• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.3
• /
• pp.293-301
• /
• 2021

LTV(Light Tactical Vehicle) operating in our military requires higher levels of performance and durability to withstand harsher conditions than general vehicles, as they must travel on both rough-train and off-road as well as on public roads. Recently, LTV development is demanded a variety of test evaluations in order to satisfy ROC (Required Operational Capability) by the military requirement. However, there is no informations of driving test course for satisfying the durability performance of Korean tactical vehicle. Therefore, this study aims to provide basic data to establish reliable drive test conditions by analyzing the main maneuvering test site at the domestic and foreign country in order to select the representative drive course. These studies will provide a more scientific and systematic evaluation solution for the development of tactical vehicles, and can be effectively used to establish a certified system for military vehicle test evaluation in the future

• Journal of Engineering Education Research
• /
• v.24 no.3
• /
• pp.50-57
• /
• 2021

Although the mathematics education process in AI education is a very important issue, little cases are reported in developing effective methods on AI and mathematics education at the university level. The universities cover all fields of mathematics in their curriculums, but they lack in connecting and applying the math knowledge to AI in an efficient manner. Students are hardly interested in taking many math courses and it gets worse for the students in humanities, social sciences and arts. But university education is very slow in adapting to rapidly changing new technologies in the real world. AI is a technology that is changing the paradigm of the century, so every one should be familiar with this technology but it requires fundamental math knowledge. It is not fair for the students to study all math subjects and ride on the AI train. We recognize that three key elements, SW knowledge, mathematical knowledge, and domain knowledge, are required in applying AI technology to the real world problems. This study proposes a modular approach of studying mathematics knowledge while connecting the math to different domain problems using AI techniques. We also show a modular curriculum that is developed for using math for AI-driven autonomous driving.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.1-8
• /
• 2023

The artificial intelligence of robot is the weakness of digital intelligence of a person who able to train, self-realize and to develop competences, creative, professional and behavioral skills. A new methodology proposed for managing robots inside the mines using an electronic system designed for driving robots to injured people in seas, mines or wells who can not be reached by human force. This paper also explains the concept of managing and remote-controlling the process of searching and helping the injured. The user controls the robot through an application that receives all the reports that the robot sends from the injured person. The robot's tasks are to take a sample of the blood of the injured person, examine it, and measure the percentage of oxygen underground and send it to the user who directs the robot to pump a specific percentage of oxygen to the injured person. The user can also communicate with the person The patient and determine his condition through the camera connected to the robot equipped with headphones to communicate with the injured and the user can direct the camera of the robot and take x-rays from the injured.

• International conference on construction engineering and project management
• /
• 2020.12a
• /
• pp.122-128
• /
• 2020

This research investigates how a Virtual Reality (VR)-based simulation could be used to train safe operation skills for forklift operators. Forklift operation is categorized as high-risk work by many occupational health and safety regulators and authorities due to high injury and fatality rates involved with forklifts. Therefore, many safety guidelines have been developed for forklift operators. Typically, forklift operation safety training is delivered based on instructional texts or videos, which have limitations in influencing people's safety behavior. Against this background, we propose a VR-based forklift simulator that can enable safe operation skills training through a feedback system. The training program consists of several modules to teach how to perform the basic tasks of forklift operation, such as driving, loading and unloading, following the safety guidelines. The system provides instantaneous instructions and feedback regarding safe operation. This training system is based on the model of "learning-by-doing". The user can repeat the training modules as many times as necessary before being able to perform the given task without violating any safety guidelines. The last training module tests the user's acquisition of all safety skills required. The user feedback from several demonstration sessions showed the potential usefulness of the proposed training system.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.822-835
• /
• 2023

Recent semantic segmentation frameworks usually combine low-level and high-level context information to achieve improved performance. In addition, postlevel context information is also considered. In this study, we present a Context ReFinement Network (CRFNet) and its training method to improve the semantic predictions of segmentation models of the encoder-decoder structure. Our study is based on postprocessing, which directly considers the relationship between spatially neighboring pixels of a label map, such as Markov and conditional random fields. CRFNet comprises two modules: a refiner and a combiner that, respectively, refine the context information from the output features of the conventional semantic segmentation network model and combine the refined features with the intermediate features from the decoding process of the segmentation model to produce the final output. To train CRFNet to refine the semantic predictions more accurately, we proposed a sequential training scheme. Using various backbone networks (ENet, ERFNet, and HyperSeg), we extensively evaluated our model on three large-scale, real-world datasets to demonstrate the effectiveness of our approach.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.157-165
• /
• 2020

Aiming at the development of neural network and self-driving data set, it is also an idea to improve the performance of network model to detect moving objects by dividing the data set. In Darknet network framework, the YOLOv4 (You Only Look Once v4) network model was used to train and test Udacity data set. According to 7 proportions of the Udacity data set, it was divided into three subsets including training set, validation set and test set. K-means++ algorithm was used to conduct dimensional clustering of object boxes in 7 groups. By adjusting the super parameters of YOLOv4 network for training, Optimal model parameters for 7 groups were obtained respectively. These model parameters were used to detect and compare 7 test sets respectively. The experimental results showed that YOLOv4 can effectively detect the large, medium and small moving objects represented by Truck, Car and Pedestrian in the Udacity data set. When the ratio of training set, validation set and test set is 7:1.5:1.5, the optimal model parameters of the YOLOv4 have highest detection performance. The values show mAP50 reaching 80.89%, mAP75 reaching 47.08%, and the detection speed reaching 10.56 FPS.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.385-391
• /
• 2023

