• Journal of Advanced Research in Ocean Engineering
• /
• v.3 no.3
• /
• pp.136-148
• /
• 2017

The paper presents dual arm ROV manipulation using deep reinforcement learning. The purpose of this underwater manipulator is to investigate and excavate natural resources in ocean, finding lost aircraft blackboxes and for performing other extremely dangerous tasks without endangering humans. This research work emphasizes on a self-learning approach using Deep Reinforcement Learning (DRL). DRL technique allows ROV to learn the policy of performing manipulation task directly, from raw image data. Our proposed architecture maps the visual inputs (images) to control actions (output) and get reward after each action, which allows an agent to learn manipulation skill through trial and error method. We have trained our network in simulation. The raw images and rewards are directly provided by our simple Lua simulator. Our simulator achieve accuracy by considering underwater dynamic environmental conditions. Major goal of this research is to provide a smart self-learning way to achieve manipulation in highly dynamic underwater environment. The results showed that a dual robotic arm trained for a 3DOF movement successfully achieved target reaching task in a 2D space by considering real environmental factor.

• Journal of Practical Engineering Education
• /
• v.8 no.1
• /
• pp.63-68
• /
• 2016

As job mismatch intensifies in the employment market, the work based learning has emerged as a solution to this social problem. The so called Korean dual education system is expected to raise jobseekers equipped with knowledge and skills demanded by industry. This study intends to recommend an effective operational model for contracted education programs at higher institutions in order to support soft landing of Korean dual education system. Students of contracted education programs are adults graduated from vocational high schools and working for companies during weekdays. In consideration of the difficulties experienced by employee students, more flexible operation of academic affairs is required. To address these challenges, the recommended contracted education model includes such beneficial features as S-OJT courses, standard curriculum based on National Competency Standard, recognition of prior learning, flipped classroom, and tutoring system. In the last part, several directions for implementing the new contracted education model are discussed.

• Journal of the Korea Safety Management & Science
• /
• v.24 no.3
• /
• pp.61-74
• /
• 2022

In a rapidly changing environment, it is essential for companies to develop their capabilities for sustainable growth. To this end, informal learning at work plays an important role for small and medium-sized enterprises that are having difficulties in human resource development. If the GRIT of workers and the spiritual leadership of leaders are maintained at an appropriate level, the effect of informal learning will increase. This study derived the following results through a survey of workers participating in the Work-learning Dual System in Korea. First, experience and feedback among informal learning activities have a significant positive effect on competency development. Second, the higher the level of the GRIT of workers, the greater the magnitude of the positive effect of experience and feedback on competency development, showing a positive moderating effect. Third, in the case of workers with high GRIT, capacity development is strengthened as experience or feedback increases, and the higher the level of spiritual leadership, the greater the scope of reinforcement. On the other hand, it can be seen that in the case of workers with low GRIT, competency development weakens as experience or feedback increases, and the higher the level of spiritual leadership, the greater the extent of the weakening.

• Journal of Korea Multimedia Society
• /
• v.23 no.4
• /
• pp.525-538
• /
• 2020

A practical animal face classification system that classifies animals in image and video data is considered as a pivotal topic in machine learning. In this research, we are proposing a novel method of fully connected dual Deep Convolutional Neural Network (DCNN), which extracts and analyzes image features on a large scale. With the inclusion of the state of the art Batch Normalization layer and Exponential Linear Unit (ELU) layer, our proposed DCNN has gained the capability of analyzing a large amount of dataset as well as extracting more features than before. For this research, we have built our dataset containing ten thousand animal faces of ten animal classes and a dual DCNN. The significance of our network is that it has four sets of convolutional functions that work laterally with each other. We used a relatively small amount of batch size and a large number of iteration to mitigate overfitting during the training session. We have also used image augmentation to vary the shapes of the training images for the better learning process. The results demonstrate that, with an accuracy rate of 92.0%, the proposed DCNN outruns its counterparts while causing less computing costs.

• Journal of vocational education research
• /
• v.35 no.6
• /
• pp.113-141
• /
• 2016

The apprenticeship training is given as a solution of the youth unemployment problem. The apprenticeship is a system that companies hire and instruct trainees and is implemented in various countries. In Korea, there is a Korean work-learning dual system which is managed by the government rather than the company itself. This Study analyzed Germany, Switzerland and Austria's apprenticeship training system and compared common factors of apprenticeship system, a trainer in company and cost-benefit analysis. The result of the study can be shown as follows. First, Germany operates practical training, meanwhile Switzerland and Austria provide a pre-education course before students enter the upper-secondary VET course. Second, Germany and Switzerland give a single type apprenticeship training, whereas Austria runs a modularization system, which enables students to specify their skills and improves clarity. Third, social partners as well as the government, states, companies/schools actively participate in apprenticeship. Fourth, there is a suitable trainer's qualification system in all countries. Finally, Germany and Austria's companies cause larger cost than benefit in apprenticeship. On the other hand, Switzerland gets larger net-benefit than cost. We can give implications to the Korean work-learning dual system based on this study.

• Korean Journal of Comparative Education
• /
• v.27 no.1
• /
• pp.247-284
• /
• 2017

The purpose of this study is to analyze firm's motivation of participating and investing in apprenticeship in Germany and Korea, and to investigate institutional factors influencing firm's motivation. By comparing institutional factors of the two countries, it aims to drawing out policy implications for improving Korean apprenticeship. The main method for data collection was comprehensive literature review on international organizations, each countries' government and research institutes' policy materials, statistical data, research outputs and media resources related to each countries' apprenticeship. Considering whether firm's motivation for participating and investing in apprenticeship is production-oriented or investment-oriented, Germany is more inclined to investment motivation with firm's covering net cost during apprenticeship period. On the other hand, Korea is more inclined toward production orientation with firm's expectation of gaining net profit during the training period. Why is firm's training motivation different in these two countries? The author tried to find the reason from the difference of institutional factors of the countries by dividing institutional factors into 4 categories: context(tripartite relations, legal framework), input (flexibility of the system, government incentive), process(training contents, training duration, quality assurance), and output(completion/retention rate, apprentice's productivity). The key implication from the comparative analysis of institutional factors is that it is necessary to enforce companies to have "accountability" on the minimum critical elements, but also to ensure them to have "autonomy" on the rest of the elements.