• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.10
• /
• pp.365-370
• /
• 2014

Although Internet of Things already became a new paradigm on service on network, we should pay more effort for studying about its testing method, since humans, things and environments in IoT are connected to each other without any restrictions. Earlier researches based on emulators showed that such virtual devices on emulators had unavoidable gap between them and real things. Furthermore, growth of connection complexity between the devices and loosing of restrictions make the gap wider. Accordingly, in this paper, we suppose a method of test coverage measurement based on BitTorrent for IoT environment. It has cooperation features among homogeneous devices with avoiding the overlapping on each part of whole test process.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.3
• /
• pp.272-280
• /
• 2021

We research and develop a SW-based virtual testing product that can be commonly used in the design/development of non-communication EW systems before the production of physical test products. Through this study, we have developed M&S technology to improve the accuracy of EW weapon system analysis/design and to verify and predict the performance of EW equipment, and to develop proven engineering module models and model base systems. It proposes a technology to build an EW M&S framework that can flexibly link/integrate various engineering/engage-level EW heterogeneous M&S systems.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.121-129
• /
• 2022

In this paper, the virtual environment using rack server type HPC and 3U VPX server type HPC was applied and tested to the basic functions of the Jangbogo-III class submarine combat system developed for the first time in Korea. Based on this test results, the possibility of applying virtualization to the domestic submarine combat system to be developed in the future is confirmed. Existing studies have been limited to deriving applicable virtualization solutions through simple performance analysis of virtualization solutions or applying virtualization to some functions of the surface ship combat system, but in this paper, the application of virtualization is expanded to the submarine combat system through testing.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.264-272
• /
• 2022

Human-robot cooperative tasks are increasingly required in our daily life with the development of robotics and artificial intelligence technology. Interactive reinforcement learning strategies suggest that robots learn task by receiving feedback from an experienced human trainer during a training process. However, most of the previous studies on Interactive reinforcement learning have required an extra feedback input device such as a mouse or keyboard in addition to robot itself, and the scenario where a robot can interactively learn a task with human have been also limited to virtual environment. To solve these limitations, this paper studies training strategies of robot that learn table balancing tasks interactively using deep reinforcement learning with human's facial expression feedback. In the proposed system, the robot learns a cooperative table balancing task using Deep Q-Network (DQN), which is a deep reinforcement learning technique, with human facial emotion expression feedback. As a result of the experiment, the proposed system achieved a high optimal policy convergence rate of up to 83.3% in training and successful assumption rate of up to 91.6% in testing, showing improved performance compared to the model without human facial expression feedback.

• Steel and Composite Structures
• /
• v.49 no.4
• /
• pp.393-405
• /
• 2023

In this paper, the energy absorption capability of a novel cruciform composite lattice structure was evaluated through the simulation of compression tests. For this purpose, several test samples of Polylactic acid cellular reinforced with continuous glass fibers were prepared for compression testing using the additive manufacturing method of material extrusion. Using a conventional path design for material extrusion, multiple debonding is probable to be occurred at the joint regions of adjacent cells. Therefore, an innovative printing path design was proposed for the cruciform lattice structure. Afterwards, quasistatic compression tests were performed to evaluate the energy absorption behaviour of this structure. A finite element model based on local material property degradation was then developed to verify the experimental test and extend the virtual test method. Accordingly, different combinations of unit cells' dimensions using the design of the experiment were numerically proposed to obtain the optimal configuration in terms of the total absorbed energy. Having brilliant energy absorption properties, the studied cruciform lattice with its optimized unit cell dimensions can be used as an energy absorber in crashworthiness applications. Finally, a cellular structure will be suitable with optimal behavior in crush load efficiency and high energy absorption.