As the use of electric vehicles has increased to minimize carbon emissions, the analyzing the state and performance of lithium-ion batteries that is instrumental in electric vehicles have been important. Comprehensive analysis using not only the voltage, current and temperature of the battery pack, which can affect the condition and performance of the battery, but also the driving data and charging pattern data of the electric vehicle is required. Therefore, a thorough analysis is imperative, utilizing electric vehicle operation data, charging pattern data, as well as battery pack voltage, current, and temperature data, which collectively influence the condition and performance of the battery. Therefore, collection and preprocessing of battery data collected from electric vehicles, collection and preprocessing of data on driver driving habits in addition to simple battery data, detailed design and modification of artificial intelligence algorithm based on the analyzed influencing factors, and A battery analysis and evaluation model was designed. In this paper, we gathered operational data and battery data from real-time electric buses. These data sets were then utilized to train a Random Forest algorithm. Furthermore, a comprehensive assessment of battery status, operation, and charging patterns was conducted using the explainable Artificial Intelligence (XAI) algorithm. The study identified crucial influencing factors on battery status, including rapid acceleration, rapid deceleration, sudden stops in driving patterns, the number of drives per day in the charging and discharging pattern, daily accumulated Depth of Discharge (DOD), cell voltage differences during discharge, maximum cell temperature, and minimum cell temperature. These factors were confirmed to significantly impact the battery condition. Based on the identified influencing factors, a battery analysis and evaluation model was designed and assessed using the Random Forest algorithm. The results contribute to the understanding of battery health and lay the foundation for effective battery management in electric vehicles.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.6
• /
• pp.120-129
• /
• 2023

To evaluate the vertical loads on railway bridges, conventional load tests are typically conducted. However, these tests often entail significant costs and procedural challenges. Railway conditions involve nearly identical load profiles due to standardized rail systems, which may appear straightforward in terms of load conditions. Nevertheless, this study aims to validate load tests conducted under operational train conditions by comparing the results with those obtained from conventional load tests. Additionally, static and dynamic structural behaviors are extracted from the measurement data for evaluation. To ensure the reliability of load testing, this research demonstrates feasibility through comparisons of existing measurement data with sensor attachment locations, train speeds, responses between different rail lines, tendency analysis, selection of impact coefficients, and analysis of natural frequencies. This study applies to the Dongho Railway Bridge and verifies the applicability of the proposed method. Ten operational trains and 44 sensors were deployed on the bridge to measure deformations and deflections during load test intervals, which were then compared with theoretical values. The analysis results indicate good symmetry and overlap of loads, as well as a favorable comparison between static and dynamic load test results. The maximum measured impact coefficient (0.092) was found to be lower than the theoretical impact coefficient (0.327), and the impact influence from live loads was deemed acceptable. The measured natural frequencies approximated the theoretical values, with an average of 2.393Hz compared to the calculated value of 2.415Hz. Based on these results, this paper demonstrates that for evaluating vertical loads, it is possible to measure deformations and deflections of truss railway bridges through load tests under operational train conditions without traffic control, enabling the calculation of response factors for stress adjustments.

• Journal of Korean Society of Transportation
• /
• v.34 no.5
• /
• pp.465-475
• /
• 2016

Since inaccurate demand estimation for recent urban rail construction may result in financial burden to cities, precise prediction for operating cost as well as construction costs is necessary to avoid or reduce budget loss of the local or central government. The operating cost is directly related to the public fare and affect a policy to determine the rate system. Therefore, there is a pressing need to develop an estimating model for reliable operating cost of urban railway. This study introduces a new model to estimate the operating cost with new variables. It provides a better prediction in accuracy and reliability compared to the existing model, considering the feature of urban railway. For verification of our model, railway operation data from a few cities for the last five years were comprehensively examined to determine variables that affect the operating cost. The operating cost was estimated in a dummy regression model using five independent variables, which were average distance between stations, daily trains distance, total passenger capacity of a train in a train, driving mode(manned/unmanned), and investment type(financial/private).

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.1860-1871
• /
• 2011

Presently, The Regular Block System(Automatic, Interlocking block system) is usually used during the operation of block section. However, In case that the regular block system fails because of the failure of the fixed signals and block equipment or in case of the unexpected emergent situation which should drive on the single-track due to the accidents in the double-track section or the repair work of the one-track, the Substitute Block System to make use of the driving permission license(mapping ticket, mapping paper) is used. In case of the operation of the opposite line and the temporary one-track, the safety gets worse and the SBS may cause the fatal accidents such as a head-on & a rear-end collision. Also, the unmanned railroad stations has recently increased owing to the effective operation of the stations, for it is difficult to execute the SBS in their absence. As a result, the increase of the operation time made the train delayed. Being on the rise of these problems, in this study, we analyzed the problems and difficulties of the SBS on the single line which is lacking stability and safety and on the sections combined between maned and unmaned railroad stations. And we proposed the method to improve the existing drive permission license used for 50 years into the brand-new one with state-of-the art technology and scientific way. In the era of the 21th century, Carrying out the new SBS equipped with stability and safety, we will contribute to the effective operation of trains and the satisfaction of our customers in the future.