• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1999년도 제14차 학술회의논문집
• /
• pp.206-209
• /
• 1999

Conventional target tracking techniques are primarily based on Kalman filtering or probabilistic data association(PDA). But it is difficult to perform well under a high cluttered tracking environment because of the difficulty of measurement, the problem of mathematical simplification and the difficulty of combined target detection for tracking association problem. This paper deals with an analysis of target tracking problem using fuzzy-logic theory, and determines fuzzy rules used by a fuzzy tracker, and designs the fuzzy tracker by using fuzzy rules and Kalman filtering.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1989년도 가을 학술발표회 논문집
• /
• pp.34-39
• /
• 1989

Recent trends in design standards development have encouraged the use of probabilistic limit sate design concepts. Reliability analysis adopted in those advanced countries have the potentials that they afford for symplifying the design Process arid placing it on a consistent reliability based for various construction materials. This study is proposed in the reliability analysis of plane frame structures using second-order moment method(Level-II they). Lind-Hasofer's minimum distance method is use in the derivation of an mathematical algorithm as well as an determination of Correlation cofficients, reliability index and total reliability index depending on the multiple failure modes. In addition. This study is employed as a practical tool for the approximate reliability analysis. Results of the numerincal analysis showed that the difference between the reliability index of the failure probability of the multiple failure modes and the total reliability index of the failure probability with the simultaneous failure modes deviated nearly 3∼10% depending on tile performance functions.

• 한국게임학회 논문지
• /
• 제14권6호
• /
• pp.109-118
• /
• 2014

이미지 디자인 등에 사용하기에 용이한 정다각형의 회전대칭성을 갖는 프랙탈 생성에 대해 연구하였다. Loocke의 논문[13]에서 사용한 방법과 같이 회전, 축소 아핀함수를 기반으로 하되 제곱근(square root)함수 대신 줄리아 셋(Julia set)을 생성하는 함수들로 확장하여 IFS(iterated function systems)를 구성하였다. 그 결과 줄리아 셋의 모양에 바탕을 둔 회전 대칭적 프랙탈을 생성할 수 있었으며, 줄리아 셋의 모양이 잘 나타나지 않는 경우에는 IFS 생성 알고리즘의 확률적 함수선택 부분을 변경하여 줄리아 셋의 모양이 뚜렸해지도록 할 수 있음을 보였다. 또한 줄리아 셋의 모양을 지수의 변화를 통해 변형하는 방법을 제안하였다.

• The Journal of the Acoustical Society of Korea
• /
• 제17권4E호
• /
• pp.3-10
• /
• 1998

This paper concerns the dynamical behavior, in probabilistic sense, of a feedforward neural network performing auto association for novelty. Networks of retinotopic topology having a one-to-one correspondence between and output units can be readily trained using back-propagation algorithm, to perform autoassociative mappings. A novelty filter is obtained by subtracting the network output from the input vector. Then the presentation of a "familiar" pattern tends to evoke a null response ; but any anomalous component is enhanced. Such a behavior exhibits a promising feature for enhancement of weak signals in additive noise. As an analysis of the novelty filtering, this paper shows that the probability density function of the weigh converges to Gaussian when the input time series is statistically characterized by nonsymmetrical probability density functions. After output units are locally linearized, the recursive relation for updating the weight of the neural network is converted into a first-order random differential equation. Based on this equation it is shown that the probability density function of the weight satisfies the Fokker-Planck equation. By solving the Fokker-Planck equation, it is found that the weight is Gaussian distributed with time dependent mean and variance.

• Geomechanics and Engineering
• /
• 제7권4호
• /
• pp.431-458
• /
• 2014

Construction of a new cavern close to an existing cavern will result in a modification of the state of stresses in a zone around the existing cavern as interaction between the twin caverns takes place. Extensive plane strain finite difference analyses were carried out to examine the deformations induced by excavation of underground twin caverns. From the numerical results, a fairly simple nonparametric regression algorithm known as multivariate adaptive regression splines (MARS) has been used to relate the maximum key point displacement and the percent strain to various parameters including the rock quality, the cavern geometry and the in situ stress. Probabilistic assessments on the serviceability limit state of twin caverns can be performed using the First-order reliability spreadsheet method (FORM) based on the built MARS model. Parametric studies indicate that the probability of failure $P_f$ increases as the coefficient of variation of Q increases, and $P_f$ decreases with the widening of the pillar.

• Structural Engineering and Mechanics
• /
• 제41권5호
• /
• pp.575-589
• /
• 2012

This paper presents a reliability analysis of steel I-beams with rectangular web openings, based on a combination of the common probabilistic reliability methods, such as RSM, FORM and SORM and using data obtained from experimental tests performed at the Istanbul Technical University. A procedure is proposed to obtain the optimum design load that can be applied to this type of structural members, by taking into account specified target values of reliability indices for ultimate and serviceability limit states. The goal of the paper is to present an algorithm to obtain more realistic and economical design of beams and to demonstrate that it can be applied efficiently to steel I-beams with web openings. Finally, a sensitivity analysis is performed allowing to ranking the random variables according to their importance in the reliability analysis.

