• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.991-997
• /
• 2020

강화학습은 주로 순차적인 의사 결정 문제에 적용되어 왔다. 특히 최근에는 신경망과 결합한 형태로 기존에는 해결하지 못한 분야에서도 성공적인 결과를 내고 있다. 하지만 신경망을 이용하는 강화학습은 현장에서 즉각적으로 사용하기엔 너무 복잡하다는 단점이 있다. 본 논문에서는 학습이 쉬운 강화학습 알고리즘 중 하나인 Q-learning을 이용하여 이동 로봇의 경로를 생성하는 알고리즘을 구현하였다. Q-table을 미리 만드는 방식의 Q-learning은 명확한 한계를 가지기 때문에 실시간으로 Q-table을 업데이트하는 실시간 Q-learning을 사용하였다. 탐험 전략을 조정하여 실시간 Q-learning에 필요한 학습 속도를 얻을 수 있었다. 마지막으로 실시간 Q-learning과 DQN의 성능을 비교하였다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집B
• /
• pp.724-729
• /
• 2001

This paper discusses the composition of the theory of reinforcement learning, which is applied in real-time learning, and evolutionary strategy, which proves its the superiority in the finding of the optimal solution at the off-line learning method. The individuals are reduced in order to learn the evolutionary strategy in real-time, and new method that guarantee the convergence of evolutionary mutations are proposed. It possible to control the control object varied as time changes. As the state value of the control object is generated, applied evolutionary strategy each sampling time because the learning process of an estimation, selection, mutation in real-time. These algorithms can be applied, the people who do not have knowledge about the technical tuning of dynamic systems could design the controller or problems in which the characteristics of the system dynamics are slightly varied as time changes. In the future, studies are needed on the proof of the theory through experiments and the characteristic considerations of the robustness against the outside disturbances.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집B
• /
• pp.105-109
• /
• 2001

This paper discuss the real-time self-tuning learning control based on evolutionary computation, which proves its the superiority in the finding of the optimal solution at the off-line learning method. The individuals are reduced in order to learn the evolutionary strategy in real-time, and new method that guarantee the convergence of evolutionary mutations are proposed. It possible to control the control object varied as time changes. As the state value of the control object is generated, applied evolutionary strategy each sampling time because the learning process of an estimation, selection, mutation in real-time. These algorithms can be applied, the people who do not have knowledge about the technical tuning of dynamic systems could design the controller or problems in which the characteristics of the system dynamics are slightly varied as time changes.

• 대한기계학회논문집A
• /
• 제27권9호
• /
• pp.1463-1468
• /
• 2003

This paper presented the real-time self-tuning learning control based on evolutionary computation, which proves its superiority in finding of the optimal solution at the off-line learning method. The individuals of the populations are reduced in order to learn the evolutionary strategy in real-time, and new method that guarantee the convergence of evolutionary mutations is proposed. It is possible to control the control object slightly varied as time changes. As the state value of the control object is generated, evolutionary strategy is applied each sampling time because the learning process of an estimation, selection, mutation is done in real-time. These algorithms can be applied; the people who do not have knowledge about the technical tuning of dynamic systems could design the controller or problems in which the characteristics of the system dynamics are slightly varied as time changes.

• 한국위성정보통신학회논문지
• /
• 제12권3호
• /
• pp.69-73
• /
• 2017

본 논문에서는 기계학습 기반의 실시간 이미지 인식 알고리즘을 개발하고 개발한 알고리즘의 성능을 테스트 하였다. 실시간 이미지 인식 알고리즘은 기계 학습된 이미지 데이터를 바탕으로 실시간으로 입력되는 이미지를 인식한다. 개발한 실시간 이미지 인식 알고리즘의 성능을 테스트하기 위해 자율주행 자동차 분야에 적용해보았고 이를 통해 개발한 실시간 이미지 인식 알고리즘의 성능을 확인해보았다.

• 4차산업연구
• /
• 제3권1호
• /
• pp.13-20
• /
• 2023

Purpose - By designing a PEF(Personalized Education Feedback) system for real-time prediction of learning achievement and motivation through real-time EEG analysis of learners, this system provides some modules of a personalized adaptive learning system. By applying these modules to e-learning and offline learning, they motivate learners and improve the quality of learning progress and effective learning outcomes can be achieved for immersive self-directed learning Research design, data, and methodology - EEG data were collected simultaneously as the English test was given to the experimenters, and the correlation between the correct answer result and the EEG data was learned with a machine learning algorithm and the predictive model was evaluated.. Result - In model performance evaluation, both artificial neural networks(ANNs) and support vector machines(SVMs) showed high accuracy of more than 91%. Conclusion - This research provides some modules of personalized adaptive learning systems that can more efficiently complete by designing a PEF system for real-time learning achievement prediction and learning motivation through an adaptive learning system based on real-time EEG analysis of learners. The implication of this initial research is to verify hypothetical situations for the development of an adaptive learning system through EEG analysis-based learning achievement prediction.

