건설현장의 디지털 트윈 구성을 위한 효율적인 3D 스캐닝 방법에 관한 연구

김성훈,김태한,엄이레,원종철 한국BIM학회 2022 KIBIM Magazine Vol.12 No.3

3D scan technology can utilize real spatial information as it is in virtual space, so it can be usefully used in various fields such as reverse engineering of buildings and process management. Recently, with the development of ICT technology, more precise scan data can be obtained, and scan processing time has also been greatly reduced. In addition, the combination of software and scanning equipment used in 3D scanning technology is very diverse, and results are very different depending on which technology is used. Accordingly, there is a problem that it is difficult for a user who has no experience in 3D scanning technology to determine which technology and equipment should be used to obtain good results. In this study, 3D scan technologies mainly used at home and abroad are investigated, classified, and tested at actual construction sites to suggest considerations and suitable 3D scan methods when using 3D scans in construction sites. The test results were analyzed to evaluate the time it takes to scan, the final quality, and the user's convenience according to each technology method.

GPU 기반 실시간 RGB-D 3D SLAM

이동화,김형진,명현 제어로봇시스템학회 2012 제어로봇시스템학회 각 지부별 자료집 Vol.2012 No.12

This paper proposes a real-time RGB-D (red-green-blue depth) 3D SLAM (simultaneous localization and mapping) with GPU (graphic processing unit) acceleration. RGB-D data contains 2D image and per-pixel depth information. 6-DOF (degree-of-freedom) visual odometry is obtained through the 3D-RANSAC (three-dimensional random sample consensus) algorithm with 2D image features and depth data. For speed up extraction of features, parallel computation is performed with GPU acceleration. After a feature man-ager detects loop closure, a graph-based SLAM algorithm optimizes full trajectory of the sensor.

Towards Automated 3D reconstruction in SME factories and Digital Twin Model generation

Mathias Minos-Stensrud,Ole Henrik Haakstad,Olav Sakseid,Baard Westby,Alex Alcocer 제어로봇시스템학회 2018 제어로봇시스템학회 국제학술대회 논문집 Vol.2018 No.10

This paper presents preliminary results towards the development of digital twin models for Small to Medium Enterprise (SME) factories in a partially automated and cost-effective manner. In many cases it is desirable to obtain a 3D model of a factory floor and machinery, that can be used for visualization of Digital Twin models. Current commercial 3D reconstruction solutions involve the use of high-end LiDAR sensors which increase the cost of the 3D scanning process and suppose a barrier for SME factories on their path towards Industry 4.0. The paper presents a comparison of 3D reconstruction results using low-cost sensors including a Zenfone AR mobile phone, an Intel RealSense ZR300 and a Kinect v2. The small size and weight of the sensors make it possible to be mounted on small unmanned aerial vehicles and enable future 3D reconstruction in an autonomous manner. The data was processed using an open source Simultaneous Localization and Mapping (SLAM) library RTAB-Map. The results were compared with a professional 3D scan using a GeoSLAM LiDAR. Experimental results from a scanning of a university research lab with a small simulated production line and two UR3 industrial manipulators is presented. The obtained 3D model was used to generate a simple Digital Twin model that can be visualized using a VR headset.

