This paper proposes a pneumatic tubular body fixture for a wearable assistive device in order to hold a user body in simple wearing actions and with comfort. The pneumatic body fixture, called active cuff, is composed of multiple sets of a pneumatic actuator module which are actively transformed by compressed air. This transformed module wraps and holds a part of a user's body so that a wearable assistive device can be mounted on the user's body with less effort in a short time. The elastic material of the pneumatic actuator module also contributes a tactile and pressure sensation of a contact with the environment as if we could feel those sensations through our clothes. The pneumatic actuator is modeled by a beam structure and then deformation of the pneumatic actuator is simulated by finite element method for an active cuff design. The maximum holding force of the developed active cuff is about 6 kilograms, which is enough to hold an upper limb. The active cuff changes the holding points to keep blood flow of the holding area iteratively and alternatively using three couples of pneumatic actuator modules of the active cuff. Through some experiments, we confirmed that the active cuff enhanced the convenience and comfort of the wearable robot from the viewpoints of easiness to wear and the contact pressure distribution.

To ensure the safety of the handle type electric wheelchair at the time of day or night, we propose the hazard detection system by using the Kinect sensor for game, and carry out the performance evaluation of the system. In particular, we propose the detection method of the hazard object at the time of day or night, the size determination of the hazard object to be detected and the hazard detection area on the basis of the reaction time of the elderly people and Japan Industrial Standards (JIS). And then, after we build the hazard detection system, we conduct driving experiments that are the moving hazardous object detection in the time of day, and the stationary and the moving hazardous object in the time of day and night, and perform the evaluation of the detection ratio and the estimation accuracy of the system. These results show the feasibility of the proposed system in the handle type electric wheelchair.

In this paper, we propose a simple unevenness (bump/dent) detection system using a monocular camera to ensure safety of a handle type electric wheelchair and carry out the quantitative evaluation of a proposal system. In particular, the evaluation in two types of unevenness, three types of approach angle (90°, 45°, 0°) against the unevenness and with or without of installation-calibration error is performed. We show the effectiveness of the proposal system.

Discovering the significant relations embedded in documents would be very useful not only for information retrieval but also for question answering and summarization. Prior methods for relation discovery, however, needed large annotated corpora which cost a great deal of time and effort. We propose an unsupervised method for relation discovery from large corpora. The key idea is clustering pairs of named entities according to the similarity of context words intervening between the named entities. Our experiments using one year of newspapers reveals not only that the relations among named entities could be detected with high recall and precision, but also that appropriate labels could be automatically provided for the relations.

This paper describes an automatic content indexing system for news programs, with a special emphasis on its segmentation process. The process can successfully segment an entire news program into topic-centered news stories; the primary tool is a linguistic topic segmentation algorithm. Experiments show that the resulting speech-based segments are fairly accurate, and scene change points supplied by an external video processor can be of help in improving segmentation effectiveness.

Position recognition of vehicles is important in ITS (intelligent transport system). This paper describes a vehicle position detection scheme for traffic flow images with a small depression angle. It uses template images divided into several color regions in the HSV (hue, saturation and value) color transform to reduce computational complexity. Finally, this proposed system has mostly performed in estimation of the vehicles' positions. However, the initial templates generated in the initial template generation module affects the detection performance.

This paper describes a system to realize seamless high speed RVC (road-to-vehicle communications). This proposed system differs from the conventional system in considering cooperation with seamless as well as high-speed data communications. For performance evaluation of this proposed system, a simulator that include the protocol and handover of this system is constructed. The results show that this system can provide high speed communications in usual time, and also provide communications with the required speed even under the traffic congestion environment.

This paper proposes the Advanced Demand Signals scheme (ADS) that is a traffic signal control scheme by vehicle demands, and evaluates its performance. Each vehicle in some range transmits the relative distance between the vehicle and the intersection using seamless and real-time road-to-vehicle communications. The traffic signal decides the light according to the demand that is calculated using the transmitted data. Performance evaluations are carried out using the simulator including three intersections with traffic signals. This simulator uses the average idling time as a performance index, and examines the basic performance effect of communication range, and effect of performance function. Consequently, it is shown that the ADS scheme is effective in almost all vehicle densities, and, in particular, the ADS scheme shortens the average idling time more than by a maximum of 90% compared with the most basic conventional control scheme with low vehicle density.

In this paper, a frame synchronization method using the differential detector and racing counters is proposed. Moreover, the synchronization performance of the proposed system is discussed. In particular, the lose lock time, the recovery time and the BER performance considering the synchronizing performance are analyzed. In consequence, the BER performance considering the synchronization performance can approach the lower bound of the spread spectrum constrained spreading code (SS-CSC) system by tuning the number of the stages in the racing counters.

The performance of M-ary spread spectrum (M-ary/SS) communication systems is discussed. Firstly, the initial acquisition time is evaluated. Secondly, the retention time, which is the average number of frames holding correct frame timing, and the recovery time, which is the average number of frames required to establish synchronization, are derived. Lastly, the bit-error rate (BER) performance is evaluated. M-ary/SS communication systems, which have more than one spreading code, can improve the BER performance under conditions in which there is additive white Gaussian noise (AWGN). However, the synchronization of M-ary/SS communication systems is difficult because they have several spreading codes. The frame synchronization method uses a Hadamard matrix and "racing counters." As a result, the retention time becomes longer than the recovery time when the size of the lower counter differs greatly from that of the upper counter in the racing counters. Then the BER gets close to the performance which is achieved under complete synchronization.

We propose a new speech communication system to convert oral motion images into speech. We call this system "the image input microphone." It provides high security and is not affected by acoustic noise because it is not necessary to input the actual utterance. This system is especially promising as a speaking-aid system for people whose vocal cords are injured. Since this is a basic investigation of media conversion from image to speech, we focus on vowels, and conduct experiments on media conversion of vowels. The vocal-tract transfer function and the source signal for driving this filter are estimated from features of the lips. These features are extracted from oral images in B learning data set, then speech is synthesized by this filter inputted with an appropriate driving signal. The performance of this system is evaluated by hearing tests of synthesized speech. The mean recognition rate for the test data set was 76.8%. We also investigate the effects of practice by iterative listening. The mean recognition rate rises from 69.4% to over 90% after four tests over four days. Consequently, we conclude the proposed system has potential as a method of nonacoustic communication.>

We propose a detector using a Hopfield (1984) network for multiuser detection in M-ary/SSMA communications. It is shown that the Hopfield network can be used by considering the optimum multiuser detection problem in M-ary/SSMA as an optimization problem. The proposed detector has a simple structure. Firstly, we consider the multiuser detector using a Hopfield network of a discrete model in M-ary/SSMA communications and it is shown by computer simulations that it has better performance in near-far situations than the conventional detectors do. Secondly, we consider the detector using a Hopfield network of a continuous model and it is shown that the performance of the proposed detector is nearly equal to that of the optimum detector.

An improved scheme using vertical block allocation to embed data in industrial-quality monochrome analog pictures by modulo masking is investigated. The video signal on each scan line is sampled, and a data bit is inserted into a block of three pels by a scrambling of the luminance level of only one pel in the block. The performance of the system is compared to that of a conventional system. The number of data bits embedded in an image for the proposed system is about 1.3 times as large as that of the conventional system. In addition, the SNR (signal-to-noise ratio) of the recovered image in the proposed system is increased by about 3-4 dB.>