• 전기학회논문지P
• /
• 제57권2호
• /
• pp.109-114
• /
• 2008

In this paper, a dynamic learning controller for robot manipulator is implemented using real-time operating system with capabilities of multitasking, intertask communication and synchronization, event-driven, priority-driven scheduling, real-time clock control, etc. The controller hardware system with VME bus and related devices is developed and applied to implement a dynamic learning control scheme for robot manipulator. Real-time performance of the proposed dynamic learning controller is tested and evaluated for tracking of the desired trajectory and compared with the conventional servo controller.

• Corrosion Science and Technology
• /
• 제18권6호
• /
• pp.253-257
• /
• 2019

A deep learning based real-time painting surface inspection algorithm is proposed herein, designed for developing an autonomous inspection drone. The painting surface inspection is usually conducted manually. However, the manual inspection has a limitation in obtaining accurate data for correct judgement on the surface because of human error and deviation of individual inspection experiences. The best method to replace manual surface inspection is the vision-based inspection method with a camera, using various image processing algorithms. Nevertheless, the visual inspection is difficult to apply to surface inspection due to diverse appearances of material, hue, and lightning effects. To overcome technical limitations, a deep learning-based pattern recognition algorithm is proposed, which is specialized for painting surface inspections. The proposed algorithm functions in real time on the embedded board mounted on an autonomous inspection drone. The inspection results data are stored in the database and used for training the deep learning algorithm to improve performance. The various experiments for pre-inspection of painting processes are performed to verify real-time performance of the proposed deep learning algorithm.

• 대한임베디드공학회논문지
• /
• 제19권3호
• /
• pp.151-158
• /
• 2024

Although on-device artificial intelligence (AI) has gained attention to diagnosing machine faults in real time, most previous studies did not consider the model retraining and redeployment processes that must be performed in real-world industrial environments. Our study addresses this challenge by proposing an on-device AI-based real-time machine fault diagnosis system that utilizes continual learning. Our proposed system includes a lightweight convolutional neural network (CNN) model, a continual learning algorithm, and a real-time monitoring service. First, we developed a lightweight 1D CNN model to reduce the cost of model deployment and enable real-time inference on the target edge device with limited computing resources. We then compared the performance of five continual learning algorithms with three public bearing fault datasets and selected the most effective algorithm for our system. Finally, we implemented a real-time monitoring service using an open-source data visualization framework. In the performance comparison results between continual learning algorithms, we found that the replay-based algorithms outperformed the regularization-based algorithms, and the experience replay (ER) algorithm had the best diagnostic accuracy. We further tuned the number and length of data samples used for a memory buffer of the ER algorithm to maximize its performance. We confirmed that the performance of the ER algorithm becomes higher when a longer data length is used. Consequently, the proposed system showed an accuracy of 98.7%, while only 16.5% of the previous data was stored in memory buffer. Our lightweight CNN model was also able to diagnose a fault type of one data sample within 3.76 ms on the Raspberry Pi 4B device.

• 부동산연구
• /
• 제27권3호
• /
• pp.71-86
• /
• 2017

본 연구의 목적은 딥 러닝 방법을 부동산가격지수 예측에 적용해보고, 기존의 시계열분석 방법과의 비교를 통해 부동산 시장 예측의 새로운 방법으로서 활용가능성을 확인하는 것이다. 딥 러닝(deep learning)방법인 DNN(Deep Neural Networks)모형 및 LSTM(Long Shot Term Memory networks)모형과 시계열분석 방법인 ARIMA(autoregressive integrated moving average)모형을 이용하여 여러 가지 부동산가격지수에 대한 예측을 시도하였다. 연구결과 첫째, 딥 러닝 방법의 예측력이 시계열분석 방법보다 우수한 것으로 나타났다. 둘째, 딥 러닝 방법 중에서는 DNN모형의 예측력이 LSTM모형의 예측력보다 우수하나 그 정도는 미미한 수준인 것으로 나타났다. 셋째, 딥 러닝 방법과 ARIMA모형은 부동산 가격지수(real estate price index) 중 아파트 실거래가격지수(housing sales price index)에 대한 예측력이 가장 부족한 것으로 나타났다. 향후 딥 러닝 방법을 활용함으로써 부동산 시장에 대한 예측의 정확성을 제고할 수 있을 것으로 기대된다.