실내 자율 비행을 위한 마크 기반의 3차원 지도 개발

문성태,조동현,한상혁 한국항공우주학회 2013 한국항공우주학회 학술발표회 논문집 Vol.2013 No.4

최근 자율 비행에 대한 관심이 증가하면서 다양한 센서를 통한 자기 위치 인식 연구가 진행되고 있다. 특히 GPS 와 같은 자기 위치를 확보할 수 없는 실내 환경의 경우, 다른 방법을 통해 자기 위치를 파악해야 한다. 한편, 컴퓨터 비전 분야에서는 이미지 처리를 통해 현재 위치 인식과 동시에 지도를 작성할 수 있는 SLAM 기술이 연구되어 왔다. 특히 최근 저가의 RGB-D 센서가 대중화 되면서 항공 분야에서 SLAM 기술을 적용하기 시작하였다. 하지만 3 차원 지도 생성은 정확성이 많이 떨어져 실제 환경에 적용하기에는 많은 문제가 있다. 본 논문에서는 RGB-D 센서의 한계점에 대해 설명하고, 실내 자율 비행에서 자기 위치를 파악하기 위한 3 차원 지도를 마크를 사용하여 보다 정확하게 생성하는 알고리즘은 제안한다. Recently, the position recognition research has been studied using various sensors according to increased interest in autonomous navigation flight. In case of indoor environment which cannot support GPS information, we have to look for another way to recognize current position. In computer vision, SLAM technology has been studied to look for current position and make map simultaneously. Especially, the SLAM technology is trying to be applied to aviation area by popularization of low-cost RGB-D sensor. However it is difficult to use the sensor in various environments because of low accuracy. In this paper, the author explains the limitation of RGB-D sensor and proposes improved 3D map creation algorithm which uses mark

GPU 가속화를 통한 이미지 특징점 기반 RGB-D 3차원 SLAM

이동화(Donghwa Lee),김형진(Hyongjin Kim),명현(Hyun Myung) 제어로봇시스템학회 2013 제어·로봇·시스템학회 논문지 Vol.19 No.5

This paper proposes an image feature-based real-time RGB-D (Red-Green-Blue Depth) 3D SLAM (Simultaneous Localization and Mapping) system. RGB-D data from Kinect style sensors contain a 2D image and per-pixel depth information. 6-DOF (Degree-of-Freedom) visual odometry is obtained through the 3D-RANSAC (RANdom SAmple Consensus) algorithm with 2D image features and depth data. For speed up extraction of features, parallel computation is performed with GPU acceleration. After a feature manager detects a loop closure, a graph-based SLAM algorithm optimizes trajectory of the sensor and builds a 3D point cloud based map.

키넥트 거리센서를 이용한 실내 이동로봇의 위치인식 및 3 차원 다각평면 지도 작성

권대현(Dae-Hyeon Gwon),김병국(Byung-Kook Kim) 제어로봇시스템학회 2016 제어·로봇·시스템학회 논문지 Vol.22 No.9

We suggest an efficient Simultaneous Localization and 3D Polygon Map Building (SLAM) method with Kinect depth sensor for mobile robots in indoor environments. In this method, Kinect depth data is separated into row planes so that scan line segments are on each row plane. After grouping all scan line segments from all row planes into line groups, a set of 3D Scan polygons are fitted from each line group. A map matching algorithm then figures out pairs of scan polygons and existing map polygons in 3D, and localization is performed to record correct pose of the mobile robot. For 3D map-building, each 3D map polygon is created or updated by merging each matched 3D scan polygon, which considers scan and map edges efficiently. The validity of the proposed 3D SLAM algorithm is revealed via experiments.

A RGB-D Vision based Indoor SLAM using 2.5D Map by Multiple UAVs

Hyunseung. Kang,Kyomun. Ku,Jaehong. Shim 제어로봇시스템학회 2021 제어로봇시스템학회 국제학술대회 논문집 Vol.2021 No.10

This paper presents an approach to build an indoor 2.5D map with multiple UAVs(Unmanned Aerial Vehicles). In an indoor environment, GPS system could not be used so each UAV adopted a tracking camera(Intel RealSense T265) to localize itself and 3D data for map building is acquired by the stereo depth camera(Intel RealSense D435). However, the raw data of the above sensors can have significant noise and large size of data, so the octree filter is applied as a solution. Usually indoor aerial robots have not enough flying time to build entire map of the floor so this paper suggested an improved way to build entire map in 2.5D with multiple UAVs. Each UAV can build a map for part of the global floor and each 3D local maps are converted to 2.5D, then merged together. To make a global map from several local maps, feature points are extracted that can be detected in 2.5D of the indoor environment. Through a series of experiments, the proposed method creates a map that takes into account obstacles and complex structures in the indoor environment, and shows that the ground robot can autonomously drive using this map.

