• Broadcasting and Media Magazine
• /
• v.27 no.2
• /
• pp.43-50
• /
• 2022

한 장의 이미지로부터 장면의 깊이 정보를 추정하는 기술은 자율 주행, 실내외 로봇 기반 서비스 등 다양한 응용 분야에서 널리 적용되고 있다. 심층 학습을 이용한 알고리즘이 활발히 연구되면서 이러한 단안 깊이 추정 기술의 산업 분야 적용 범위는 확대되고 있는 추세이다. 그러나, 깊이 경계 정보를 정밀하게 예측하는데 여전히 많은 어려움이 있으며, 다양한 실제 환경에서 획득한 3차원 깊이 정보 구축 또한 많은 비용이 소모되는 문제점이 있다. 본 고에서는 이러한 문제를 해결하기 위해 최근 활발히 연구되고 있는 심층신경망 기반 단안 깊이 추정 연구의 최신 동향을 소개하고자 한다. 지도 학습 기반 방법부터 최근 활발히 연구되고 있는 비지도 학습 방법까지 상세히 살펴본다. 이와 더불어 대표 방법에 대한 성능 평가 결과도 간략히 제시하고자 한다.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.148-148
• /
• 2017

방역 차량의 약액탱크, 차량의 연료, 워셔액 등의 탱크 내부에는 잔존량을 측정하기 위해 기둥과 floating box로 이루어진 부력식 수위레벨센서가 사용되고 있으나 액체레벨에 따라 float이 상하로 움직이는 측정원리상 차량 주행 중 정확성이 매우 떨어진다(Park et al. 2016). 방역차량이 주행 중 분사할 때, 슬로싱 현상과 방역소독기의 노즐과 펌프에서 발생하는 진동으로 인해 기존의 부력식 센서를 이용한 약제 살포량 측정방법은 정확성이 매우 떨어지는 경향이 있다. 본 연구의 목적은 방역차량이 주행하면서 분사할 때, 수위레벨 센서를 이용한 약제살포량 측정의 정확성을 개선하는 것으로 디지털 칼만필터, Low pass filter와 댐퍼를 제작하여 이용했다. 본 연구에서는 압력식 레벨센서를 이용해 약액탱크의 높이당 단면적과 수위를 측정하여 약제살포량을 계산했다. Python 2.7을 이용해 디지털 칼만필터와 Low pass filter(LPF)를 구현하였으며 3D프린터를 이용해 댐퍼를 제작했다. 실내에서 슬로싱 현상을 인공적으로 만들어 필터와 댐퍼의 수위 측정 정확성 개선효과를 확인 후 실제 방역차량에 부착하여 비포장도로에서 주행하면서 분사할 때 필터와 댐퍼의 효과를 확인하였다. 댐퍼의 공극률(p)을 바꿔가며 수위 측정 정확성 개선효과를 확인하였다. 실내, 현장 실험 결과, 칼만필터가 LPF보다 개선효과가 더 크지만 데이터 50개 처리에 1.71초의 시간지연이 발생했다. 댐퍼는 수위센서를 고정시키고 유체의 운동을 방해하여 이상치와 큰 오차제거에 효과적이었다. 칼만필터와 댐퍼를 동시에 이용할 경우, 수위 측정정확성 $R^2$는 0.9985, 0.9981로 ${\pm}4.3cm$의 범위내에서 수위를 측정할 수 있었다. 필터의 시간지연과 수위 측정정확성을 고려하여 데이터 기록간격을 3초로 설정하면 ${\pm}3cm$이내에서 약탱크 내 수위를 측정할 수 있었다. 공극률(p)가 0.294, 0.291, 0.17에서 측정정확성 $R^2$는 각각 0.9897, 0.9858, 0.9872 로 p가 0.294에서 개선효과가 가장 좋았으나 개선효과의 차이는 크지 않았다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.867-868
• /
• 2013

본 논문은 물류창고와 같은 실내 공간에서 RFID와 초음파 센서등을 이용하여 이동로봇이 자율적으로 자신의 위치를 파악하고 관리자가 지정한 목표 물체를 인식하여 간단한 업무를 보조할 수 있는 기법을 제안한다. 실내공간엔 RFID를 지면과 목표물체에 설치하고 로봇은 RFID의 리더기와 물체 접근시 활용이 가능한 추가적 센서를 갖춤으로써 이동시 자기 위치를 실시간으로 파악하고 물체로부터도 고유정보를 얻는다. 초음파 센서 신호의 귀환시간을 활용하여 근접한 물체와의 상대 거리를 추출하고 바닥의 RFID로부터 이미 획득한 자기 위치를 조합하여 목표 물체의 절대 위치를 구한다. 이는 이동 로봇을 중심으로 한 경로지도를 실시간으로 작성하며 동시에 실내의 이동 가능 구조 및 목표 물체의 파악이 가능하여 이동로봇의 자율적 탐색을 위한 최적 경로 계획 수립에 활용 가능하다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.3
• /
• pp.421-432
• /
• 2013

