• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.11
• /
• pp.89-98
• /
• 1998

The enhanced resource-allocating network(ERAN) that adaptively generates hidden units of radial basis function(RBF) network for systems modeling has been proposed. The ERAN is an improved version of the resource-allocating network(RAN) that allocates new hidden units based on the novelty of observation data. The learning process of the ERAN involves allocation of new hidden units and adjusting the network parameters. The network starts with no hidden units. As observation data are received, the network adds a hidden units only if the three network growth criteria are satisfied. The network parameters are adjusted by the LMS algorithm. The performance of the ERAN is compared with the RAN for nonlinear static systems modeling problem with sequential and random learning. For two simulations, the ERAN has been shown to realize RBF networks with better accuracy with fewer hidden units.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.8
• /
• pp.1937-1947
• /
• 1996

In the case of developing the real-time system using object-oriented method, k the problem of the timing constraints is certainly considered. we propose the method of modeling the object-oriented real-time system using the OMT methodology and the SDL. And we also present the predictable time table that reflects the constraints of real-time system into dynamic model of OMTs and the predicatable time formula of the sequence, repeat, and parallel routine. The proposed method is applied to the estimate of the maximum process time of the ATMs(Automatic teller machines) and is used to specifying the functional specification for the user interface of the ATMs using the SDL syntax and the object interaction graph.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.5
• /
• pp.512-519
• /
• 2003

In this paper, we propose a neuro-fuzzy modeling to improve the performance using the hierarchical clustering and Gaussian Mixture Model(GMM). The hierarchical clustering algorithm has a property of producing unique parameters for the given data because it does not use the object function to perform the clustering. After optimizing the obtained parameters using the GMM, we apply them as initial parameters for Adaptive Network-based Fuzzy Inference System. Here, the number of fuzzy rules becomes to the cluster numbers. From this, we can improve the performance index and reduce the number of rules simultaneously. The proposed method is verified by applying to a neuro-fuzzy modeling for Box-Jenkins s gas furnace data and Sugeno's nonlinear system, which yields better results than previous oiles.

• Journal of Korea Society of Industrial Information Systems
• /
• v.4 no.2
• /
• pp.73-80
• /
• 1999

A major concern in Concurrent Engineering is the control and management of workload As a general rule, leveling the peak of workload in a period is difficult because concurrent processing is comprised of various processed, including overlapping, paralleling and looping and so on. Therefore workload management with resource constraints is so beneficial that effective methods to analyze design process are momentous. This paper presents a procedure to transform the IDEF3 process model into the precedence network model for more useful assessment of the process. This procedure is expected to facilitate resolving resource constrained scheduling problems more systematically in Concurrent Engineering environment.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.4
• /
• pp.7-16
• /
• 2009

A robot must be able to generate various skills to achieve given tasks intelligently and reasonably. The robot must first learn affordances to generate the skills. An affordance is defined as qualities of objects or environments that induce actions. Affordances can be usefully used to generate skills. Most tasks require sequential and goal-oriented behaviors. However, it is usually difficult to accomplish such tasks with affordances alone. To accomplish such tasks, a skill is constructed with an affordance and a soft behavioral motivation switch for reflecting goal-oriented elements. A skill calculates a behavioral motivation as a combination of both presently perceived information and goal-oriented elements. Here, a behavioral motivation is the internal condition that activates a goal-oriented behavior. In addition, a robot must be able to execute sequential behaviors. We construct skill networks by using generated skills that make action selection feasible to accomplish a task. A robot can select sequential and a goal-oriented behaviors using the skill network. For this, we will first propose a method for modeling and learning Bayesian networks that are used to generate affordances. To select sequential and goal-oriented behaviors, we construct skills using affordances and soft behavioral motivation switches. We also propose a method to generate the skill networks using the skills to execute given tasks. Finally, we will propose action-selection-mechanism to select sequential and goal-oriented behaviors using the skill network. To demonstrate the validity of our proposed methods, "Searching-for-a-target-object", "Approaching-a-target-object", "Sniffing-a-target-object", and "Kicking-a-target-object" affordances have been learned with GENIBO (pet robot) based on the human teaching method. Some experiments have also been performed with GENIBO using the skills and the skill networks.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.4
• /
• pp.315-321
• /
• 2012

In this study, a sequential multiscale homogenization method to characterize the effective thermal conductivity of nano particulate polymer nanocomposites is proposed through a molecular dynamics(MD) simulations and a finite element-based homogenization method. The thermal conductivity of the nanocomposites embedding different-sized nanoparticles at a fixed volume fraction of 5.8% are obtained from MD simulations. Due to the Kapitza thermal resistance, the thermal conductivity of the nanocomposites decreases as the size of the embedded nanoparticle decreases. In order to describe the nanoparticle size effect using the homogenization method with accuracy, the Kapitza interface in which the temperature discontinuity condition appears and the effective interphase zone formed by highly densified matrix polymer are modeled as independent phases that constitutes the nanocomposites microstructure, thus, the overall nanocomposites domain is modeled as a four-phase structure consists of the nanoparticle, Kapitza interface, effective interphase, and polymer matrix. The thermal conductivity of the effective interphase is inversely predicted from the thermal conductivity of the nanocomposites through the multiscale homogenization method, then, exponentially fitted to a function of the particle radius. Using the multiscale homogenization method, the thermal conductivities of the nanocomposites at various particle radii and volume fractions are obtained, and parametric studies are conducted to examine the effect of the effective interphase on the overall thermal conductivity of the nanocomposites.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.166-166
• /
• 2017

