Recently, we have presented a method of probabilistic prediction of chaotic time series. The method employs learning machines involving strong learners capable of making predictions with desirably long predictable horizons, where, however, usual ensemble mean for making representative prediction is not effective when there are predictions with shorter predictable horizons. Thus, the method selects a representative prediction from the predictions generated by a number of learning machines involving strong learners as follows: first, it obtains plausible predictions holding large similarity of attractors with the training time series and then selects the representative prediction with the largest predictable horizon estimated via LOOCV (leave-one-out cross-validation). The method is also capable of providing average and/or safe estimation of predictable horizon of the representative prediction. We have used CAN2s (competitive associative nets) for learning piecewise linear approximation of nonlinear function as strong learners in our previous study, and this paper employs bagging (bootstrap aggregating) to improve the performance, which enables us to analyze the validity and the effectiveness of the method.

In dense mobile wireless sensor networks (MWSNs), a number of major applications require monitoring of the geographical boundaries of sensor readings among physical phenomena with as low traffic as possible. In this paper, we propose a periodical boundary detection method reducing traffic by exchanging sensor data only among nodes near the boundaries. Our proposed method achieves this by incorporating recording the time when a node has crossed the boundary and overhearing sensor data. We confirmed the effectiveness of our proposed method through simulation experiments.

In dense mobile wireless sensor networks, it is desirable to gather sensor data with as low traffic as possible. When an application requires geographical boundaries of sensor readings, it is able to satisfy the requirement while reducing traffic by gathering sensor data only from nodes located close to the boundaries. In order to do so, it is necessary to detect nodes located close to the boundaries (boundary detection). Until now, many boundary detection methods have been proposed, assuming networks constructed by fixed nodes. In these methods, each node preliminarily recognizes locations of itself and all its neighboring nodes. When applying this approach to a dense and mobile environment, it takes a large amount of traffic for exchanging information on locations of nodes. In this paper, we propose an efficient boundary detection method in dense mobile wireless sensor networks, which makes use of overhearing sensor data. In our proposed method, each node determines whether it exists close to a boundary or not based on sensor data overheard from its neighboring nodes. When a node determines that it dose not exist close to the boundary, it stops transmitting its own sensor data. By doing so, traffic for boundary detection can be reduced. We confirmed the effectiveness of our proposed method through simulation experiments.

In dense mobile wireless sensor networks, it is desirable to reduce traffic for data gathering as much as possible while satisfying an application requirement. In this paper, focusing on an application which monitors the time-variant shape of events, we discuss an approach to achieve an efficient monitoring with small traffic in dense mobile wireless sensor networks.

In dense mobile sensor networks, to reduce traffic for gathering sensor data, it is desirable to efficiently gather sensor data from the minimum number of mobile sensor nodes necessary to guarantee the sensing granularity, which the application requires. For this aim, in our previous work, we have proposed a data gathering method using mobile agents (DGUMA). However, DGUMA does not consider geographical distribution of data values, and all agents have to send their own readings to the sink. Thus, multiple readings with the same value are sent to the sink. In this paper, we extend DGUMA in order to further reduce the traffic considering geographical distribution of data values. In the extended method, mobile agents aggregate multiple readings with the same value into one reading and forward the aggregated data to reduce the traffic. Moreover, our extended method dynamically controls communication routes from each agent to the sink so that the agents can effectively aggregate the readings. We confirmed the effectiveness of our proposed method through some simulation experiments.

In dense mobile wireless sensor networks, it is desirable to gather sensor data from the minimum number of nodes necessary to guarantee the sensing granularity. For this aim, in our previous work, we have proposed a data gathering method using mobile agents. However, this method cannot gather all the sensor data in the situation where some mobile agents cannot be deployed due to low node density. In this paper, we extend our previous method in order to gather all the sensor data even when some mobile agents disappear. First, when a mobile agent disappears, the extended method redeploys the disappeared agent. In addition, when the redeployment is failed, the extended method regathers sensor data which is missed by the disappearance of the agent. We confirmed the effectiveness of the extended method through some simulation experiments.

This paper describes an object tracking system for a robot working in an everyday environment, which tracks and recognizes everyday objects. Passive RFID (Radio Frequency IDentification) tags are attached to the objects for object recognition. The system consists of static sensors embedded in the environment and mobile sensors mounted on mobile robots. By utilizing the different characteristics and advantages of these sensors, the system achieves good performance in an everyday environment. Although the tag ID and the position of an object carried by a person is not measurable by static sensors or by mobile sensors, the system can estimate them by using an SIR (Sequential importance resampling) particle filter that integrates the data obtained by the static sensors and the mobile sensors. In the experiment, the system successfully tracked 20 objects, some of which were held by a person.

This paper describes a tactile sensing to estimate the position and orientation of a joint-axis of a linked object. This tactile sensing is useful when a multi-jointed multi-fingered robotic hand manipulates a tool which has a joint. This estimation requires sensing of the location of a contact point and the direction of an edge of the tool as contact information measured by a robotic fingertip. A conventional hard fingertip with a force sensor can measure only the location of a contact point. In contrast, we have already developed a robotic fingertip with a force sensor and a soft skin, and it can measure not only the location of a contact point but also the direction of an edge of an object. The estimation of a joint-axis of a linked object is demonstrated by using the soft fingertip.

This paper proposes a segmentation method of human manipulation task based on measurement of contact force imposed by a human hand on a grasped object. We define an index measure for segmenting a human manipulation task into primitives. The indices are calculated from the set of the contact forces measured at all the contact points during a manipulation task. Then, we apply the EM algorithm to the set of the indices in order to segment the manipulation task into primitives. These primitives are mapped onto the robotic hand to impose appropriate contact forces on a grasped object. In the experiments, manipulation tasks performed in daily human life have been successfully segmented.

The present paper describes a decision method for the placement of tactile elements for manipulation task recognition. Based on the mutual information of the manipulation tasks and tactile information, an effective placement of tactile elements on a sensing glove is determined. Although the effective placement consists of a small number of tactile elements, it has a recognition performance that is as high as that of a placement consisting of many tactile elements. The effective placement of tactile elements decided by the proposed method has been evaluated through experiments involving the recognition of grasp type from grasp taxonomy defined by Kamakura [1].