• International Journal of Internet, Broadcasting and Communication
• /
• 제13권1호
• /
• pp.210-218
• /
• 2021

CNN (Convolutional Neural Network), a type of deep learning algorithm, is a type of artificial neural network used to analyze visual images. In deep learning, it is classified as a deep neural network and is most commonly used for visual image analysis. Accordingly, an AI autonomous driving model was constructed through real-time image processing, and a crosswalk image of a road was used as an obstacle. In this paper, we proposed a low-cost model that can actually implement autonomous driving based on the CNN model. The most well-known deep neural network technique for autonomous driving is investigated and an end-to-end model is applied. In particular, it was shown that training and self-driving on a simulated road is possible through a practical approach to realizing lane detection and keeping.

• 한국자동차공학회논문집
• /
• 제21권6호
• /
• pp.123-134
• /
• 2013

Strengthening vehicle emission regulation is one of important policies to improve air quality in urban area. Due to the limitation of specified driving cycles for certification test to reflect real driving conditions, additional off-cycle emission regulations have been adopted in US and being developed in Europe. The driving cycles of US or Europe have been used in emission certification for Korean light-duty vehicles, but it has not been known how well the driving cycles reflect various real driving patterns in Korea. In that point of view, it is required to estimate vehicle emission based on real road driving conditions to raise the effectiveness of vehicle emission regulation in Korea. In this study, real driving emission measurements have been conducted for three Korean light-duty vehicles with PEMS. The driving routes consisted of urban, rural and motorway in Seoul and Incheon. The data have been analyzed with various averaging methods including moving averaging windows method and compared to emission limits set with emission certification modes applied to tested vehicles. The results have shown that the real driving pollutant emissions of a gasoline and a LPG vehicles have been ranged quite lower than those of emission limits on CVS-75 driving cycle. But real driving NOx of a light duty diesel vehicle has been considerably higher than emission limit of NEDC driving cycle. The higher than expected NOx emission of a diesel vehicle might be caused by different strategy to control EGR in real driving condition from NEDC driving.

• 대한교통학회지
• /
• 제21권4호
• /
• pp.91-101
• /
• 2003

본 연구에서는 현장실험 실시의 제약을 극복하기 위해 3차원 가상현실(Virtual Reality)로 4차로 고속도로를 모델링한 후, 피실험자가 차량시뮬레이터(Driving Simulator)에 탑승하고 주행하면서 혼합곡선부의 선형변화에 대한 동적인 반응 데이터를 획득함과 동시에 도로 기하구조 이미지 인지특성 데이터를 획득하였다. 특히, 본 실험에 적용된 차량시뮬레이터의 그래픽 모듈은 동역학 해석 모듈로 얻은 데이터에 기초하여 운동재현기와 가상환경의 일치감을 최대로 하여 피실험자로 하여금 Simulator Sickness를 최소화하도록 하였으며, 감성요인으로 사용된 어의는 국어사전 및 전문가의 견해를 바탕으로 추출된 7개 항목의 형용사를 바탕으로 감성평가를 실시하였다. 또한 감성적인 측면에서 도로 선형과 안전성에 대한 인과관계를 규명하기 위하여 LISREL(Linear Structural Relationships) 모형을 이용하여 정적인 도로 안전성 모형과 동적인 도로 안전성 모형을 구축하여 복합선형구간에서 도로 기하구조의 변화에 따른 운전자들의 안전성에 영향을 미치는 요인을 규명하였다. 그 결과, 도로 설계시 운전자의 감성적인 측면을 고려한 도로 설계가 필요하며 이는 안전한 도로를 건설하는데 있어 매우 중요한 요인이며 이를 통하여 도로 안전성 향상에 기여할 수 있다고 생각된다.

• International Journal of Precision Engineering and Manufacturing
• /
• 제8권3호
• /
• pp.3-7
• /
• 2007

There have been numerous studies directed toward the development of driving mechanisms for off-road mobility and rescue robots. To achieve surveillance, reconnaissance, and rescue, it is necessary for robots to have a driving mechanism that can handle off-road environments, We propose a new type of single-track driving mechanism with a variable geometry for a rescue robot, This mechanism has a symmetric configuration so that the robot can advance in two directions and also remain operable when overturned. By transforming its geometry, the robot can reduce energy consumption in steering and rotating as well as maximize its ability to climb obstacles such as stairs. The robot is also designed to have a compact size and low center of gravity to facilitate driving when on a set of stairs. In this paper, we analyzed the design parameters of the robot for the four phases of climbing stairs and determined the specifications needed to enhance its adaptability.

