• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.5
• /
• pp.395-400
• /
• 2000

In this paper, we propose a new neuro-fuzzy controller based on reinforcement learning. The proposed system is composed of neuro-fuzzy controller which decides the behaviors of an agent, and dynamic recurrent neural networks(DRNNs) which criticise the result of the behaviors. Neuro-fuzzy controller is learned by reinforcement learning. Also, DRNNs are evolved by genetic algorithms and make internal reinforcement signal based on external reinforcement signal from environments and internal states. This output(internal reinforcement signal) is used as a teaching signal of neuro-fuzzy controller and keeps the controller on learning. The proposed system will be applied to controller optimization and adaptation with unknown environment. In order to verifY the effectiveness of the proposed system, it is applied to collision avoidance of an autonomous mobile robot on computer simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.11
• /
• pp.2486-2491
• /
• 2013

There are invisible defects by naked eyes in ceramic material images such as internal stomata, cracks and foreign substances. In this paper we propose a method to detect and extract such defects from ceramic pipe weld zone by applying ART2 learning. In pre-processing, we apply Ends-in Search Stretching to enhance the intensity and then perform fuzzy binarization with triangle type membership function followed by enhanced ART2 that interacts with random input patterns to extract such invisible defects. The experiment verifies that this proposed method is sufficiently effective.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.4
• /
• pp.77-86
• /
• 1995

The new architecture that fuzzy logic control(FLC) with difficulties for tuning membership function (MF) is parallel with neural networks(NN) to be learned from the output of FLC is proposed. Therefore proposed scheme has the characteristics to utilize the expert knowledge in design process, to be learned during the operation without any learning mode. In this architecture, the function of the FLC is to supply the sliding surface which is constructed on the phase plane by rule base for giving the desired control characteristics and learning criterion of NN and the stabilization of the control performance before NN is learned, The function of the NN is to let the system trajectory be tracked to the sliding surface and reached to the stable point.

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.39 no.4
• /
• pp.370-378
• /
• 2002

Fuzzy inference which combines numerical data and linguistic data has been used to design adaptive filter algorithms. In adaptive IIR filter design, the fuzzy prefilter is taken account, and applied to both direct and lattice structure. As for the fuzzy inference of the fuzzy filter, the Sugeno's method is employed. As membership functions and inference rules are recursively generated through neural network, the accuracy can be improved. The proposed adaptive algorithm, adaptive IIR filter with fuzzy prefilter, has been applied to adaptive system identification for the purposed of performance test. The evaluations have been carried out with viewpoints of convergence property and tracking properties of the parameter estimation. As a result, the faster convergence and the better coefficients tracking performance than those of the conventional algorithm are shown in case of direct structures.

• Journal of Internet Computing and Services
• /
• v.8 no.3
• /
• pp.109-118
• /
• 2007

This paper proposed a method that can be simply analyzed instead of the basic general Fuzzy rule that its insufficient characters are cut out. Based on the proposed method. Rough sets are used to eliminate the incomplete attributes included in the rule and also for a classification more precise; the agreement of the membership function's output extracted the maximum attributes. Besides, the proposed method in the simulation shows that in order to verify the validity, compare the max-product result of fuzzy before and after reducing rule hosed on the rice taste data; then, we can see that both the max-product result of fuzzy before and after reducing rule are exactly the same; for a verification more objective, we compared the defuzzificated real number section.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.9
• /
• pp.189-193
• /
• 2013

In this paper, in order to predict and prevent electrical fires that occur in the power system, we measured the informations of electric power, and then proposed a system to predict the electrical fire using these informations. To this end, we analyzed the correlations for over-current, overload and overheating. These states are caused by the grounding current and the leakage current, and are the main causes of an electrical fire. Use these correlations to derive the derivative of the fuzzy rules for membership function. The designed algorithm was simulated by utilizing the informations of the actual power of the switchgear-panel.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.35-36
• /
• 2012

The generation of fuzzy rules is inevitable in order to construct fuzzy modeling and in general, has the problem that the number of rules increases exponentially with increasing dimension. To solve this problem, we introduce the system that generate the fuzzy rules and make a inference based on FCM clustering algorithm that partition the input space in the scatter form. The parameters in the premise part of the fuzzy rules is determined as membership matrix by the FCM clustering algorithm and the consequence part of the fuzzy rules is are expressed as a polynomial function. Proposed model evaluated using the numerical data.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.3 no.4
• /
• pp.388-398
• /
• 1998

In constructing fuzzy control systems. there are many parameters such as rule base. membership functions. inference m method. defuzzification. and I/O scaling factors. To control the system in properly using fuzzy logic. we have to consider t the correlation with those parameters. This paper deals with self-tuning of fuzzy control systems. The fuzzy controller h has parameters that are input and output scaling factors to effect control output. And we propose the looklongleftarrowup table b based self-tuning fuzy controller. We propose the PMTM(Plus-Minus Tuning Method) for self tuning method, self-tuning the initial look-up table to the appropriate table by adding and subtracting the values.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.6
• /
• pp.613-620
• /
• 1992

Recently, there are lots of research work on fuzzy Information theory for many practlcal applications. As the fuzzy control systems become to be sophisticated, they demand more fuzzy parameters, membership functions and fuzzy Inference rules. Eventually, they need effective parallel computing architectures to implement those complex fuzzy inference rules. In this paper, a optical fuzzy Inference system based on 2-D spatial light modulator and digital image board Is Implemented as a new approach for real-time parallel fuzzy computing system. From its good experimental results on the practical fuzzy airconditioner system, a new real-time Opto Fuzzy Inference system Is suggested.

• Journal of Internet Computing and Services
• /
• v.10 no.5
• /
• pp.107-116
• /
• 2009

This paper presents an approach to detect arrhythmia using heart rate variability and a fuzzy neural network. The proposed algorithm diagnoses arrhythmia using 32 RR-intervals that are 25 seconds on average. We extract six statistical values from the 32 RR-intervals, which are used to input data of the fuzzy neural network. This paper uses the neural network with weighted fuzzy membership functions(NEWFM) to diagnose arrhythmia. The NEWFM used in this algorithm classifies normal and arrhythmia. The performances by Tsipouras using the 48 records of the MIT-BIH arrhythmia database was below 80% of SE(sensitivity) and SP(specificity) in both. The detection algorithm of arrhythmia shows 88.75% of SE, 82.28% of SP, and 86.31% of accuracy.