• Journal of Information Technology Applications and Management
• /
• v.31 no.3
• /
• pp.149-163
• /
• 2024

As conversational AI systems such as ChatGPT have become more advanced, researchers are exploring ways to use them in education. However, we need effective ways to evaluate these systems before allowing them to help teach students. This study proposes a detailed framework for testing conversational AI across three important criteria as follow. First, specialized benchmarks that measure skills include giving clear explanations, adapting to context during long dialogues, and maintaining a consistent teaching personality. Second, adaptive standards check whether the systems meet the ethical requirements of privacy, fairness, and transparency. These standards are regularly updated to match societal expectations. Lastly, evaluations were conducted from three perspectives: technical accuracy on test datasets, performance during simulations with groups of virtual students, and feedback from real students and teachers using the system. This framework provides a robust methodology for identifying strengths and weaknesses of conversational AI before its deployment in schools. It emphasizes assessments tailored to the critical qualities of dialogic intelligence, user-centric metrics capturing real-world impact, and ethical alignment through participatory design. Responsible innovation by AI assistants requires evidence that they can enhance accessible, engaging, and personalized education without disrupting teaching effectiveness or student agency.

• Journal of Internet Computing and Services
• /
• v.15 no.5
• /
• pp.107-121
• /
• 2014

Distributed systems are collection of physically distributed computers linked by a network. General use of wired/wireless Internet enables users to make use of distributed service anytime and anywhere. The explosive growth of distributed services strongly requires functional verification of services as well as verification of non-functional elements such as service quality. In order to verify distributed services it is necessary to build a test environment for distributed systems. Because, however, distributed systems are composed of physically distributed nodes, efforts to construct a test environment are required more than those in a test environment for a monolithic system. In this paper we propose a test framework to verify functional and non-functional features of distributed systems. The suggested framework automatically generates test cases through the message sequence charts, and includes a test driver composed of the virtual nodes which can simulate the physically distributed nodes. The test result can be checked easily through the various graphs and the graphical user interface (GUI). The test framework can reduce testing efforts for a distributed system and can enhance the reliability of the system.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.235-249
• /
• 2020

The Ministry of Land, Infrastructure and Transport established safety standards for Level 3 autonomous vehicles for the first time in the world in December 2019, and specified the safety standards for conditional autonomous driving systems. Accordingly, it is necessary to analyze the influence of various driving environments on take-over. In this study, using a driving simulator, we investigated how traffic conditions and weather conditions affect take-over time and stabilization time. The experimental procedure was conducted in the order of preliminary training, practice driving, and test driving, and the test driving was conducted by dividing into a traffic density and geometry experiment and a weather environment experiment. As a result of the experiment, it was analyzed that the traffic volume and weather environment did not affect the take-over time and take-over stabilization time, and only the curve radius affects take-over stabilization time.

• Composites Research
• /
• v.16 no.5
• /
• pp.45-53
• /
• 2003

The experimental and numerical investigations on the failure characteristics of the secondary bonded composite single-lap joints were performed. The initiations and growths of cracks were observed using CCD camera and acoustic emission sensor during the tension tests of the joint specimens. The structural behaviors of the specimens were predicted by the geometric nonlinear two-dimensional finite element analysis. The three types of observed initial cracks were included in each finite element models and the strain energy release rates of each specimen models were calculated by VCCT(Virtual Crack Closure Technique) technique. The tension tests showed that the initial cracks occurred in the 60∼90% of final failure loads and the major failure modes of the specimens were adhesive failure and the delamination between the 1st and 2nd ply of laminate. The specimens with the thicker bondline had earlier crack initiation loads but higher crack propagation resistance and eventually better loading capability. The delaminations were mostly observed in the thicker bondline specimens. The mode I values of calculated strain energy release rates were higher than the mode II values in the all specimen models considering the three types of initial cracks. The mode I and total strain energy release rates were calculated as higher values in the order of initial crack in the edge interface, comer interface and delamination between the plies of laminate.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.4
• /
• pp.31-43
• /
• 2019

Take-overs occur in autonomous vehicles at levels 3 and 4 based on SAE. For safe take-over, it is necessary to set the time required for diverse drivers to complete take-over in various road conditions. In this study, take-over time and stabilization characteristics were measured to secure safety of take-over in autonomous vehicle. To this end, a virtual driving simulator was used to set up situations similar to those on real expressways. Fifty drivers with various sexes and ages participated in the experiment where changes in traffic volume and geometry were applied to measure change in takeover time and stabilization characteristics according to various road conditions. Experimental results show that the average take-over time was 2.3 seconds and the standard deviation was 0.1 second. As a result of analysis of stabilization characteristics, there was no difference in take-over stabilization time due to the difference of traffic volume, and there was a significant difference by curvature changes.