• 정보보호학회논문지
• /
• 제32권1호
• /
• pp.39-48
• /
• 2022

본 논문에서는 무인 비행체 내부 네트워크의 이상 징후를 탐지하는 시퀀스 기반 이상 탐지 시스템을 제안한다. 제안하는 이상 탐지 시스템은 무인 비행체가 지상 통제 시스템에 주기적으로 전송하는 상태 메시지 시퀀스들 간의 유사도를 측정하여 이상 징후를 탐지한다. 본 연구에서는 무인 비행체 내부 네트워크에서 수행 가능한 악의적인 메시지 주입 공격 세 가지를 정의하고, 해당 공격 기법들을 Pixhawk4 쿼드콥터에서 시뮬레이션하였다. 결과적으로, 제안하는 이상 탐지 시스템은 96% 이상의 정확도로 비정상 시퀀스를 탐지할 수 있었다.

• 대한공간정보학회지
• /
• 제25권2호
• /
• pp.21-29
• /
• 2017

최근 활발히 연구가 진행 중인 자율주행차량이나 첨단운전자보조시스템의 효율적이고 안정적인 동작을 위해서 차량 위치를 정확하게 결정하는 것이 중요하다. 주로 사용되는 위성 기반 항법은 신호 수신이 어려운 영역에서 위치 정확도가 매우 떨어지는 한계가 있다. 이를 극복하기 위해 INS 등 추가센서를 이용하는 방안이 모색되고 있지만 높은 비용이 문제가 된다. 이에 본 연구는 고가의 센서를 추가하지 않고 차량에 이미 내장된 센서와 저가의 영상센서를 통합하여 차량의 위치를 정확하게 추정하는 알고리즘을 개발하였다. 차량 내장 센서로부터 제공되는 속력, 각속도와 단영상후방교차법로 결정된 카메라의 위치, 자세를 함께 활용하여 차량의 위치를 추정하였다. 알고리즘의 성능평가를 위해 시범 시스템을 구축하였고, 시험 데이터를 취득하여 주행경로를 추정하였다. 차량 내장 센서만을 이용하였을 경우에 비해 단영상후방교차법 결과를 함께 이용하였을 경우 약 40% 높은 정확도로 차량 경로추정이 가능하였다.

• 한국자동차공학회논문집
• /
• 제21권1호
• /
• pp.1-8
• /
• 2013

Recently, the automotive technology has developed with electronics and information technology as convergence technology while vehicles had been regarded as machines. Moreover, vehicles are becoming more intelligent and safer devices, assembly of advanced technologies by customers' demand. Even though all of installations of vehicle have attracted as diverting devices, it cause drivers' mistakes like delay of response on traffic condition. Here, we proposed the Field Operational Test (FOT) environment which could be used as driving and road conditions collector(Vehicle motion, Traffic condition, Driver input, Driver state, etc.) for researches about Driver Friendly Intelligent System(SCC, LDWS, etc.), Human Vehicle Interface(Driving Workload, etc.) and Economic Drive Model. Furthermore driving patten and fuel consumption patten of drivers were analyzed by measured data and direction of future research was suggested.

• 한국자동차공학회논문집
• /
• 제23권5호
• /
• pp.478-485
• /
• 2015

Availability of the climbing angle information is crucial for the intelligent vehicle system. However, the climbing angle information can't be measured with the sensor mounted on the vehicle. In this paper, climbing angle estimation system is proposed. First, longitudinal acceleration obtained from gyro-sensor is compared with the actual longitudinal acceleration of the vehicle. If the vehicle is in yawing motion, actual longitudinal acceleration can't be approximated from time derivative of wheel speed, because lateral velocity and yaw rate affect actual longitudinal acceleration. Wheel speed and yaw rate can be obtained from the sensors mounted on the vehicle, but lateral velocity can't be measured from the sensor. Therefore, lateral velocity is estimated using unknown input observer with nonlinear tire model. Simulation results show that the compensated results using lateral velocity and yaw rate show better performance than uncompensated results.

• 자동차안전학회지
• /
• 제7권4호
• /
• pp.20-25
• /
• 2015

The basic point of a vehicle recall is to remove vehicle defects as soon as possible and thus prevent possible road traffic accidents caused by the defects beforehand. Therefore, the core of vehicle recall under the self-certification system consists of a timely response and fast remedy of defects. The present study aimed to deduce a plan for improvement of the system necessary for the fast remedy of defects through a phased analysis of defect investigation procedure based on defect investigation statistical data. There will be a need to make the TSB(Technical Service Bulletin) or service campaign data submission of a manufacturer compulsory for the collection of broad defect information in the stage of information analysis and to impose a higher penalty when the manufacturer violates the data submission in the investigation stage. In addition, it is considered that an active service campaign should be induced and a punishment for late recall will be needed for consumer protection.

