• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.4 no.2
• /
• pp.161-164
• /
• 2004

The yo-yo system has been introduced as an interesting plant to demonstrate the effectiveness of intelligent controllers. Having nonlinear and asymmetric characteristics, the yo-yo plant requires a controller quite different from conventional controllers such as PID. In this paper is presented an adaptive method of controlling the yo-yo system. Fuzzy logic controller based on human expertise is referred at first. Then, an adaptive fuzzy controller which has adaptation features against the variation of plant parameters is proposed. Finally, experimental results are presented.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.839-842
• /
• 2000

A new and modified neural network model Is proposed for CT image reconstruction from four projection directions only. The model uses the Resilient Back-Propagation (Rprop) algorithm, which is derived from the original Back-Propagation, for adaptation of its weights. In addition to the error in projection directions of the image being reconstructed, the proposed network makes use of errors in pixels between an image which passed the median filter and the reconstructed one. Improved reconstruction was obtained, and the proposed method was found to be very effective in CT image reconstruction when the given number of projection directions is very limited.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.887-890
• /
• 1993

By introducing the notion of constraint-oriented fuzzy inference, we will show that it provides us ways of fuzzy control methods that has abilities of adaptation, learning and self-organization. The basic supporting techniques behind these abilities are“hard”processing by Artificial Intelligence or traditional computational framework and“soft”processing by Neural Network or Genetic Algorithm techniques. The reason that these techniques can be incorporated to fuzzy control systems is that the notion of“constraint”itself has two fundamental properties, that is, the“modularity”property due to its declarativeness and the“logicality”property due to its two-valuedness. From the former property, the modularity property, decomposing and integrating constraints can be done easily and efficiently, which enables us to carry out the above“soft”processing. From the latter property, the logicality property, Qualitative Reasoning and Instance Generalization by Symbolic Reasoning an be carried out, thus enabling the“hard”processing.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.562-562
• /
• 2016

신경망 모형에서 학습이란 주어진 입출력시스템에 대하여 원하는 동작을 수행할 수 있도록 연결 강도를 최적의 상태로 적응(adaptation)시키는 과정을 의미한다. 따라서 강수와 지하수위의 관계를 연계시킨 인공신경망기법은 선택적으로 예측 지하수위에 영향을 미치는 변수들을 학습에 의하여 택함으로써 예측모형을 구성할 수 있다. 즉, 예측 지하수위와의 상관관계에 의하여 입력되는 변수와의 연결강도를 조정하여 매개변수 조정 및 모형의 최적화를 자동화할 수 있다. 본 연구에서는 지하수위에 영향을 주는 요소는 지하수위와 강우량이라고 가정하고, 지하수위의 입출력과정을 시계열 분석에 의하여 모형화하였으며 예측지하수위는 강우 및 지하수위의 선행조건과 매우 밀접한 관계를 갖는다. 따라서 선행강우 및 지하수위의 상태에 따라 이를 입력하여 미래의 지하수위를 예측하게 된다. 이 모형을 제주지역의 관측소에 적용한 결과 관측소별로 타당한 예측결과를 도출하였다.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.156-165
• /
• 2021

In this paper, we propose an histogram weighted centroid based Siamese region proposal network for object tracking. The original Siamese region proposal network uses two identical artificial neural networks which take two different images as the inputs and decide whether the same object exist in both input images based on a similarity measure. However, as the Siamese network is pre-trained offline, it experiences many difficulties in the adaptation to various online environments. Therefore, in this paper we propose to incorporate the histogram weighted centroid feature into the Siamese network method to enhance the accuracy of the object tracking. The proposed method uses both the histogram information and the weighted centroid location of the top 10 color regions to decide which of the proposed region should become the next predicted object region.

