• 대한조선학회논문집
• /
• 제61권2호
• /
• pp.106-114
• /
• 2024

In the 21st century, the rapid development of automation and artificial intelligence technologies is driving innovative changes in various industrial sectors. In the transportation industry, this is evident with the commercialization of autonomous vehicles. Moreover research into autonomous navigation technologies is actively underway in the aviation and maritime sectors. Consequently, for the practical implementation of autonomous ships, an effective collision avoidance algorithm has become a crucial element. Therefore, this study proposes a collision avoidance algorithm based on the Obstacle Zone by Target(OZT), which visually represents areas with a high likelihood of collisions with other ships or obstacles. The A-star algorithm was utilized to represent obstacles on a grid and assess collision risks. Subsequently, a collision avoidance algorithm was developed that performs fuzzy control based on calculated waypoints, allowing the vessel to return to its original course after avoiding the collision. Finally, the validity of the proposed algorithm was verified through collision avoidance simulations in various encounter scenarios.

• International Journal of Advanced Culture Technology
• /
• 제10권2호
• /
• pp.300-306
• /
• 2022

Currently, Korea is building traffic infrastructure using Intelligent Transport Systems (ITS), but the pedestrian traffic accident rate is very high. The purpose of this paper is to prevent the risk of traffic accidents by jaywalking pedestrians. The development of this study aims to detect pedestrians who trespass using the public data set provided by the Artificial Intelligence Hub (AIHub). The data set uses training data: 673,150 pieces and validation data: 131,385 pieces, and the types include snow, rain, fog, etc., and there is a total of 7 types including passenger cars, small buses, large buses, trucks, large trailers, motorcycles, and pedestrians. has a class format of Learning is carried out using YOLOv5 as an implementation model, and as an object detection and edge detection method of an input image, a canny edge model is applied to classify and visualize human objects within the detected road boundary range. In this study, it was designed and implemented to detect pedestrians using the deep learning-based YOLOv5 model. As the final result, the mAP 0.5 showed a real-time detection rate of 61% and 114.9 fps at 338 epochs using the YOLOv5 model.

• 한국재난정보학회 논문집
• /
• 제19권2호
• /
• pp.421-431
• /
• 2023

연구목적: 건설기계의 안전사고를 예방하기 위해 사용하는 스마트 안전 시스템의 안전성과 재해 예방효과를 분석하여 그 안전성을 실증하고자 한다. 연구방법:건설기계 중 재해다발 및 위험성이 높은 굴삭기를 대상으로 스마트 안전 시스템의 유무에 따른 근로자의 행동 패턴을 분석하였다. 연구결과: 건설기계에 스마트 안전 시스템을 설치하였을 때 건설기계와 협착 및 충돌 등에 의한 재해로부터 근로자의 안전성이 확보되었다. 결론:건설기계에 설치된 스마트 안전 시스템은 건설기계 관련 재해 감소와 중대 재해예방에 실효성을 증대시킬 수 있을 것으로 판단된다.

• 한국컴퓨터정보학회논문지
• /
• 제28권11호
• /
• pp.21-28
• /
• 2023

다품종 소량 생산을 중심으로 하는 제조 기업의 경우 하나의 공정 라인에서 여러 가지의 제품들을 생산하기 때문에 자동화된 검수 보다는 작업자에 의해서 불량품을 선별하고 있다. 따라서 일정한 기준 없이 작업자의 경험이나 숙련도에 의해 선별 기준이 조금씩 차이가 있어 잘못 선별이 이루어질 가능성이 높다. 또한, 크기나 모양 등이 정형화되지 않은 유연물체의 경우 선별 기준에 대한 편차가 더 커질 수 있는 문제가 있다. 이러한 문제를 해결하기 위해 본 논문에서는 인공지능 기반의 비지도 학습 방법을 적용한 품질 검사 시스템을 설계하고 실제 제조현장에서 획득한 데이터 셋을 기반으로 정확도를 실험하는 연구를 진행하였다.

• Journal of Multimedia Information System
• /
• 제6권3호
• /
• pp.139-146
• /
• 2019

This paper, wireless internet such as Wi-Fi has a vulnerability to security. Blockchain also means a 'Ledger' in which transaction information that occurs on a public or private network is encrypted and shared among the network participants. Blockchain maintains information integrity by making it impossible for a particular node to tamper with information arbitrarily, a feature that would result in changes in the overall blockchain hash value if any one transaction information that constitutes a block was changed. The complete sharing of information through a peer-to-peer network will also cripple hacking attempts from outside, targeting specialized nodes, and prepare for the "single point of failure" risk of the entire system being shut down. Due to the value of these Blockchain, various types of Blockchain are emerging, and related technology development efforts are also actively underway. Various business models such as public block chains such as Bitcoin, as well as private block chains that allow only certain authorized nodes to participate, or consortium block chains operated by a select few licensed groups, are being utilized. In terms of technological evolution, Blockchain also shows the potential to grow beyond cryptocurrency into an online platform that allows all kinds of transactions with the advent of 'Smart Contract'. By using Blockchain technology, the company makes suggestions to overcome the vulnerability of wireless Internet.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.252-254
• /
• 2021