모돈의 건강 상태를 정량 지수화 하기 위한 연구를 수행 중이다. 지제이상, 섭식 불량, 수면 패턴 등의 운동 특성 분석을 위하여 복수의 초음파 센서를 이용하였다. 시계열 계측 신호를 분석하여 정량 지수화를 수행하는 과정에서 주파수 도메인 분석을 시도하였다. 이 과정에서 주파수 도메인의 분해능에 따른 편차 극복을 위한 비선형 모델링을 수행하였다. 또한 인접한 시계열 데이터 구간 간의 상관성 분석이 가능하면 대용량 데이터의 실시간 처리로 인한 지연 시간 극복 및 기대되는 예후에 대한 조기 진단이 가능할 것이다. 본 연구에서는 구글에서 제공하는 Tensorflow와 NVIDIA에서 제공하는 CUDA 엔진을 동시 적용한 심층 학습 시스템을 이용하였다. 전 처리를 위하여 주파수 분해능 (2분, 3분, 5분, 7분, 11분, 13분, 17분, 19분)에 따른 데이터 집합을 1단계로 두고, 상위 10 순위 안에 드는 파워 스펙트럼 밀도의 크기를 2단계로 하여, 총 2~10개의 입력 노드를 순차적으로 선정하였고, 동일한 방식으로 인접한 시계열의 파워 스펙터럼 밀도를 순위를 변화시켜 지정하였다. 대표적인 심층학습 모델인 Softmax regression with a multilayer convolutional network를 이용하여 Recursive feature selection 경우의 수를 $8{\times}9{\times}9$로 총 648 가지 선정하고, Epoch는 10,000회로 지정하였다. Calibration 모델링의 경우 Cost function이 10% 이하인 경우 해당 경우의 학습을 중단하였으며, 모델 간 상호 교차 검증을 수행하기 위하여 $_8C_2{\times}_8C_2{\times}_8C_2$ 경우의 수에 대한 Verification test를 수행하였다. Calibration 과정 상 모든 경우에 대하여 10% 이하의 Cost function 값을 보였으나, 검증 테스트 과정에서 모든 경우에 대하여 $r^2$ < 0.5 인 결정 계수 값이 나타났다. 단적으로 심층학습 모델의 과도한 적합(Over fitting) 방식의 한계를 보인 것이라 판단할 수 있다. 적합한 Feature selection 및 심층 학습 모델에 대한 지속적이고 추가적인 고려를 통해 과도적합을 해소함과 동시에 실효적이고 활용 가능한 Classification을 위한 입, 출력 노드 단의 전후 Indexing, Quantization에 대한 고려가 필요할 것이다. 이를 통해 모돈 생체 정보 정량화를 위한 지능형 현장 진단 기술 연구를 지속할 것이다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.1
• /
• pp.32-42
• /
• 2000

This paper presents an on-line learning algorithm for sequential construction of generalized radial basis function networks (GRBFNs) to model nonlinear systems from empirical data. The GRBFN, an extended from of standard radial basis function (RBF) networks with constant weights, is an architecture capable of representing nonlinear systems by smoothly integrating local linear models. The proposed learning algorithm has a two-stage learning scheme that performs both structure learning and parameter learning. The structure learning stage constructs the GRBFN model using two construction criteria, based on both training error criterion and Mahalanobis distance criterion, to assign new hidden units and the linear local models for given empirical training data. In the parameter learning stage the network parameters are updated using the gradient descent rule. To evaluate the modeling performance of the proposed algorithm, simulations and their results applied to two well-known benchmarks are discussed.

• Aerospace Engineering and Technology
• /
• v.2 no.1
• /
• pp.140-152
• /
• 2003

Scenario is a guiding principle of launch operation and control for rocket and ground support system. Therefore, developing a scenario is the first step to prepare for rocket launch, which is a critical task for success of KSR-III flight test. The launch scenario for KSR-III flight test is a procedural sequence of command and control signals to be given to rocket and ground support systems. In this paper, the UML object modeling method is applied to development of a launch scenario. First, the subsystems of the launch system are modeled by objects, and then the interfaces between each two subsystems are modeled by association links. The finally obtained object diagram of KSR-III launch system is used to analyzing flow of data and commands and control signals, and interactions. The scenario includes the sequences of pre-launch/launch operations and emergency operations.

• Journal of the Korea Society for Simulation
• /
• v.25 no.3
• /
• pp.133-146
• /
• 2016

To engage multiple missiles in single launcher against multiple targets, launcher system has to operate for optimized launch angle to each target sequentially. If the launch angle sequence is simply defined according to the target assignment order only, overall engagement time would be increased, and even in some engagement scenarios, it could be possible to miss some moving targets being out of proper engagement area. Therefore, the study on methodology for a real-time decision of optimized launch angle sequence is necessary. In this paper, the automatic decision model of launch angle sequence was suggested to minimize total engagement time by analyzing the simulation results of all engagement sequence set for multiple moving target scenario. Performance of proposed methodology for decision of optimal launch angle sequence was verified by comparing with the optimal or suboptimal sequence obtained from simulation results.