• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.3955-3971
• /
• 2020

Due to the significant increase in the use of autonomous car technology, it is essential to integrate this technology with high-precision digital map data containing more precise and accurate roadway information, as compared to existing conventional map resources, to ensure the safety of self-driving operations. While existing map technologies may assist vehicles in identifying their locations via Global Positioning System, it is however difficult to update the environmental changes of roadways in these maps. Roadway vision algorithms can be useful for building autonomous vehicles that can avoid accidents and detect real-time location changes. We incorporate a hybrid architectural design that combines unsupervised classification of vision data with supervised joint fusion classification to achieve a better noise-resistant algorithm. We identify, via a deep learning approach, an intelligent hybrid fusion algorithm for fusing multimodal vision feature data for roadway classifications and characterize its improvement in accuracy over unsupervised identifications using image processing and supervised vision classifiers. We analyzed over 93,000 vision frame data collected from a test vehicle in real roadways. The performance indicators of the proposed hybrid fusion algorithm are successfully evaluated for the generation of roadway digital maps for autonomous vehicles, with a recall of 0.94, precision of 0.96, and accuracy of 0.92.

• Journal of Digital Convergence
• /
• v.19 no.7
• /
• pp.227-236
• /
• 2021

This study aims to develop an AI education program for secondary school students to help understand AI and to provide an experience of solving real-life problems by using AI, and to analyze the effectiveness of education. The education program based on the AI education system for K-12 developed in the previous study was composed of a total of 12 lessons by selecting the self-driving cars, which is emerging as a recent issue among real life problems, as the main topic. Classes were conducted for secondary school students who had experience in software education, and the effectiveness of education and class satisfaction were analyzed. As a result of the analysis, it was confirmed that the understanding of AI and the sense of AI efficacy were improved, and the class satisfaction was high in all items such as educational content, fun in class, difficulty of class, and interest in AI. Based on these results, implications for AI education for secondary students were proposed.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.907-917
• /
• 2023

Recently, a self-driving car have applied deep learning technology to advanced driver assistance system can provide convenience to drivers, but it is shown deep that learning technology is vulnerable to adversarial evasion attacks. In this paper, we performed five adversarial evasion attacks, including MI-FGSM(Momentum Iterative-Fast Gradient Sign Method), targeting the object detection algorithm YOLOv5 (You Only Look Once), and measured the object detection performance in terms of mAP(mean Average Precision). In particular, we present a method applying morphology operations for YOLO to detect objects normally by removing noise and extracting boundary. As a result of analyzing its performance through experiments, when an adversarial attack was performed, YOLO's mAP dropped by at least 7.9%. The YOLO applied our proposed method can detect objects up to 87.3% of mAP performance.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.145-155
• /
• 2024

As the importance of software and artificial education is emphasized on the digital transformation era, various educational materials are being developed and distributed. To achieve the purpose of software education, various software education programs suitable for school settings need to be provided. In this paper, we developed a software education program using self-driving that can be applied to secondary school software education and applied it to secondary school students. The developed software education program is a physical computing program consisting of various motion control programs such as object detection, line tracing using various sensors, focusing on experience and practice. As a result of the survey, students' attitudes and career orientation toward software and artificial intelligence, and satisfaction with software education were over 90%, and satisfaction with the proposed program was over 95%.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.83-89
• /
• 2021

To improve the performance of self-driving cars, the introduction of V2X, a communication technology that connects vehicles, infrastructure, and vehicles, is essential. Even if traffic information of the other vehicle is known, the structure of the intersection and a distance calculation algorithm are required for accurate calculations at roundabouts. This paper proposes a design algorithm for a rotating intersection and implemented in Matlab that complies with the national design rules and enables accurate calculations. Assuming the roundabout and the entrance/exit path to be a circle, a method for measuring the distance between vehicles at an arbitrary point was proposed using the horizontal shift of the entrance circle to the main circle. The algorithm could be used in fully autonomous vehicles by designing a roundabout suitable for the terrain by arbitrarily varying the angle between branches and the radius of curvature of the entrance and exit roads, and transmitting a warning signal when a collision between two driving vehicles is expected.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.3
• /
• pp.93-102
• /
• 2023

Data processing through the cloud causes many problems, such as latency and increased communication costs in the communication process. Therefore, many researchers study edge computing in the IoT, and autonomous driving is a representative application. In indoor self-driving, unlike outdoor, GPS and traffic information cannot be used, so the surrounding environment must be recognized using sensors. An efficient autonomous driving system is required because it is a mobile environment with resource constraints. This paper proposes a machine-learning method using neural networks for autonomous driving in an indoor environment. The neural network model predicts the most appropriate driving command for the current location based on the distance data measured by the LiDAR sensor. We designed six learning models to evaluate according to the number of input data of the proposed neural networks. In addition, we made an autonomous vehicle based on Raspberry Pi for driving and learning and an indoor driving track produced for collecting data and evaluation. Finally, we compared six neural network models in terms of accuracy, response time, and battery consumption, and the effect of the number of input data on performance was confirmed.

