http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
최정광(Jungkwang Choi),윤문영(Moonyoung Yoon),정재업(Jaeeop Jung),부광석(Kwangseok Boo),김흥섭(Heungseob Kim) 대한기계학회 2012 대한기계학회 춘추학술대회 Vol.2012 No.11
Recently the automotive industry to predict the risk factors with the sensor to the driver in an accident be avoided by delivering systems that are now underway to study through a combination of various sensors and car, and that during the course of study as part of the study vehicle’s blind-spot detection system is actively underway. Intelligent sensors developed in this study of 15°~45° vehicle blind spot region is recognized by the infrared sensor. In order to increase the sensitivity of the sensor to receive infrared signals mixed with the carrier frequency and the center frequency of the signal consists of a mixed-frequency sensitivity and improved features. SILS environment was built according to international ISO standards based on specifications and configuration of presented BSDS. The environment of the road was built using PreScan and Matlab / simulink was configured using the control logic. In order to verify the applicability of the developed BSDS system, BSDS systems receive information of vehicle distance, steering angle, speed such as actual driving conditions. The applicability of the system presented safe conditions and accident situations by a visual representation and data.
최정광(Jungkwang Choi),부광석(Kwangsuck Boo),윤문영(Moonyoung Yoon),정재업(Jaeeop Jung),김흥섭(Heungseob Kim) 한국자동차공학회 2012 한국자동차공학회 학술대회 및 전시회 Vol.2012 No.11
In this study, the proposed BSD system consists of multiple infrared LEDs which are cheap and robust to weather conditions. It is installed a housing of outside rear-view mirror system to detect other vehicles and objects of the rear left and right-side in the range of 15 to 45 degrees. The layout like as horizontal and vertical angle of emitter/detector, sensing distance and angle of each channel is determined by many simulations and experiments. To maximize the detection distance of infrared LED, the current of an emitter is driving at the modulated frequency which is mixed the carrier frequency with a center frequency of each channel. The detector removes the carrier frequency from measured signal and distinguishes the channel with center frequency. The performance of BSD system is evaluated with SIL simulation and practical field test. The SILS system consists of PreScan and Matlab/Simulink. PreScan yields a realistic driving environments and road conditions. Vehicle model dynamics and collision warning is controlled by Matlab/Simulink. The control and warning logic is intensively evaluated through blind spot warning test and lane change warning test of ISO 17387 standards. Also the developed BSD system is installed on a housing of outside rear-view mirror of vehicle and showed good detection results under various weather conditions.
차량 측후방 비전센서를 이용한 선회차량 인식 시스템에 대한 연구
백승환(Seunghwan Baek),최정광(Jungkwang Choi),김흥섭(Heungseob Kim),윤문영(Moonyoung Yoon),부광석(Kwangsuck Boo) 한국자동차공학회 2012 한국자동차공학회 학술대회 및 전시회 Vol.2012 No.11
When a vehicle changes the lane, the vehicle side and rear crash can be prevented by a blind spot detection system (BSDS) for assisting the driver’s field of view and informing the approaching speed of the other vehicles to the driver. This paper presents a robust method to estimate the relative distance and speed with the approaching vehicles in side and rear blind region of the vehicle for a BSDS with a vision system. A distance from a vehicle in the blind region can be determined by recognition of the pixel positions of the vehicle image on the camera coordinate frame via the coordinate transformation using the perspective camera model under the assumption of uniform elevation of the road and contact of the vehicle on the road surface. The vehicles at the same pixel position on the blind spot image may have different relative distance and speed according to the rear side road curvature, because the vehicle with same relative distance can be represented at different pixel position on the image frame according to the road curvature. Therefore, the rear side lane must be recognized in advance before the vehicles on each lane.