its architecture to realize these characteristics, based on reinforcement learning and associative memories.

View full abstract

When we solve a problem, we firstly have no knowledge and gradually acquire some piece of knowledge by observing new data, and at last arrive at complete knowledge for solving the problem. To implement such kind of learning mechanism, we proposed a learning method of switching reasoning methods and rule generation methods. In the previous paper, we proposed a method that aquires meta-rules for switching with a reinforcement learning. The aquired meta-rules may be independent from data sets and can be commonly used for learning many data sets. In this paper, we apply the switching meta-rules acquired from the iris plants data set to learning the wine recognition data set. We got a result that has the smaller number of rules than the conventional switching method with the same correct ratio.

View full abstract

Recently, biometrics technologies have been widely studied in order to remove the drawbacks of conventional information security technologies. Current biometrics technologies are almost based on static authentication, for example, fingerprint, iris, blood vessel and so on. In order to develop this area more, it is necessary to include dynamic authentication technique by using on-line signature, body movement and gesture. This paper shows a novel on-line handwriting authentication method by a new dissimilarity measurement and writing trajectories in pen-up as well as pen-down. As the result of simple experiment, proposed measure works well for distinguishing a registered handwriting from an unregistered handwriting.

View full abstract

SOM(Self-Organizing Map) by T.Kohonen is one of the methods for classifying data of a large number of attributes automatically, where output units are restrictedly placed on the grid points. We propose a real-number SOM(RSOM), where output units are randomly placed on the real-number coordinates and visualized as a Voronoi diagram. Voronoi diagram partitions the plane into many areas that contains the points that are closest to the corresponding output unit. RSOM can visualize more expressively and place output units on curved surfaces easily. To make more understandable diagram, we also propose two methods to move output units to the boundary from the best matching units and a method to remove the units that do not affect the boundary.

View full abstract

Ritter developed the morphological associative memory (MAM) which is able to achieve the perfect recall. The MAM is a powerful associative memory compared to an ordinary associative memory. However, the MAM needs the kernel image which is difficult to design. On the contrary, the associative memory with small-world network is known as effective network which is able to improve recall rate and reduce memory matrices compared to regular network. The small-world network is produced without difficulty so that the network is useful. We try to improve of the perfect recall rate and reduce the memory matrices using the small-world network.

View full abstract

This paper suggests the hybrid Neuro-Fuzzy Multinomial Logit model for use in transportation studies. The model is applied for estimating travelers behavior in the context of the transport modal choice problem, where the modes of bus, subway and automobile are investigated. The model is evaluated by comparing its results with the results of a Multinomial Logit model. Moreover, the probabilities of selecting a mode obtained by applying the two models are compared with the actual transport modal choices, showing the efficiency of the suggested model in practical transportation problems.

View full abstract

In this paper, we consider a traffic flow model where the information about the actual travel time for each alternative route is not available when each driver performs route selection. For such a traffic flow model, we examine the effectiveness of two routing methods for minimizing the average travel time over all vehicles running in the model. One method is to minimize the average travel time globally. We employ a genetic algorithm to search for a near-optimal route choice of each driver. The other method is to minimize the average travel time locally by each driver. We employ a neural network to estimate the travel time for each route. Through computational experiments, we compare the two methods with each other and demonstrate the characteristic features of each method.

View full abstract

Recently, various camera systems have been developed and installed for security. It is generally accepted that use of multiple cameras each of which are synchronized and cooperated is effective. In such camera systems, the number of camera installed in the field is limited due to space and cost. Thus, to obtain useful information using fewer cameras is desired. In this research, we implemented an algorithm to find an efficient camera allocation into a microscopic road traffic simulator which is named MITRAM, and determined the camera allocation.

View full abstract

This paper presents dynamic output feedback control designs for discrete Takagi-Sugeno fuzzy systems. A dynamic output feedback controller is constructed based on the concept of dynamic parallel distributed compensation (DPDC). Design conditions for optimal dynamic feedback control are obtained in terms of linear matrix inequalities (LMIs). In application to a vehicle with triple trailers setup, we utilize the optimal design conditions to avoid the jack-knife phenomenon. Our results demonstrate that the optimal dynamic output feedback design effectively achieves the backing up control of the vehicle with triple trailers while avoiding the jackknife phenomenon. More importantly, we demonstrate that the designed optimal control can achieve the backing up control without two potentiometers that were employed to measure the relative angles (of a vehicle with triple trailers) in our previous experiments.

View full abstract

What do the drivers who can understand the routes well pay attention to when they are driving cars? If we know this relation, we are able to develop a fine car navigation system that provides effective information for drivers. This paper presents the experimental results of the relation between drivers' situation awareness and their spatial travel understanding. In the experiment, we used SAGAT to measure the drivers' situation awareness and we conducted route recall test to measure the drivers' spatial travel understanding. The qualitative analysis of these measurements reveals effective expression of the car navigation system.

