• 인터넷정보학회논문지
• /
• 제12권5호
• /
• pp.101-111
• /
• 2011

WSN기반 주차관리 시스템에서 대부분의 연구는 주차장에서 사건을 통제하기 위해 무선 센서를 이용하지만, 주차장에서의 차량충돌에 대한 연구는 거의 수행되지 않았다. 시간에 따른 자세한 차량의 위치는 충돌 사건을 분석하는데 매우 중요하다. 본 연구는 주차장에서 차량 충돌사건을 감지하여 분석하고, 이를 차주에게 통보하는 충돌감지 방법을 제시한다. 차량의 위치 및 이동 방향을 감지하기 위해, 움직임 센서로부터의 정보를 활용하며, 빠른 OBB 교차 테스트를 사용하여 검증을 위한 객체를 추적한다. 성능평가 결과 위치추적 기법은 센서를 추가함에 따라 좀 더 정확함을 보였고, 제안한 OBB 충돌 테스트가 일반적인 OBB 교차테스트에 비해 속도가 향상됨을 나타내었다.

• 실천공학교육논문지
• /
• 제13권3호
• /
• pp.491-496
• /
• 2021

본 논문에서 기존 주차장들에 설치되어 있는 CCTV를 활용하여 주차장의 현황을 파악하고 안드로이드 애플리케이션을 통해 사용자에게 실시간 정보를 제공하고자 한다. YOLO V3를 활용하여 주차구역에 관심 영역을 설정하는 방법과 설정된 관심 영역을 통해 실시간으로 변경되는 빈자리 수를 제공하는 방법에 대해 서술하고, ImageZMQ와 파이어베이스를 이용하여 CCTV-서버-사용자 간에 연동하는 방법에 대해 서술한다. 사용자는 애플리케이션을 통해 목적지에 도착하기 전 목적지 근처 주차장 실시간 상황을 파악할 수 있으며 그에 따른 다양한 방안을 마련할 수 있다.

• 한국자동차공학회논문집
• /
• 제11권5호
• /
• pp.176-182
• /
• 2003

The researches for autonomous vehicle have been implemented in many studies, but most studies were confined to the lane fol1owing and changing. This paper addresses a problem of autonomous lane following parking a nonholonomic vehicle. The algorithm for image processing by the hough transform and controlling a steering angle and speed to park a nonholonomic vehicle is developed. The developed system which integrated the control algorithm for parking and vision algorithm for line traction tested with RC car and verified by the performance of the detection of parking area and the reactive parking without collisions.

• 국제초고층학회논문집
• /
• 제3권2호
• /
• pp.155-165
• /
• 2014

This study measured and compared the variation of ventilation rate and fan energy consumption according to various control strategies after installing wireless sensor-based pilot ventilation system in order to verify the applicability of demand-controlled ventilation (DCV) strategy that was efficient ventilation control strategy for underground parking lot. The underground parking lot pilot ventilation system controlled the ventilation rate by directly or indirectly tracking the traffic load in real-time after sensing data, using vehicle detection sensors and carbon monoxide (CO) and carbon dioxide ($CO_2$) sensor. The ventilation system has operated for 9 hours per a day. It responded real-time data every 10 minutes, providing ventilation rate in conformance with the input traffic load or contaminant level at that time. A ventilation rate of pilot ventilation system can be controlled at 8 levels. The reason is that a ventilation unit consists of 8 high-speed nozzle jet fans. This study proposed vehicle detection sensor based demand-controlled ventilation (VDS-DCV) strategy that would accurately trace direct traffic load and CO sensor based demand-controlled ventilation (CO-DCV) strategy that would indirectly estimate traffic load through the concentration of contaminants. In order to apply DCV strategy based on real-time traffic load, the minimum required ventilation rate per a single vehicle was applied. It was derived through the design ventilation rate and total parking capacity in the underground parking lot. This is because current ventilation standard established per unit floor area or unit volume of the space made it difficult to apply DCV strategy according to the real-time variation of traffic load. According to the results in this study, two DCV strategies in the underground parking lot are considered to be a good alternative approach that satisfies both energy saving and healthy indoor environment in comparison with the conventional control strategies.

• ETRI Journal
• /
• 제37권6호
• /
• pp.1220-1230
• /
• 2015

An autonomous valet parking (AVP) system is designed to locate a vacant parking space and park the vehicle in which it resides on behalf of the driver, once the driver has left the vehicle. In addition, the AVP is able to direct the vehicle to a location desired by the driver when requested. In this paper, for an AVP system, we introduce technology to recognize a parking space using image sensors. The proposed technology is mainly divided into three parts. First, spatial analysis is carried out using a height map that is based on dense motion stereo. Second, modelling of road markings is conducted using a probability map with a new salient-line feature extractor. Finally, parking space recognition is based on a Bayesian classifier. The experimental results show an execution time of up to 10 ms and a recognition rate of over 99%. Also, the performance and properties of the proposed technology were evaluated with a variety of data. Our algorithms, which are part of the proposed technology, are expected to apply to various research areas regarding autonomous vehicles, such as map generation, road marking recognition, localization, and environment recognition.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 G
• /
• pp.3206-3208
• /
• 1999

This paper describes the remote fault monitoring system for tower parking facilities. This system consists of central station, remote monitoring equipments and communication equipments. The central station is developed under client-server architecture which composed a DB server, a fault detection client, a status collection client and a A/S client. And the remote monitoring systems are connected to central station by LAN using RAS(Remote Access Service) which is constructed PSTN(Public Switched Telephone Network). This system offers real-time fault detection and status data acquisition of tower parking system.

