• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 F
• /
• pp.2610-2612
• /
• 1999

In controlling vector of induction motor, PID controller is required much time as the expert should control manually a gain of controller according to plant or a change of circumstances. Accordingly, this paper has gotten a gain of PID controller used neural network by self-funning method in order to settle above problem. The neural network can describe an input/output features in spite of non-linear system which is hard to get mathematical model by controlling the strength of connection by learning. It has a strong character against a distortion and noise of input information, and is suitable modeling of diver-variable system which is composed of several input/output. This paper has represented the self-tunning method for gain of PID controller used neural network when using PID controller to control speed of induction motor, and has checked strong characters against distortion and noise of input information through simulation.

• E2M - 전기 전자와 첨단 소재
• /
• 제10권9호
• /
• pp.916-921
• /
• 1997

In this study Characteristics of 3-D underwater object recognition independent of translation using the self-made ultrasonic sensor fabricated with porous piezoelectric resonator and presented. The sensor was satisfied with requirement of ultrasonic sensor. The recognition rates for the training data and the testing data are 97.45 and 91.25[%] respectively using the self-made ultrasonic sensor and SCL(Simple Competitive Learning) neural network. According to the experimental results It is believed that the self-made ultrasonic sensor can be applied as sensor of SONAR system.

• 제어로봇시스템학회논문지
• /
• 제4권2호
• /
• pp.209-214
• /
• 1998

In this paper, we are dealing with the problem of controlling unknown nonlinear dynamical system by using neural networks. A novel error self-recurrent(ESR) neural model is presented to perform black-box identification. Through the various outcome of the experiment, a new neural network is seen to be considerably faster than the BP algorithm and has advantages of being less affected by poor initial weights and learning rate. These characteristics make it flexible to design the controller in real-time based on neural networks model. In addition, we design an adaptive PID controller that Keyser suggested by using ESR neural networks, and present a method on the implementation of adaptive controller based on neural network for practical applications. We obtained good results in the case of robot manipulator experiment.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.29.1-29
• /
• 2001

Neuro-fuzzy systems are multi-layered connectionist networks that realize the elements and functions of traditional fuzzy logic control/decision systems. A trained neuro-fuzzy system is isomorphic to a fuzzy logic system, and fuzzy IF-THEN rule knowledge can be explicitly extracted from the network. This talk presents a brief introduction to self-adaptive neuro-fuzzy systems and addresses some recent research results and applications. Most of the existing neuro-fuzzy systems exhibit several major drawbacks that lead to performance degradation. These drawbacks are the curse of dimensionality (i.e., fuzzy rule explosion), inability to re-structure their internal nodes in a changing environment, and their lack of ability to extract knowledge from a given set of training data. This talk focuses on our investigation of network architectures, self-adaptation algorithms, and efficient learning algorithms that will enable existing neuro-fuzzy systems to self-adapt themselves in an unstructured and uncertain environment.

• 제어로봇시스템학회논문지
• /
• 제15권2호
• /
• pp.192-197
• /
• 2009

In this paper, we proposed the obstacle avoidance system using a single camera image and LM(Levenberg-Marquart) neural network fuzzy controller. According to a robot technology adapt to various fields of industry and public, the robot has to move using self-navigation and obstacle avoidance algorithms. When the robot moves to target point, obstacle avoidance is must-have technology. So in this paper, we present the algorithm that avoidance method based on fuzzy controller by sensing data and image information from a camera and using the LM neural network to minimize the moving error. And then to verify the system performance of the simulation test.

• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.687-701
• /
• 2022

Inevitable response loss under complex operational conditions significantly affects the integrity and quality of measured data, leading the structural health monitoring (SHM) ineffective. To remedy the impact of data loss, a common way is to transfer the recorded response of available measure point to where the data loss occurred by establishing the response mapping from measured data. However, the current research has yet addressed the structural condition changes afterward and response mapping learning from a small sample. So, this paper proposes a novel data driven structural response reconstruction method based on a sophisticated designed generating adversarial network (UAGAN). Advanced deep learning techniques including U-shaped dense blocks, self-attention and a customized loss function are specialized and embedded in UAGAN to improve the universal and representative features extraction and generalized responses mapping establishment. In numerical validation, UAGAN efficiently and accurately captures the distinguished features of structural response from only 40 training samples of the intact structure. Besides, the established response mapping is universal, which effectively reconstructs responses of the structure suffered up to 10% random stiffness reduction or structural damage. In the experimental validation, UAGAN is trained with ambient response and applied to reconstruct response measured under earthquake. The reconstruction losses of response in the time and frequency domains reached 16% and 17%, that is better than the previous research, demonstrating the leading performance of the sophisticated designed network. In addition, the identified modal parameters from reconstructed and the corresponding true responses are highly consistent indicates that the proposed UAGAN is very potential to be applied to practical civil engineering.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2024년도 제69차 동계학술대회논문집 32권1호
• /
• pp.45-46
• /
• 2024