View full abstract

A family of automatic container loading problems is studied and lgorithms are proposed. The algorithms are constructed with metaheuristics and include at and/or vertical loading schemes, loading effciency, stability of loaded objects, and computational requirement. Handling groups of objects in a metaheuristic scheme is moreover considered.

View full abstract

We have proposed the active learning method (ALM) as asoft computing methodology.The ALM is algorithmically modeled onthe information-handling processes of the human mind. In this paper,using the ALM, we attempted the acquisition of control knowledge of parking a motor vehicle.In this learning application, the control based on linear feedback is insufficient since vehicles possesses non-Horonomic characteristics. The simulation results demonstrated that the proposed method can acquire simple control knowledge of parking a motor vehicle by accumulation of experience.

View full abstract

The traffic safety planning has received much attention in terms of reduction of danger in driving. The causality of the plural factors has to be clarified in the estimation of traffic accidents at the intersection. Simultaneously, the high estimation accuracy in the modeling is necessary. The model with fuzzy reasoning was introduced to the estimation of the number of traffic accident in the intersection. It concluded that time and spatial distribution of traffic accidents are determined quite accumulate. The parameter of membership function is optimized by learning with the neural-network and genetic algorithm.

View full abstract

In this paper it is presented a mixed control approach for stabilising a Furuta pendulum in upright position. The full Monty controller consists partly of a state feedback controller and partly of a Lyapunov controller. A fuzzy controller takes the decision to vary between of the structures.

View full abstract

This paper is concerned with output feedback stabilization of discrete-time Takagi-Sugeno fuzzy systems with immeasurable premise variables. Takagi-Sugeno fuzzy systems can usually approximate nonlinear systems and even describe a certain class of nonlinear system exactly. When we consider Takagi-Sugeno fuzzy systems, the selection of premise variables plays an important role. If the premise variable is the state of the system, then a fuzzy system describes a wide class of nonlinear systems. However, the state is not measurable in the output feedback control problem. In this case, a control design of the underlying nonlinear system based on its fuzzy system description is difficult. In this paper, we consider the output feedback stabilization problem. In this problem, the premise variable may not be unknown. Thus we formulate the output feedback stabilization problem of fuzzy systems as robust stabilization of uncertain systems. Numerical examples illustrate our theory.

View full abstract

Fuzzy model-based control mainly deals with dynamical systems which affinely depend on control inputs. In this paper, we propose the switching fuzzy model construction and controller design for dynamical systems which is permitted to have nonlinearity not only in the states, but also in the inputs. By utilizing the dynamic state feedback control approach, we show that we can deal with membership functions which depend on control inputs and that the number of the fuzzy model rules which fire simultaneously can be decreased in comparison with the ordinary switching fuzzy model. We design the switching fuzzy dynamic state feedback controller and derive the controller design conditions based on the switching Lyapunov function in terms of LMIs. A design example illustrates the utility of this approach.

View full abstract

A method of designing Takagi-Sugeno fuzzy control systems based on the parameterization of quadratically stabilizing controllers is presented. Conception of Youla parameterization of LTI systems are extended to T-S fuzzy system with respect to quadratic stability. The parameterization of the close-loop systems, which are affine with arbitrary stable Q-parameter, is then described. This description enables the application of the Q-parameter approach to various T-S fuzzy control-systems. Above all, a design scheme of Q to obtain L2-gain performance is clarified.

View full abstract

Fuzzy system is applied to the anti-rolling device of a high-speed catamaran boat in order to improve the passenger comfort. A PID control was initially employed in the system, however no significant improvement was observed for passengers although instrumental data showed good results. The present fuzzy system has shown not only good instrumental data but also provided good comfort ride for passengers. The experiment results demonstrate the effectiveness of the fuzzy system for the anti-rolling system of a high-speed catamaran boat, superseding the conventional PID Control. We also propose the prediction model of the anti-rolling system using neural networks.

View full abstract

This paper presents a concept to design a Lyapunov function candidate and the stabilizing controllers of nonlinear discrete-time systems. A stabilization algorithm for linear step-varying discrete-time system is proposed via piecewise Lyapunov functions based on a tracking trajectory control. This tracking control can be applied to arbitrary trajectories. The solutions can be solved efficiently because the design conditions are represented as linear matrix inequality (LMI) conditions. We apply the algorithm to nonlinear discrete-time systems. An illustrated example is given to show the validity of the proposed method.

View full abstract

This paper proposes a tendon-driven mechanism with a spring in order to improve a grasping stability of a robotic hand. The properties of the tendon drive system with an elasticity are investigated from the viewpoint of finger posture in numerical simulations. The grasping stability generally decreases when a grasping object is displaced by an external force affected. The stability decrement of the tendon-driven mechanism becomes less than that of the conventional joint drive system, because the joint drive system allows a fingertip to rotate following to the external force. As a result, the fingertip tends to slip at grasping point. On the other hand, a fingertip of the tendon-driven mechanism rotates in the opposite direction in the same case. Besides the tendon-driven system mechanically and simultaneously responds to the disturbance without a time delay because it does not requre any sensors and feedback systems.

View full abstract