산업안전보건연구원에 따르면 2019년 산업재해자 수가 109,242명으로 2018년에 비해 6.8% 증가하였다. 이러한 산업 안전보건 분야는 질병보다 사고가 더 자주 발생하고 있다. 이러한 상황에서 정부와 기업은 건설 시공 분야에서 ICT 기반 현장 안전사고 예방 핵심 기술 개발이 논의되고 있는 실정이다. 이러한 분야에서 최근 컴퓨터 비전과 인공지능을 활용한 기술들이 많이 사용되고 있다. 본 논문에서는 공장 근로자들의 안전관리를 위한 학습 데이터를 구축하고 YOLOv4를 기반으로 모델을 학습시켰다. 이를 통해 공장에서 근로자들의 위험 상황을 예측하는 초기 연구로써 활용할 수 있을 것으로 사료된다.

• 한국안전학회지
• /
• 제28권3호
• /
• pp.118-122
• /
• 2013

For safe evacuation in the fire disaster, the evacuees must find the exit and evacuate quickly. Especially, if the evacuees don't know the location of the exit, they have to depend on the evacuation guidance system. Because the more smoke spread, the less visibility is decreasing, it is difficult to find the way to the exit by the naked eye. For theses reasons, the evacuation guidance system is highly important. However, the evacuation guidance system without change of direction has the risk that introduce to the dangerous area. In the evacuation safety assessment scenario by the evacuation simulation has the same problem. Because the evacuee in the simulation evacuate by the shortest route to the exit, the simulation result is same like the evacuation without the evacuation guidance system. In this study, it was used with MAS (Multi Agent System)-based simulation program including the evacuation guidance system to implement the change of evacuation by fire. Using this method, confidence of evacuation safety assessment can be increase.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.54-57
• /
• 2022

대한민국은 산불의 위험으로부터 취약한 환경을 가지고 있으며, 이로 인해 매년 큰 피해가 발생하고 있다. 이를 예방하기 위해 많은 인력을 활용하고 있으나 효과가 미흡한 실정이다. 만약 인공지능 기술을 통해 산불을 조기 발견해 진화된다면 재산 및 인명피해를 막을 수 있다. 본 논문에서는 산불의 피해를 최소화하기 위한 오브젝트 디텍션 모델을 제작하는 과정에서 발생하는 데이터 수집과 가공 과정을 최소화하는 목표로 한 객체 이중 탐지 방법을 연구했다. YOLOv5에서 한정된 이미지를 학습한 단일 모델을 통해 일차적으로 원본 이미지를 탐지하고, 원본 이미지에서 탐지된 객체를 Crop을 통해 잘라낸다. 이렇게 잘린 이미지를 재탐지하는 객체 이중 탐지 방법을 통해 오 탐지 객체 탐지율의 개선 가능성을 확인했다.

• 방사성폐기물학회지
• /
• 제20권1호
• /
• pp.99-110
• /
• 2022

In decommissioning a nuclear power plant, numerous concrete structures need to be demolished and decontaminated. Although concrete decontamination technologies have been developed globally, concrete cutting remains problematic due to the secondary waste production and dispersion risk from concrete scabbling. To minimize workers' radiation exposure and secondary waste in dismantling and decontaminating concrete structures, the following conceptual designs were developed. A micro-blast type scabbling technology using explosive materials and a multi-dimensional contamination measurement and artificial intelligence (AI) mapping technology capable of identifying the contamination status of concrete surfaces. Trials revealed that this technology has several merits, including nuclide identification of more than 5 nuclides, radioactivity measurement capability of 0.1-10⁷ Bq·g^-1, 1.5 kg robot weight for easy handling, 10 cm robot self-running capability, 100% detonator performance, decontamination factor (DF) of 100 and 8,000 cm²·hr^-1 decontamination speed, better than that of TWI (7,500 cm²·hr^-1). Hence, the micro-blast type scabbling technology is a suitable method for concrete decontamination. As the Korean explosives industry is well developed and robot and mapping systems are supported by government research and development, this scabbling technology can efficiently aid the Korean decommissioning industry.

• 한국기계가공학회지
• /
• 제20권12호
• /
• pp.1-7
• /
• 2021

The deep learning image processing technique was used to prevent accidents in lathe work caused by worker negligence. During lathe operation, when the chuck is rotated, it is very dangerous if the operator's hand is near the chuck. However, if the chuck is stopped during operation, it is not dangerous for the operator's hand to be in close proximity to the chuck for workpiece measurement, chip removal or tool change. We used YOLO (You Only Look Once), a deep learning image processing program for object detection and classification. Lathe work images such as hand, chuck rotation and chuck stop are used for learning, object detection and classification. As a result of the experiment, object detection and class classification were performed with a success probability of over 80% at a confidence score 0.5. Thus, we conclude that the artificial intelligence deep learning image processing technique can be effective in preventing incidents resulting from worker negligence in future manufacturing systems.