• The Journal of Korean Physical Therapy
• /
• v.18 no.3
• /
• pp.37-45
• /
• 2006

Purpose: This study analyzed changes in spinal neuron and action potential of motor unit depending on voluntary contraction on spinal neuron adaptation. Methods: It selected 80 university students in their twenties and divided into experimental groups of 25% MVIC (I), 50% MVIC (II), 75% MVIC (III) and 100% MVIC (IV) depending on maximum voluntary isometric contraction (MVIC) and performed isometric exercise of plantar flexor muscle to each experimental group with given contraction for 20 times. It measured Mmax, H/Mmax, Hmax latency, V/Mmax, V wave latency before and after exercise, compared method and volume of contraction. Results: H/Mmax ratio showed significant difference in comparison among groups (p<0.01) and there was difference in I and IV groups. V/Mmax ratio showed significant difference in comparison among experimental groups (p<0.05) and there was difference in I and IV groups. When voluntary contraction level was maximum, changes were greatest. However, no significantly difference was to Mmax, H latency and V wave latency. Conclusion: These results suggest that amplitude changes of voluntary contraction level, spinal neuron and supra-spinal neuron had a dose connection that the more contraction level, the better central activation seem to decrease highly for a short time.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.3
• /
• pp.27-38
• /
• 2000

This paper proposes how the recurrent neural network controller for a Khepera mobile robot with an obstacle avoiding ability can be determined by co-adaptation of the evolution and learning, The proposed co-adaptation scheme consists of two folds： a population of NN controllers are evolved by the genetic algorithm so that the degree of obstacle avoidance might be reduced through the global searching and each NN controller is trained by CRBP learning so that the running behavior is adapted to its outer environment through the local searching. Experimental results shows that the NN controller coadapted by evolution and learning outperforms its non-learning equivalent evolved by only genetic algorithm in both the ability of obstacle avoidance and the convergence speed reaching to the required running behavior.

• Journal of Power Electronics
• /
• v.13 no.4
• /
• pp.516-527
• /
• 2013

This paper presents a new overall system for state-of-available-power (SoAP) prediction for a lithium-ion battery pack. The essential part of this method is based on an adaptive network architecture which utilizes both fuzzy model (FIS) and artificial neural network (ANN) into the framework of adaptive neuro-fuzzy inference system (ANFIS). While battery aging proceeds, the system is capable of delivering accurate power prediction not only for room temperature, but also at lower temperatures at which power prediction is most challenging. Due to design property of ANN, the network parameters are adapted on-line to the current battery states (state-of-charge (SoC), state-of-health (SoH), temperature). SoC is required as an input parameter to SoAP module and high accuracy is crucial for a reliable on-line adaptation. Therefore, a reasonable way to determine the battery state variables is proposed applying a combination of several partly different algorithms. Among other SoC boundary estimation methods, robust extended Kalman filter (REKF) for recalibration of amp hour counters was implemented. ANFIS then achieves the SoAP estimation by means of time forward voltage prognosis (TFVP) before a power pulse occurs. The trade-off between computational cost of batch-learning and accuracy during on-line adaptation was optimized resulting in a real-time system with TFVP absolute error less than 1%. The verification was performed on a software-in-the-loop test bench setup using a 53 Ah lithium-ion cell.

• Journal of Korea Game Society
• /
• v.6 no.2
• /
• pp.13-22
• /
• 2006

As the development of games continues, the intelligence of NPC is becoming more and more important. Nowadays, the NPCs of MMORPGS are not only capable of simple actions like moving and attacking players, but also utilizing variety of skills and tactics as human-players do. This study suggests a method that grants characters used in RPG(Role-Playing Game) an ability of training and adaptation using Neural network and Genetic Algorithm. In this study, a simple game-play model is constructed to test how suggested intellect characters could train and adapt themselves to game rules and tactics. In the game-play model, three types of characters(Tanker, Dealer, Healer) are used. Intellect character group constructed by NN and GA, and trained by combats against enemy character group constructed by FSM. As the result of test, the proposed intellect characters group acquire an appropriate combat tactics by themselves according to their abilities and those of enemies, and adapt change of game rule.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.36 no.2
• /
• pp.125-130
• /
• 2012

The nematode Caenorhabditis elegans is a widely used model organism in biological research. Thanks to the availability of well-established knowledge about its neural connectivity, a wide range of studies have been attempted to uncover the relationship between behaviors and the responsible neurons. In our research, the adaptive behavior of C. elegans in solid environments with different surface rigidities is investigated, where the worm adapts to different mechanical stiffnesses by modulating its crawling waveform. The amplitude and wavelength of the crawling waveform decrease as the environment becomes more rigid. Interestingly, the mechanosensation-defective mutant shows different responses to the surface rigidity compared to those of the wild-type worm. To explain the adaptation process in mechanically different environments, we suggest a plausible neural circuit model.