• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.3
• /
• pp.603-610
• /
• 2000

In this paper, the direct and indirect adaptive controller are combined based on the Lyapunov synthesis approach. The Proposed controller is constructed from RBF Neural Network and weighting parameters are adjusted on-line according to some adaptation law. In this scheme, fuzzy IF-THEN rules are used to decide the combined weighting factor. In the results, proposed controller has the main advantages of both the direct adaptive controller and the indirect adaptive controller. The effectiveness of the proposed control scheme is demonstrated through simulation results of control for one-link rigid robotics manipulator.

• Journal of Intelligence and Information Systems
• /
• v.3 no.1
• /
• pp.69-81
• /
• 1997

In this paper, first we propose an architecture of fuzzy neural networks with triangular fuzzy weights. The proposed fuzzy neural network can handle fuzzy input vectors. In both cases, outputs from the fuzzy network are fuzzy vectors. The input-output relation of each unit of the fuzzy neural network is defined by the extention principle of Zadeh. Also we define a cost function for the level sets(i. e., $\alpha$-cuts)of fuzzy outputs and fuzzy targets. Then we derive a learning algorithm from the cost function for adjusting three parameters of each triangular fuzzy weight. Finally, we illustrate our a, pp.oach by computer simulation examples.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1887-1888
• /
• 2008

본 논문에서는 퍼지 추론 기반의 다항식 RBF 뉴럴네트워크(Polynomial Radial Basis Function Neural Network; pRBFNN)를 설계하고 PSO(Particle Swarm Optimization) 알고리즘을 이용하여 모델의 파라미터를 동정한다. 제안된 모델은 "IF-THEN" 형식으로 기술되는 퍼지 규칙에 의해 조건부, 결론부, 추론부의 기능적 모듈로 표현된다. 조건부의 입력공간 분할에는 HCM 클러스터링에 기반을 두어 구조가 결정되며, 기존에 주로 사용된 가우시안 함수를 RBF로 이용하고, 원뿔형태의 선형 함수를 제안한다. 또한 입력공간 분할시 데이터 집합의 특성을 반영하기 위해 분포상수를 각 입력마다 고려하여 설계함으로서 공간 분할의 정밀성을 높인다. 결론부에서는 기존 상수항의 연결가중치를 다항식 형태로 표현하는 pRBFNN을 제안한다. 제안한 모델의 성능을 평가하기 위해 Box와 Jenkins가 사용한 가스로 시계열 데이터를 적용하고, 기존 모델과의 근사화와 일반화 능력에 대하여 토의한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.130-132
• /
• 2001

신경망을 이용한 추천 기술은 항목이나 사용자간의 가중치를 학습할 수 있고, 자료 유형에 상관없이 데이터 처리가 용이하다. 또한 최근 연구를 통해서 그 우수성이 입증되고 있다. 그러나 사용자간의 상관관계로 추천하는 사용자 신경망 모델과 항목간의 상관관계로 추천하는 항목 신경망 모델이 서로 다른 관점으로 다른 선호도를 제시한 경우에 선택한 모델의 선호도에 따라 시스템의 성능이 좌우된다. 그러므로 효율적이고 성능이 우수한 추천 시스템을 위해 사용자와 항목 신경망 모델의 통합 방법을 제안한다. 두 모델 사이에 우선 순위를 결정하여 통합하는 순차적 통합 방법과 두 모델을 동시에 고려하는 병렬적 통합방법을 제안한다. 그러나 두 통합 방법은 선호도 예측 기준에 있어서 정적이고, 문제에 대한 적응성이 없다. 그러므로 신경망(퍼셉트론, 다층 퍼셉트론)을 이용한 통합 방법을 제안한다. 또한 퍼지의 소속함수를 이용하여 퍼지 추론를 적용한 통합 방법을 제안하고, 패턴 인식 분야에서 사용하는 BKS 방법을 적응하여 두 신경망 모델을 통합하여 실험한다. 본 논문에서는 사용자와 항목 신경망 모델을 통합함으로써 기존의 추천 기술인 연관 규칙과 단일 신경망 모델을 이용한 추천보다 우수함을 보이고 있다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.2
• /
• pp.7-12
• /
• 2009

The new approach was proposed which uses feature extraction based on fuzzy integral in the process of face verification using morphological shape decomposition. The centre of area was used with image pixels related with structure element and its weight in an attempt to consider neighborhood information. Therefore the morphological operators were defined for feature extraction. And then the number of decomposition images were more about 4 times than the conventional. Finally in the simulations with the extractions for face verification it was proved that the approach in this paper was even more good than the conventional in stability of feature extraction and threshold value.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.4
• /
• pp.507-512
• /
• 2015

In the wireless localization systems using range measurements, iteration method-based approximated solutions have been used. Also, linear closed-form solutions have been investigated in the light of local minimum problem and computational load. However, each closed-form solution has individual error factors that cause usage limit of the solutions. In this paper, a fusion method integrating two representative closed-form solutions is presented. The presented method cancels the error factors of each solution out. Weights for integrating the standalone solutions are determined using the error factors-based fuzzy method. The performance of the proposed method is verified using some simulation results.

• Archives of design research
• /
• v.13 no.1
• /
• pp.131-138
• /
• 2000

The evaluation of design is one of the most important steps in design process. However, because in most cases, design evaluation has been disregarded in comparison with other steps and has been performed by designer's subjective judgment. there are few studies for the development of systematic methodology. In this study, we developed the method of design evaluation using the fuzzy theory and the multi-criteria dedsion making model. This method consists of two steps. The first step is to obtain relative importance weights of design evaluation criteria, and the second step is to obtain and integrate scores for design evaluation criteria. AHP(analytic hierarchy process) and fuzzy theory were applied to the first and the second step, respectively. We applied the developed method to the sensible evaluation of cellular phone design. As the result, we verified that the developed method could be effectively used in the evaluation of product design.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2001.06a
• /
• pp.141-144
• /
• 2001

In this paper, we proposed a neural network controller with a fuzzy compensator which compensate a output of neural network controller. Even if learn by neural network controller, it can occur a bad results from disturbance or load variations. So in order to adjust above case. we used the fuzzy compensator to get an expected results. And the weight of main neural network can be changed with the result of learning an inverse model neural network of plant, so a expected dynamic characteristics of plant can be got. In order to confirm a performance of the proposed controller, we implemented the controller using the DSP processor and applied in a hydraulic servo system. And then we observed an experimental results.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.4
• /
• pp.75-79
• /
• 2009

In this paper the method was proposed which uses ${\lambda}$-fuzzy measure to detect the edge of the features of the face region. In the conventional method the features was founded using valley, brightness and edge. This method had its drawbacks that it is so sensitive to the external noises and environments. This paper proposed ${\lambda}$-fuzzy measure to cope with this drawbacks. By considering each weight of the pixels the integral evaluation was considered using the center of area method. Thus the continuity of the edge was kept by way of the neighborhood information and the reduction of time complexity wad resulted in.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.4
• /
• pp.530-536
• /
• 2008

Localization is one of the fundamental problems in wireless sensor networks (WSNs) that forms the basis for many location-aware applications. Localization in WSNs is to determine the position of node based on the known positions of several nodes. Most of previous localization method use triangulation or multilateration based on the angle of arrival (AOA) or distance measurements. In this paper, we propose an enhanced centroid localization method based on edge weights of adjacent nodes using fuzzy modeling and genetic algorithm when node connectivities are known. The simulation results shows that our proposed centroid method is more accurate than the simple centroid method using connectivity only.