Journal of Intelligent & Fuzzy Systems - Volume 15, issue 3-4

Authors: Palade, Vasile | Jain, Lakhmi C.

Article Type: Research Article

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 137-138, 2004

Authors: Suksmono, Andriyan Bayu | Hirose, Akira

Article Type: Research Article

Abstract: This paper describes an intelligent beamforming (IBF) system based on complex-valued neural network (CVNN). A multilayer network structure with complex-valued neurons has been used. The system employs the complex-valued backpropagation algorithm (CVBPA) to intelligently adapt incoming signals impinging to sensors array. Performance of the CVNN-IBF system is compared with that of the conventional single-layer adaptive system using complex-valued least mean square (CLMS) algorithm. Experiments for multiple beam-pointing and multiple null-steering demonstrate that …the CVNN-IBF outperforms the CLMS one in terms of convergence speed and interferences suppression levels. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 139-147, 2004

Authors: Matsui, Nobuyuki | Isokawa, Teijiro | Kusamichi, Hiromi | Peper, Ferdinand | Nishimura, Haruhiko

Article Type: Research Article

Abstract: Quaternion neural networks are models in which computations of the neurons are based on quaternions, the four-dimensional equivalents of imaginary numbers. This paper shows by experiments that the quaternion-version of the Back Propagation (BP) algorithm achieves correct geometrical transformations in three-dimensional space, as well as in color space for an image compression problem, whereas real-valued BP algorithms fail. The quaternion neural network also performs superior in terms of convergence speed to a real-valued neural …network with respect to the 3-bit parity check problem, as simulations show. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 149-164, 2004

Authors: Matsumura, Yuji | Fukumi, Minoru | Akamatsu, Norio | Nakaura, Kazuhiro

Article Type: Research Article

Abstract: Information terminals in recent years, including cellular phones, have high performances due to new advances in IT. If a standard (such as Bluetooth) is used, it enables us to collect and to perform operational interfaces of various apparatuses using one equipment only. For example, a cellular phone can be turned on and off, made into manners mode, a CD player's volume can be easily regulated, and so on. We call this "total operation device". However, such …a device is not available yet. Therefore, we propose a recognition system based on wrist movements by focusing on ElectroMyoGram (EMG), using the body signals generated by voluntary movements of subject muscles, as the initial stage for construction of the total operation device. This paper tries to recognize EMG signals using neural networks (NNs). The electrodes under the dry state are attached to wrists and then EMG signals are measured. These EMG signals are classified using NNs into seven categories: neutral, up and down, right and left, inside twist, outside twist. The NN learns the FFT spectra of these signals in order to classify them. Moreover, we introduce a modular structure of the NN for improving the recognition accuracy. Computer simulations show that our approach is effective to classifying the EMG signals. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 165-171, 2004

Authors: Korekado, Keisuke | Morie, Takashi | Nomura, Osamu | Ando, Hiroshi | Nakano, Teppei | Matsugu, Masakazu | Iwata, Atsushi

Article Type: Research Article

Abstract: Hierarchical convolutional neural networks represent a well-known robust image-recognition model. In order to apply this model to robot vision or various intelligent vision systems, its VLSI implementation with high performance and low power consumption is required. This paper proposes a VLSI convolutional network architecture using a hybrid approach composed of pulse-width modulation (PWM) and digital circuits. We call this approach merged/mixed analog-digital architecture. The VLSI chip includes PWM neuron circuits, PWM/digital converters, …digital adder-subtracters, and digital memory. We have designed and fabricated a VLSI chip by using a 0.35 μm CMOS process. The VLSI chip can perform 6-bit precision convolution calculations for an image of 100 × 100 pixels with a receptive field area of up to 20 × 20 pixels within 5 ms, which means a performance of 2 GOPS. Power consumption of PWM neuron circuits was measured to be 20 mW. We have verified successful operations using a fabricated VLSI chip. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 173-179, 2004

Authors: Stefanoiu, Dan

Article Type: Research Article

Abstract: This article aims to introduce a non-conventional method for extracting, decoding and classifying the information regarding defects that could appear and develop in components of rotating machinery (e.g. in bearings). Such an information is encoded by vibrations and regards the type and the severity degree of possible defects. Though vibration is not as complex as speech or seismic signals, decoding this information is not easy. Unlike many Signal Processing applications, where the noise is attenuated or …compensated, one focuses here on the noisy component of vibrations, where defects are encoded. The classification of defects is based upon some statistical and fuzzy concepts described within the article. A succinct comparison between Fuzzy-Statistical and Envelope Analysis Methods is performed as well. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 181-194, 2004

Authors: Bocaniala, Cosmin Danut | Sa da Costa, Jose | Palade, Vasile

Article Type: Research Article

Abstract: This paper introduces a novel fuzzy classification methodology for fault diagnosis. The main advantages of the proposed fuzzy classifier are the high accuracy of defining the areas corresponding to different categories and the fine precision of discrimination inside overlapping areas. The fuzzy sets used by the classifier are built upon a similarity measure between the objects in the problem space. Another advantage of the classifier is its capability to handle either single or hybrid similarity measures. …The methodology has been validated by application to a fault diagnosis problem. The classifier has shown excellent performances in diagnosing faults to a control flow valve from an industrial device. Show more

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 195-205, 2004

Authors: Lee, S.H. | Howlett, R.J. | Walters, S.D.

Article Type: Research Article

Abstract: Small spark-ignition gasoline-fuelled internal-combustion engines can be found all over the world performing in various roles including power generation, agricultural applications and motive power for small boats. To attain low cost, these engines are typically air-cooled, use simple carburettors to regulate the fuel supply, and employ magneto ignition systems. Electronic control, of the sort found in automotive engines, has seldom proved cost-effective for use with small engines. However, the future trend towards …engines that have low levels of polluting exhaust emissions will make electronic control necessary, even for small engines. This paper describes a fuzzy control system applied to a small engine to achieve regulation of the fuel injection system. The system determines the amount of fuel required by a fuzzy algorithm that uses the engine speed and manifold air pressure as input values. The parameters of this fuzzy control paradigm were a collection of rules and fuzzy-set membership functions. These were intuitively comprehensible by the operator. This facilitated the calibration process, leading to quick and convenient tuning. Experimental results show that a considerable improvement in fuel regulation was achieved compared to the original carburettor-based engine configuration. In addition measurements of HC and CO emissions show a corresponding reduction. Show more

Keywords: engine management systems, engine control, fuzzy control, emissions reduction, intelligent system, applied artificial intelligence

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 207-217, 2004

Authors: Watabe, Hirokazu | Kawaoka, Tsukasa

Article Type: Research Article

Abstract: To design behaviors of a mobile robot for realizing given tasks, a designer has to make a set of rules which generates a proper action from a state of sensors. In general, however, it is difficult for the designer to make the complete set of rules since the number of rules is very large and the proper action for a given state of sensors is not clear. Therefore, the robot must learn and construct the knowledge base of actions by …itself. This paper proposes a learning algorithm to construct the knowledge of action in order to achieve tasks that are given to the mobile robot. The action to achieve a task in an environment is generated by a genetic algorithm. It is also shown that repeating the knowledge extraction will make the construction of the Action Knowledge-Base possible, concerning the task in any situation. Show more

Keywords: mobile robot, genetic algorithms, motion learning, action knowledge-base

Citation: Journal of Intelligent & Fuzzy Systems, vol. 15, no. 3-4, pp. 219-223, 2004