• 한국공작기계학회논문집
• /
• 제14권3호
• /
• pp.110-115
• /
• 2005

Driving performance is affected by a steering mechanism and characteristics of the ground at off-road skateboarding. In order to drive on off-road, it is necessary off-road wheel and high performance steering mechanism to adapt on various configuration of the ground. In this paper, design factors are studied to affect to steering radius such as inclination angle of a king-bolt, distance of a wheel axle, and rolling angle of a deck plate. A steering system is adhered to inclination face of the deck plate. And, inclination angle is existed between the king-bolt and the flat face of the deck plate. Therefore, the wheel axle of the steering system can be steered by control of the rolling angle of the deck plate.

• 드라이브 ㆍ 컨트롤
• /
• 제20권1호
• /
• pp.8-15
• /
• 2023

With the recent trend of intelligence and automation of construction work, a double-blade road cutter is being developed that automatically enables cutting along the cutting line marked on the road using a vision system. The road cutter can recognize the cutting line through the camera and correct the driving route in real-time, and it detects the load of the cutting blade in real-time to control the driving speed in case of overload to protect workers and cutting blades. In this study, a vision system mounted on a double-blade road cutter was developed. A cutting route recognition technology was developed to stably recognize cutting lines displayed on non-uniform road surfaces, and performance was verified in similar environments. In addition, a vision sensor protection module was developed to prevent foreign substances (dust, water, etc.) generated during cutting from being attached to the camera.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2006년도 춘계공동학술대회
• /
• pp.429-436
• /
• 2006

Standardization strategy is considered as an important approach for the organization growth and competitiveness. Major contents of standardization are the selection of standardization area and management of standardization timing under limited resources. The developments of standardization road map and its driving strategy in construction area are important policy subjects in view of investment scale and national economy. In this paper, we aim to survey on the standardization theory and its application in construction field and suggest research direction for driving strategy of standardization road map development.

• 한국품질경영학회:학술대회논문집
• /
• 한국품질경영학회 2004년도 품질경영모델을 통한 가치 창출
• /
• pp.284-288
• /
• 2004

Standardization strategy is considered as an important approach for the organization growth and competitiveness. Major contents of standardization are the selection of standardization area and management of standardization timing under limited resources. The developments of standardization road map and its driving strategy in construction area are important policy subjects in view of investment scale and national economy. In this paper, we aim to survey on the standardization theory and its application in construction field and suggest research direction for driving strategy of standardization road map development.

• 한국자동차공학회논문집
• /
• 제8권5호
• /
• pp.188-195
• /
• 2000

This study was conducted to develop the mathematical model and the computer simulation program(TPPMWV) for predicting the tractive performance of off-road wheeled vehicles operated on various soil conditions. The model takes into account main design parameters of a wheeled vehicle, including the radius and width of front and rear tires, the weight of vehicle, wheelbase and driving type(4WD, 2WD). Soil characteristics, such as the peressure-sinkage and shearing characteristics and the response to repetitive loading, are also taken into consideration. The effectiveness of the developed model was verified by comparing the predicted drawbar pulls using TPPMWV with measured ones obtained by field tests for two different driving types of wheeled vehicle. As a results, the drawbar pulls predicted by the TPPMWV were well matched to the measured ones within the absolute errors of 5.25%(4WD) AND 9.42%(2WD)for two different driving types, respectively.

• 한국공작기계학회논문집
• /
• 제14권5호
• /
• pp.55-61
• /
• 2005

In this paper, we propose the visual simulation system for virtual reality motion rider system. The visual simulation system can apply verity virtual reality system. This paper deals with programs on 3D automatic creation of terrain, road design, and a realtime rendering program for the virtual reality system. For the 3D automatic creation of terrain, DEM data and rectangular grid method are applied. We can make two different road object with the road design program. One of them includes road definition, and the other is obtained by using 'NURBS curve.' Visual simulation is consisted by additional modeling and real-time rendering. We can apply the programs made in this way to visual system of driving simulator.