핸드형 및 백팩형 레이저 스캐너의 모델링 활용방안

문석현(Moon, Seok Hyun),남옥우(Nam, Ock woo),최윤수(Choi, Yun Soo) 한국측량학회 2023 한국측량학회지 Vol.41 No.6

Today, it is widely used in important decision-making through design based on 3D drawings and the final results through 3D modeling, rather than designs based solely on 2D drawings. This is a positive function of the digital twin. Currently, 3D spatial information modeling can be acquired with a variety of equipment depending on the purpose. The various purposes include those that require rapid acquisition, those that require high accuracy and precision, and those that increase or decrease the resolution of data that will determine the quality of the data. The 3D modeling construction equipment that makes digital twin possible allows acquisition of data by appropriately converging LiDAR(Light Detection And Ranging), GPS(Global Positioning System), IMU(Inertial Measurement Unit), and cameras according to the purpose. Currently, terrestrial LiDAR is a product that shows high quality and high accuracy by converting indoor spaces into digital twins. In order to secure high location accuracy in indoor spaces, terrestrial LiDAR is used using T/S (total station) used in existing surveying. Research has been conducted to verify the location accuracy of the data acquired, and high-quality indoor spatial information has been produced by creating 3D modeling based on the verified accuracy. In this study, we compared the acquisition speed and data processing speed of two types of newly researched and developed mobile scanners in the same space, using terrestrial LiDAR, backpack-type laser scanner, and hand-held laser scanner, and compared based on terrestrial LiDAR. We verify the performance and conduct research on the use of 3D modeling using the acquired 3D Pointcloud. As a result of the study, it was confirmed that the point cloud created by each laser scanner shows a certain level of error and that 3D modeling can also be used.

건축물 실시간 원격 스캔을 위한 SLAM 시스템 개발 시 고려사항

강태욱 한국BIM학회 2020 KIBIM Magazine Vol.10 No.1

In managing building facilities, spatial information is the basic data for decision making. However, the method of acquiring spatial information is not easy. In many cases, the site and drawings are often different due to changes in facilities and time after construction. In this case, the site data should be scanned to obtain spatial information. The scan data actually contains spatial information, which is a great help in making space related decisions. However, to obtain scan data, an expensive LiDAR (Light Detection and Ranging) device must be purchased, and special software for processing data obtained from the device must be available.Recently, SLAM (Simultaneous localization and mapping), an advanced map generation technology, has been spreading in the field of robotics. Using SLAM, 3D spatial information can be obtained quickly in real time without a separate matching process. This study develops and tests whether SLAM technology can be used to obtain spatial information for facility management. This draws considerations for developing a SLAM device for real-time remote scanning for facility management. However, this study focuses on the system development method that acquires spatial information necessary for facility management through SLAM technology. To this end, we develop a prototype, analyze the pros and cons, and then suggest considerations for developing a SLAM system.

SLAM을 위한 3차원 지도 인터페이스

김범성(BeomSeong Kim),김동엽(Dong Yeop Kim),황재필(Jae Pil Hwang),김은태(Euntai Kim),김영욱(Young-Ouk Kim) 한국지능시스템학회 2009 한국지능시스템학회 학술발표 논문집 Vol.19 No.1

SLAM에서 나오는 Data들은 모두 3차원 좌표로 존재하게 된다. 하지만 기존에 3차원 Map Building에서 주로 사용해 왔던 방법들은 속도가 느리고 높은 Hardware 사양을 요구하기 때문에 실시간으로 로봇을 구동하고 data를 처리하는 데에는 불편함이 많았다. 그래서 OpenGL을 이용하여 그래픽 API로 SLAM에 적합한 3D Map building Interface를 개발하게 되었다. 개발된 Interface는 Feature를 특성에 맞게 분류하여 표현하기 때문에 좀 더 나은 개발 환경을 제공하게 될 것이다.