• 한국정보처리학회논문지
• /
• 제6권2호
• /
• pp.357-366
• /
• 1999

Inheritance reasoning has been widely used in the area of common sense reasoning in artificial intelligence. Although many inheritance reasoners have been proposed in artificial intelligence literature, most previous reasoning systems are lack of clear semantics, thus sometimes provide anomalous conclusions. In this paper, we describe a set-oriented inheritance reasoner and propose a method of resolving conflicts with clear semantics of defeasible rules. The semantics of default rule is provided by statistical analysis of $\chi$ method, and likelihood of rule is computed based on the evidence in the past. Two basic rules, specificity and generality, are defined to resolve conflicts effectively in the process of reasoning. We show that the mutual tradeoff between specificity and generality 추 prevent many anomalous results from occurring in traditional inheritance reasoners. An algorithm is provided. and some typical examples are given to show how the specificity/generality rules resolve conflicts effectively in inheritance reasoning.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2014년도 추계학술대회
• /
• pp.109-111
• /
• 2014

본 논문에서는 노드의 속성정보를 통해 노드의 움직임을 예측하는 PPoP(The Path Prediction algorithm based on Probability) 알고리즘을 제안한다. 기존 이동 예측 알고리즘들은 GPS(Global Positioning System)를 사용해 노드의 이동을 학습을 통해 패턴화 하여 예측한다. 이때, 노드들이 이동 패턴을 벗어날 경우 예측률이 떨어진다. 따라서 본 논문에서는 스테른 게를라흐의 실험(Stern-Gerlach experiment)을 분석하여 노드의 이동성을 예측하는 알고리즘을 제안한다. 본 논문에서 제안된 알고리즘에서는 노드의 이동 경로를 staore-carry-forward 방식으로 상황 인지에 의한 경로 설정 변경 예측 방법으로 이동 예측 확률 기법이다. 모의실험 결과 제안한 방법을 사용하여 노드의 이동성 및 패턴을 벗어난 상황에서도 노드의 예측 하고자 한다.

• Structural Engineering and Mechanics
• /
• 제61권3호
• /
• pp.347-357
• /
• 2017

The applications of active control is being more popular nowadays. Several control algorithms have been developed to determine optimum control force. In this paper, a Chaotic Particle Swarm Optimization (CPSO) technique, based on Logistic map, is used to compute the optimum control force of active tendon system. A chaotic exploration is used to search the solution space for optimum control force. The response control of Multi-Degree of Freedom (MDOF) shear buildings, equipped with active tendons, is introduced as an optimization problem, based on Instantaneous Optimal Active Control algorithm. Three MDOFs are simulated in this paper. Two examples out of three, which have been previously controlled using Lattice type Probabilistic Neural Network (LPNN) and Block Pulse Functions (BPFs), are taken from prior works in order to compare the efficiency of the current method. In the present study, a maximum allowable value of control force is added to the original problem. Later, a twenty-story shear building, as the third and more realistic example, is considered and controlled. Besides, the required Central Processing Unit (CPU) time of CPSO control algorithm is investigated. Although the CPU time of LPNN and BPFs methods of prior works is not available, the results show that a full state measurement is necessary, especially when there are more than three control devices. The results show that CPSO algorithm has a good performance, especially in the presence of the cut-off limit of tendon force; therefore, can widely be used in the field of optimum active control of actual buildings.

• 한국통신학회논문지
• /
• 제26권11A호
• /
• pp.1946-1956
• /
• 2001

본 논문에서는 관계 벡터 공간상의 특징 대응에 관한 확률적 해석에 기반한 새로운 부분 인식 기법을 제안한다. 효과적인 인식을 위해 물체를 관계 속성 그래프(Attributed Relational Graph; ARG)와 관계 벡터 공간들의 집합으로 표현한다. 또한 잡음이나 특징 소실로 인한 왜곡을 관계 벡터 공간에서의 관계 벡터 분포에 대한 왜곡으로 확률적으로 모델링한다. 제안하는 부분 인식 기법은 두 단계로 이루어진다. 우선 지역적인 특징(local feature)과 구조적인 일관성(structural consistency)을 사용하여 후보집합을 추출한다. 이렇게 추출된 후보집합 각각에 대해 관계 벡터 공간상에서의 에러 분석과 반복적인 voting 알고리즘을 통해 특징 소실을 검출한다. 실제 영상에 대한 실험 결과를 통해 제안한 알고리즘이 잡음이나 가리어짐이 심한 경우에도 강건한 성능을 보임을 알 수 있으며, 릴렉세이션(relaxation) 기법과 수행 시간 비교 분석을 통해 계산량 측면에서의 성능 향상을 확인할 수 있다.