• 산업경영시스템학회지
• /
• 제25권3호
• /
• pp.28-33
• /
• 2002

Several studies indicated that an independent visual background (IVB) reduced simulator sickness (SS) and balance disturbance associated with exposure to virtual environments (VEs) and motion simulators. A recent study showed that an IVB comprised of an earth-fixed grid was less effective in a complex driving simulator than in a simple VE. Subjects' post-experiment reports indicated that the VE motion "induced" motion of the earth-fixed grid m. This led to the suggestion that an IVB comprised of clouds would be less subject to induced motion and therefore would alleviate nausea more effectively than a grid IVB. Clouds are "natural" and are usually perceived as relatively stable, whereas a grid has no inherent stability. 12 subjects were exposed to complex motion through a simulated environment in a driving simulator under 3 IVB conditions: grid, less clouds, many clouds. They reported less nausea when the many-cloud IVB was used relative to the grid IVB condition. grid IVB condition.

• 반도체디스플레이기술학회지
• /
• 제22권3호
• /
• pp.161-165
• /
• 2023

One of the difficult works in an autonomous driving system is detecting road lanes or objects in the road boundaries. Detecting and tracking a vehicle is able to play an important role on providing important information in the framework of advanced driver assistance systems such as identifying road traffic conditions and crime situations. This paper proposes a vehicle detection scheme based on deep learning to classify and tracking vehicles in a complex and diverse environment. We use the modified YOLO as the object detector and polynomial regression as object tracker in the driving video. With the experimental results, using YOLO model as deep learning model, it is possible to quickly and accurately perform robust vehicle tracking in various environments, compared to the traditional method.

• 대한교통학회지
• /
• 제35권2호
• /
• pp.160-168
• /
• 2017

지난 10년 간 졸음운전은 전체 고속도로 사고건수의 약 23%로 교통사고 사망원인 중 가장 높은 비중을 차지하고 있다. 과속, 주시태만 등 운전자 과실이 주요원인인 일반적인 사고유형과 달리, 졸음운전은 졸음이라는 불가항력적 원인에 의해 발생한다는 점에서 타 사고유형과 차별화된 접근이 요구된다. 그 동안의 졸음운전 감소대책은 일반적인 교통사고 대책과 마찬가지로 사고다발지점과 같은 특정지점(spot)에 집중하였으나, 도로특성(해당구간의 화물차 비율 등) 또는 시간특성(누적주행시간에 따른 위험 운전행동 증가 등)을 고려한 감소대책이 필요함에 따라, 본 연구에서는 시 공간적으로 확대한 구간(link) 개념을 도입하였다. 고속도로 졸음운전 위험구간 분석을 위해 화물차 디지털 운행기록계(digital tacho graph: DTG) 자료를 활용하였으며, 이를 바탕으로 졸음운전 위험구간을 산정하였다. 위험 행동지표와 사고 발생건수 간의 상관 분석을 위해 음이항 회귀모형(negative binomial regression)을 통한 졸음사고 예측모형을 추정하였으며 모형의 결과 값을 바탕으로 경험적 베이즈(empirical Bayes: EB) 추정치와 구간별 잠재적 안전개선 지수(potential for safety improvement: PSI)를 산출하여 졸음운전 위험 구간을 선정하였다. 졸음사고 모형 추정 결과, 연평균 일교통량, 화물차 비율, DTG 수집 자료건수, 평균 과속비율(20km/h 초과), 평균 급감속비율 및 평균 급차로변경비율이 늘어날 경우 졸음운전 사고건수 역시 증가하는 것으로 분석되었다.