본 논문에서는 Self-Attention 기반 딥러닝 기법인 Dense Prediction Transformer(DPT) 모델을 MVTec Anomaly Detection(MVTec AD) 데이터셋에 적용하여 실제 산업 제품 이미지 내 이상 부분을 분할하는 연구를 진행하였다. DPT 모델의 적용을 통해 기존 Convolutional Neural Network(CNN) 기반 이상 탐지기법의 한계점인 지역적 Feature 추출 및 고정된 수용영역으로 인한 문제를 개선하였으며, 실제 산업 제품 데이터에서의 이상 분할 시 기존 주력 기법인 U-Net의 구조를 적용한 최고 성능의 모델보다 1.14%만큼의 성능 향상을 보임에 따라 Self-Attention 기반 딥러닝 기법의 적용이 산업 제품 이상 분할에 효과적임을 입증하였다.

• 한국인공지능학회지
• /
• 제6권2호
• /
• pp.23-27
• /
• 2018

Korea has a high proportion of self-employment. Many of them start the food business since it does not require high-techs and it is possible to start the business relatively easily compared to many others in business categories. However, the closure rate of the business is also high due to excessive competition and market saturation. Cafés and restaurants are examples of food business where the business analysis is highly important. However, for most of the people who want to start their own business, it is difficult to conduct systematic business analysis such as trade area analysis or to find information for business analysis. Therefore, in this paper, we predicted business status with simple information using Microsoft Azure Machine Learning Studio program. Experimental results showed higher performance than the number of attributes, and it is expected that this artificial intelligence model will be helpful to those who are self-employed because it can easily predict the business status. The results showed that the overall accuracy was over 60 % and the performance was high compared to the number of attributes. If this model is used, those who prepare for self-employment who are not experts in the business analysis will be able to predict the business status of stores in Seoul with simple attributes.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 지능정보 및 응용 학술대회
• /
• pp.142-145
• /
• 2008

본 논문에서는 self-organizing map(SOM)과 grassfire 기법을 이용한 계산 효율적인 컬러 영상 분할 방법을 제안한다. SOM에서 출력 뉴런 수를 축소하고 학습에 사용하는 입력 데이터를 줄임으로써 실행 시간을 단축 시켰다. 입력 영상을 CIE $L^*u^*v^*$ 컬러 공간으로 변환하고 3개의 입력 뉴런과 $4{\times}4$ 또는 $3{\times}3$ 출력 뉴런 구조의 SOM을 이용해 학습한다. 학습 완료 후 입력 영상의 픽셀에 대응하는 출력 값을 구하고 grassfire 기법을 이용해 지역적으로 인접하고 출력 값이 동일한 픽셀들을 하나의 영역으로 결합한다. 다양한 영상을 이용한 실험을 통해 제안한 방법이 컬러 영상 분할에서 기존의 방법에 비해 좋은 결과를 얻을 수 있음을 확인하였다.

• 스마트미디어저널
• /
• 제12권1호
• /
• pp.32-46
• /
• 2023

Natural Language Processing (NLP) has grown tremendously in recent years. Typically, bilingual, and multilingual translation models have been deployed widely in machine translation and gained vast attention from the research community. On the contrary, few studies have focused on translating between spoken and sign languages, especially non-English languages. Prior works on Sign Language Translation (SLT) have shown that a mid-level sign gloss representation enhances translation performance. Therefore, this study presents a new large-scale Korean sign language dataset, the Museum-Commentary Korean Sign Gloss (MCKSG) dataset, including 3828 pairs of Korean sentences and their corresponding sign glosses used in Museum-Commentary contexts. In addition, we propose a translation framework based on self-supervised learning, where the pretext task is a text-to-text from a Korean sentence to its back-translation versions, then the pre-trained network will be fine-tuned on the MCKSG dataset. Using self-supervised learning help to overcome the drawback of a shortage of sign language data. Through experimental results, our proposed model outperforms a baseline BERT model by 6.22%.