This paper describes implementation of functions for mobile robot localization, which is one of the vital technologies for autonomous navigation of a mobile robot. There are several function libraries for mobile robot navigation. Some of them have limited applicability for practical use since they can be used only for simulation. Our research focuses on development of functions which can be used for localization of indoor robots. The functions implement deadreckoning and motion model of mobile robots, measurement model of range sensors, and frequently used calculations on angular directions. The functions encompass various types of robots and sensors. Also, various types of uncertainties in robot motion and sensor measurements are implemented so that the user can select proper ones for their use. The functions are tested and verified through simulation and experiments.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.37 no.2
• /
• pp.181-186
• /
• 2013

In recent years, eco-friendliness and high fuel economy have become important issues for commercial tractors. Electric tractors are often required for operation in a greenhouse. However, the battery capacity limits the available operation time. To overcome this problem, a plug-in hybrid electric tractor is considered a reasonable alternative. This tractor has a basic driving ability and can operate in various working modes such as mower, rotary, loader, and trailing. This study focuses on the energy management strategy by considering various working modes.

• International Journal of Highway Engineering
• /
• v.11 no.4
• /
• pp.25-31
• /
• 2009

In general, it is well known fact that the stone mastic asphalt (SMA) pavement has a high resistance against rutting. However, performance of SMA is not well measured by general method used in the laboratory. The objective of this study is to investigate an applicability of deformation strength ($S_D$) for performance estimation of SMA, and to find out the correlation between rut depth and dynamic stability, and $S_D$ of SMA. This study carried out wheel tracking test and Kim-test with optimum asphalt content (OAC) determined by mix design. The results indicated that the $S_D$ of SMA was very poorer than those of dense-graded asphalt mixtures. $S_D$ showed similar WT dynamic stability and rut-depth level. It was found that Kim-test was not reflected higher rutting resistance of SMA like as indirect tensile strength (ITS) test and Marshall stability test. Also, it was revealed that dynamic stability and rut-depth of WT had some problems to estimate rutting resistance of SMA mixtures.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.1D
• /
• pp.127-132
• /
• 2011

Load managing method on road became necessary since overloaded vehicles occur damage on road facilities and existing systems for preventing this damage still show many problems. Accordingly, efficient managing system for preventing overloaded vehicles could be organized by using the road itself as a scale by applying genetic algorithm to analyze the load and the drive information of vehicles. First of all, accurate analysis of load using the behavior of road itself is needed for solving illegal axle manipulation problems of overloaded vehicles and for installing intelligent embedded load analyzing system. Accordingly in this study, to use the behavior of road, the transformation was measured by installing underground box type indoor model and indoor experiment was held using genetic algorithm and 10% error were checked.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.2
• /
• pp.391-399
• /
• 2012

This paper proposes a particle filter approach for SLAM(Simultaneous Localization and Mapping) of a mobile robot. The SLAM denotes estimation of both the robot location and map while the robot navigates in an unknown environment without map. The proposed method estimates robot location simultaneously with the locations of the ultrasonic beacons which constitute landmarks for navigation. The particle filter method represents the locations of the robot and landmarks in probabilistic manner by the distribution of particles. The method takes care of the uncertainty of the landmarks' location as well as that of the robot motion. Therefore, the locations of the landmarks are updated including uncertainty at every sampling time. Performance of the proposed method is verified through simulation and experiments. The method yields practically useful mapping information even if the range data from the landmarks include random noise. Also, it provides more accurate and robust estimation of the robot location than the usual least squares methods or dead-reckoning method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.42 no.2 s.302
• /
• pp.1-10
• /
• 2005

This paper presents a technique for localization of a mobile robot using a single ultrasonic sensor. The mobile robot is designed for operating in a well-structured environment that can be represented by planes, edges, corners and cylinders in the view of structural features. In the case of ultrasonic sensors, these features have the range information in the form of the arc of a circle that is generally named as RCD (Region of Constant Depth). Localization is the continual provision of a knowledge of position which is deduced from it's a priori position estimation. The environment of a robot is modeled into a two dimensional grid map. we defines a physically-based sonar sensor model and employs an extended Kalman filter to estimate position of the robot. The performance and simplicity of the approach is demonstrated with the results produced by sets of experiments using a mobile robot.

• Journal of Industrial Convergence
• /
• v.21 no.12
• /
• pp.91-99
• /
• 2023

This study is a descriptive study to invesigate people's perceptions of the living space in fully autonomous vehicle. Survey was conducted on 250 adults in their 20s to 60s who drive their own vehicle in Korea, and the final study subject were 204 participants who completed responses. Frequency analysis, independent sample t-test, and regression analysis were used to analyze the collected data. For the adequacy of using the four factors of mobility, adaptability, convertibility, and interactivity to evaluate the design of living space in fully autonomous vehicle, the scores of respondents were slightly higher, with an average of 3.76 points on a 5-point scale. Regarding the functions required for fully autonomous vehicle, 'sleep and relaxation' was the most common at 46.4% (130), followed by 'entertainment' at 19% (53), 'business and meeting' at 18.3% (51), and 'health monitoring' at 10.8% (30). It is expected that these study findings will be used as basis for the design of 'living space', a key element of fully autonomous mobility.