• Journal of Korean Academy of Nursing
• /
• v.31 no.1
• /
• pp.126-138
• /
• 2001

The purpose of this study is to act nursing intervention strategies to promote quality of life (QOL) in Spinal Cord Injury patients. Therefore, the study is designed to evaluate religion, marital status, educational background economic status, level of injury, voiding pattern, transportation by driving, rehabilitation education, self-concept, sexual status, perceived barrier, powerlessness, depression, hope, social support, self-efficacy, anxiety, ADL, perceived hopelessness, personality and pain for QOL at the same time. It was gathered empirical data was collected using a self report questionnaire from 61 patients during a six month period after SCI at 5 general hospitals, at 2 rehabilitation centers and at 2 rehabilitation schools province from Jun. 18, 1999 through Sep. 28, 1999. The reliability of the 19 instruments was tested with Cronbach's alpha which ranged from .6319 to .9769. For the data analysis a SAS program was used for Pearson correlation coefficients, and stepwise multiple regression. The results were as follows; 1. The mean of total item score in the QOL scale was 2.814, which showed a much lower score than the cancer points. 2. There was a significant correlation between self concept, transportation by driving, perceived barrier, powerlessness, depression, hope, social support, self-efficacy, anxiety, ADL and QOL.(${\gamma}$= .27~-.79, p< .05) 3. Stepwise multiple regression analysis showed that; a) Depression was the main predictor of QOL and account for 53.7% of the variance in QOL. b) Hope, ecomic status and body satisfaction accounted for 8%, 4%, 3.4% respectively and these variables combined accounted for 69.2% of the variance in QOL. In conclusion, to begin with, it is important to reduce depression and to help for SCI points to inspire hope, economic status and body satisfaction. That will be short cut for nursing strategies to enhance QOL.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.3
• /
• pp.56-61
• /
• 1996

In this paper, new buck converter of a TV power system is presented. First, we devised a revised driving circuit for an emitter-coupled type buck converter, by which it is possible to reduce the material cost of transformers and voltage stress of power device. Secondly, we adopted a hybrid oscillation technique. When TV system is in off-stage, initial standby power which is necessary for remote controllable TV system is supplied by self-oscillating mode. Main power which is necessry in TV system bing on state is provided by an externally triggered oscillating mode. The switching frequency is synchronized to the oscillating frequency of horizontal deflection in TV, by which we can reduce picture noises and the size of power transformer. Thirdly, a simple error amplifier is inserted to the feed-back loop to keep the output voltage constant which means pulse width modulatio mode is added in driving part of power device. Finally, we showed by experiments that our proposed converter performs well enough to be close to the theoretically predicted values.

• Proceedings of the KIEE Conference
• /
• 2006.07b
• /
• pp.813-814
• /
• 2006

The squirrel case induction motor has widely used in the driving of the blowers but it is low efficiency and hard to control. So, the damper is used for the control of a flow and it cause to low the driving efficiency. Our laboratory has proposed the single-phase SRM(switched reluctance motor) for driving blowers. It has salient pole structure and can be reduced a number of semiconductors than three-phase SRM. But it can not be starting by itself and has heavier torque ripple than three-phase SRM. For self-starting the single-phase SRM is required the starting device which place the rotor at the rising inductance slope. On this paper, the electromagnet starting device is designed to generate the starting torque and to fix the rotating direction of the single-phase SRM which is fabricaed to use a blower.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.5
• /
• pp.404-410
• /
• 1991

In this paper,a network interface unit of token-passing bus network and its driving software are implemented based on Mini-MAP. This network interface unit performs the function of MAC layer which is responsible for transmission and reception of frames as well as the management of logical ring. The driving software performs the management of data buffer and the report of errors, if ocoured, to the higher layer. Motorola MC68824 of is used as a TBC(Token Bus Controller) and Intel80186 as a CPU for network interface unit. The operation of network interface unit is verified by self-test which checks the functioning of TBC and CPU. In addition each module of driving software is tested to check the functions regarding transmission and reception of frames.