• 인터넷정보학회논문지
• /
• 제23권2호
• /
• pp.53-59
• /
• 2022

3단계 자율주행 차량 이후, 4, 5단계의 자율주행 기술은 차량의 완벽한 주행뿐만 아니라 탑승객의 상태를 최적으로 유지하기 위해 노력하고 있다. 그러나 현재 자율주행 기술은 LiDAR, 전방 카메라 등 시각적 정보에 과하게 의존하기 때문에 지정된 도로 이외의 도로에서 완벽하게 자율주행을 실행하기 힘들다. 따라서 본 논문은 차량이 시각 정보 외의 데이터를 사용하여 도로의 상태를 분류하고, 도로 상태와 주행 상태에 따라 최적의 제동 강도를 계산하는 BSCS (Braking Strength Calculation System)를 제안한다. 본 논문에서 제안하는 BSCS는 KNN 알고리즘을 기반으로 도로의 상태를 분류하는 RCDM (Road Condition Definition Module)과 RCDM의 결과와 현재 주행 상태를 통해 주행 중 최적의 제동 강도를 계산하는 BSCM (Braking Strength Calculation Module)로 구성된다. 본 논문의 실험 결과, KNN 알고리즘에 가장 적합한 K의 수를 찾을 수 있었고, 비지도 학습인 K-means 알고리즘보다 본 논문에서 제안한 RCDM이 더 정확한 것이 증명되었다. 해당 논문의 BSCS는 시각 정보뿐만 아니라 서스펜션에 가해지는 진동 데이터를 사용함으로써, 시각 정보가 제한되는 여러 환경에서 자율주행 차량의 제동을 더 원활하게 만들 수 있다.

• 가정과삶의질연구
• /
• 제23권1호
• /
• pp.125-137
• /
• 2005

Information technology is an important driving force that has changed consumer information environments. In order to adjust in the new environments, consumers need an innovative information system. The purpose of this study was to develop a Consumer Information System (CIS). CIS is a device that supports consumer's decision-making process and elevates consumer information competence. The CIS was constructed by the following steps: (1) organization of developers, (2) systematization of consumer information, (3) data loading, (4) integration of consumer database: data warehouse, (5) data distribution, (6) composition of data mart, (7) use of data access tools: data-mining, OLAP, statistical analysis, Q+R, (8) data visualization: web server.

• Corrosion Science and Technology
• /
• 제18권4호
• /
• pp.117-120
• /
• 2019

Considering the tendency of development of electrification vehicles, development and verification of new evaluation technology is needed because of new technology applications. Recently, as the battery package is set outdoors of an electric vehicle, such vehicles are exposed to corrosive environments. Among major components connected to the battery package, rust prevention of high-voltage cables and connectors is considered the most important issue. For example, if corrosion of high voltage cable connectors occurs, the corrosion durability assessment of using an electric vehicle will be different from general environmental corrosion phenomena. The purpose of this study is to investigate the corrosion mechanism of high voltage cable connectors of an electric vehicle under various driving environments (road surface vibration, corrosion environment, current conduction by stray current, etc.) and develop an optimal rust prevention solution. To improve our parts test method, we have proposed a realistic test method to reproduce actual electric vehicle corrosion issues based on the principle test.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.228-230
• /
• 2009

Lane extraction is basically required for a driving car to understand its external road environments via a camera. In this paper, a lane extraction method using "Sobel Intensity Profile" is described. The Sobel intensity profile is obtained using only vertical edge components of Sobel edge outputs, and used to yield fitted lines for lanes. The RANAC algorithm is applied to fit lines using only inliers. Experimental results have shown the reliability of the proposed lane extraction method.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2003년도 춘계학술대회 논문집
• /
• pp.828-831
• /
• 2003

The benefits of the pneumatic valve with piezoelectric element are faster response times, low energy consumption, and the ability to be used in hazardous environments and field bus systems. In this paper, PZT actuator, 2 and 3 stages pneumatic valve were designed and manufactured. Also. characteristics of the pneumatic valve with piezoelectric element were tested with a testing system. It is confirmed that the PZT actuator is useful one for controlling the direction of pilot valve.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2005년도 춘계 학술대회 발표논문집
• /
• pp.103-106
• /
• 2005

The goal of our research is to build a text independent speaker recognition system that can be used in any condition without any additional adaptation process. The performance of speaker recognition systems can be severally degraded in some unknown mismatched microphone and noise conditions. In this paper, we show that PCA(Principal component analysis) without dimension reduction can greatly increase the performance to a level close to matched condition. The error rate is reduced more by the proposed augmented PCA, which augment an axis to the feature vectors of the most confusable pairs of speakers before PCA

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.499-503
• /
• 1997

In this paper, we present a fuzzy ICC using a self-tuning method. To provide robustness and adaptiveness over the vehicle nonlinearities and changes of the driving environments, an on-line self-tuning scheme based on 'Interior Penalty Function' was developed. Road test and computer simulation results verify the feasible performance of the suggested ICC algorithm.