• 제어로봇시스템학회논문지
• /
• 제10권1호
• /
• pp.66-72
• /
• 2004

To assess tunnel safety, cracks in tunnel lining are measured by inspectors, who observe cracks with their naked eyes and record them. But manual inspection is slow, and measured crack data is subjective. Therefore, this study proposes inspection system fur measuring cracks in tunnel lining and providing objective crack data to be used in safety assessment. The system consists of On-vehicle system and Laboratory system. On-Vehicle system acquires image data with line CCD camera on scanning along the tunnel lining. Laboratory system extracts crack information from the acquired image using image processing. Measured crack information is crack thickness, length and orientation. To improve accuracy of crack recognition, the geometric properties and patterns of cracks in concrete structure were applied to image processing. The proposed system was verified with experiments in both laboratory environment and field environment such as subway tunnel.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.610-630
• /
• 2020

A large number of people suffered from traffic accidents each year, so people pay more attention to traffic safety. However, the traditional methods use laser sensors to calculate the vehicle distance at a very high cost. In this paper, we propose a method based on deep learning to calculate the vehicle distance with a monocular camera. Our method is inexpensive and quite convenient to deploy on the mobile platforms. This paper makes two contributions. First, based on Light-Head RCNN, we propose a new vehicle detection framework called Light-Car Detection which can be used on the mobile platforms. Second, the planar homography of projective geometry is used to calculate the distance between the camera and the vehicles ahead. The results show that our detection system achieves 13FPS detection speed and 60.0% mAP on the Adreno 530 GPU of Samsung Galaxy S7, while only requires 7.1MB of storage space. Compared with the methods existed, the proposed method achieves a better performance.

• Journal of information and communication convergence engineering
• /
• 제8권1호
• /
• pp.83-88
• /
• 2010

Recently car navigation systems which have been widely spread and evolved. The systems use various information and techniques such as real time traffic information and augmented reality technique. In order to provide a shortest path with good flow, real-time traffic information provided by DMB is required. Augmented reality technique is also introduced to give a reality to driver by displaying real images captured by camera during driving. However, these operate well when the system receives GPS data normally. Exact information about the positions of vehicles is a base that supports the above function with realities. This paper proposes a model for acquiring exact position of vehicles. When the GPS does not operate normally, the proposed model uses various data which are generated by integrated vehicle networks.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2004년도 춘계학술대회 논문집
• /
• pp.1459-1461
• /
• 2004

EMU, Electrical Multiple Unit, stores various information in the computer system installed on the vehicle when train is running. The stored information has to be moved into a computer installed on the ground. The information is used to all kinds of planning concerned EMU, and taken advantage of maintenance with information importantly. IC memory card is used to move the information from vehicle to the ground at present. But this method has many weaknesses from hours and manpower aspect because it spends much time shifting data. So this paper proposes a method to move data as quickly and correctly as possible. We propose the method to collect the information using wireless LAN automatically in this paper. And we carried out data-transfer-test on the real field using wireless LAN to confirm possibility of this method. This paper shows the result of the data-transfer-test.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2014년도 춘계학술대회
• /
• pp.319-322
• /
• 2014

최근 인터넷이 활성화 되면서 물류기업의 수 배송 물량이 증가하고 있으나, 제한된 차량과 차주를 확보하는 것이 용이하지 않기 때문에, 수 배송 업무의 합리화는 국내 기업들의 물류관리에서 가장 중요한 과제 중의 하나이다. 이에 따라, 차주와 차량의 효율적 배차 계획 수립을 위해 차량경로 문제와 차량일정계획문제를 해결하기 위한 많은 연구가 있었다. 그러나, 매일 변동하는 교통상황과 세밀한 지리적인 조건을 반영하기 어렵고 지점수가 많아질수록 방대한 크기의 데이터베이스 구축이 필요할 뿐만 아니라 계산시간의 증가로 인한 비효율성이 야기될 수 있다. 본 논문에서는 다양한 제약조건을 가지는 3PL 물류업체의 운송비용에 따른 배차계획 문제를 다루며, 문제 해결을 위해 배차 정보 구축과 운송비용을 고려한 배차 알고리즘을 제안하고 이를 물류업체에 적용한 실험 결과를 제시하였다. 개발된 운송비 기반 배차시스템을 적용해 본 결과 차주의 기존 운송비 수익이 평균 11% 상승하였다.