• 국제학술발표논문집
• /
• The 3th International Conference on Construction Engineering and Project Management
• /
• pp.1278-1282
• /
• 2009

In apartment type of housing, if resident's vehicle is registered in central control office and RFID TAG is issued, identification can be recognized from the time of entrance into parking lot and intelligent parking guide system can be activated based on the residents' profile. Parking Guide System leads a vehicle to the available parking space which is closest to the entrance gate of the vehicle's owner. And when residents forget where they parked their cars, they can query to the Parking Guide and Management System and get responses about the location. For the correct operation of this system, it is necessary to find out where the residents' cars have parked in real time and which lot is available for parking of other cars. RFID is very fancy solution for this system. RFID reader gathers the ID information in RFID TAGs in parked cars and updates the DB up to date. But, when non-residents' cars are parked inside apartment, RFID reader cannot identify them nor know the exact empty/occupied status of parking spaces because they don't react to RFID reader's query. So for the exact detection of empty/occupied status, we suggest the combined use of ultrasonic sensors and RFID. We designed a tree topology with intermediate data aggregators. The depth of tree is normally more than 3 from root (central office) to leaves (individual parking lots). The depth of 2 in tree topology brings about the bottleneck in communication and maintenance. We also designed the information fields used in RFID networks and Sensor Networks.

• 반도체디스플레이기술학회지
• /
• 제19권3호
• /
• pp.93-99
• /
• 2020

A majority of car crash is affected by careless driving that causes extensive economic and social costs, as well as injuries and fatalities. Thus, the research of precise crash detection systems is very significant issues in automotive safety. A lot of crash detection algorithms have been developed, but the coverage of these algorithms has been limited to few scenarios. Road scenes and situations need to be considered in order to expand the scope of a collision detection system to include a variety of collision modes. The proposed algorithm effectively handles the x, y, and z axes of the sensor, while considering time and suggests a method suitable for various real worlds. To reduce nuisance and false crash detection events, the algorithm discriminated between driving mode and parking mode. The performance of the suggested algorithm was evaluated under various scenarios, and it successfully discriminated between driving and parking modes, and it adjusted crash detection events depending on the real scenario. The proposed algorithm is expected to efficiently manage the space and lifespan of the storage device by allowing the vehicle's black box system to store only necessary crash event's videos.

• 전기전자학회논문지
• /
• 제13권2호
• /
• pp.232-240
• /
• 2009

본 논문은 불법 주정차 단속을 위한 정지 차량 검지 및 추적 기법을 제안한다. 제안하는 알고리즘은 크게 네 부분으로 구성되어 있다. 먼저, 입력 영상으로부터 움직이는 차량을 구분하기 위하여 향상된 코드북 물체 검지 알고리즘을 이용한 차량 검지 알고리즘을 제안한다. 두 번째로 차량의 기하학적 특징을 이용하여 차량이 아닌 물체는 제외시키는 전처리 기법을 사용한다. 그런 다음, 검지된 결과 차량들을 히스토그램 추적 기법과 칼만 필터를 결합한 추적 알고리즘을 이용하여 추적한다. 추적 결과를 더 정확하게 하기 위하여, 히스토그램 추적 결과를 칼만 필터의 측정 데이터로 사용한다. 마지막으로, 정지 차량 검지 알고리즘의 신뢰성 있고 정확한 성능을 위하여 실제 정지 카운터 (RSC)를 제안한다. 실험결과로부터 제안한 시스템은 복잡한 실제 도로 환경에서도 여러 차량을 동시에 추적할 수 있고, 정지 차량을 성공적으로 검지해냄을 확인한다.

• 한국자동차공학회논문집
• /
• 제18권4호
• /
• pp.8-17
• /
• 2010

In this paper, a fault tolerant control against sensor faults of electric parking brake (EPB) is proposed. Fault tolerant control method of EPB system is strongly demanded since sensor faults can endanger a driver's safety. In this paper, a clamp force estimation method is presented using motor's armature current and angular velocity. Clamp force estimation method is applied for fault detection method with parity equations. The goal of the detection method is to detect and identify faults in encoder, current sensor, force sensor, and parking cable. And a switching logic for fault tolerant control against the three sensor faults is suggested. Experimental results show that the proposed force estimation method satisfies the specifications of EPB system. The effectiveness of the fault detection method is validated with experimental results. Although a single sensor fault happens, EPB system with the proposed fault detection method does not develop into a failure on subsystem or system level.