• 한국군사과학기술학회지
• /
• 제17권6호
• /
• pp.810-820
• /
• 2014

Recently many studies of Radar systems mounted on ground vehicles for autonomous driving, SLAM (Simultaneous localization and mapping) and collision avoidance are reported. In near field, several hits per an object are generated after signal processing of Radar data. Hence, clustering is an essential technique to estimate their shapes and positions precisely. This paper proposes a method of grouping hits in range-doppler domains into clusters which represent each object, according to the pre-defined rules. The rules are based on the perceptual cues to separate hits by object. The morphological connectedness between hits and the characteristics of SNR distribution of hits are adopted as the perceptual cues for clustering. In various simulations for the performance assessment, the proposed method yielded more effective performance than other techniques.

• 로봇학회논문지
• /
• 제7권2호
• /
• pp.120-128
• /
• 2012

The loop closure problem is one of the most challenging issues in the vision-based simultaneous localization and mapping community. It requires the robot to recognize a previously visited place from current camera measurements. While the loop closure often relies on visual bag-of-words based on point features in the previous works, however, in this paper we propose a line-based method to solve the loop closure in the corridor environments. We used both the floor line and the anchored vanishing point as the loop closing feature, and a two-step loop closure algorithm was devised to detect a known place and perform the global pose correction. We propose an anchored vanishing point as a novel loop closure feature, as it includes position information and represents the vanishing points in bi-direction. In our system, the accumulated heading error is reduced using an observation of a previously registered anchored vanishing points firstly, and the observation of known floor lines allows for further pose correction. Experimental results show that our method is very efficient in a structured indoor environment as a suitable loop closure solution.

• 센서학회지
• /
• 제32권2호
• /
• pp.118-123
• /
• 2023

This study presents sensor system for autonomous mobile robot capable of floor-to-floor self-navigation. The robot was modified using the Turtlebot3 hardware platform and ROS2 (robot operating system 2). The robot utilized the Navigation2 package to estimate and calibrate the moving path acquiring a map with SLAM (simultaneous localization and mapping). For elevator boarding, ultrasonic sensor data and threshold distance are compared to determine whether the elevator door is open. The current floor information of the elevator is determined using image processing results of the ceiling-fixed camera capturing the elevator LCD (liquid crystal display)/LED (light emitting diode). To realize seamless communication at any spot in the building, the LoRa (long-range) communication module was installed on the self-navigating autonomous mobile robot to support the robot in deciding if the elevator door is open, when to get off the elevator, and how to reach at the destination.

• 제어로봇시스템학회논문지
• /
• 제21권10호
• /
• pp.936-943
• /
• 2015

This paper presents arc/line segments-based Simultaneous Localization and Mapping (SLAM) by updating accumulated laser sensor data with a mobile robot moving in an unknown environment. For each scan, the sensor data in the set are stored by a small constant number of parameters that can recover the necessary information contained in the raw data of the group. The arc and line segments are then extracted according to different limit values, but based on the same parameters. If two segments, whether they are homogenous features or not, from two scans are matched successfully, the new segment is extracted from the union set with combined data information obtained by means of summing the equivalent parameters of these two sets, not combining the features directly. The covariance matrixes of the segments are also updated and calculated synchronously employing the same parameters. The experiment results obtained in an irregular indoor environment show the good performance of the proposed method.

• 로봇학회논문지
• /
• 제7권1호
• /
• pp.29-34
• /
• 2012

This paper suggests a new observation model for Extended Kalman Filter based Simultaneous Localization and Mapping (EKF-SLAM). Since the EKF framework linearizes non-linear functions around the current estimate, the conventional line model has large linearization errors when a mobile robot locates faraway from its initial position. On the other hand, the model that we propose yields less linearization error with respect to the landmark position and thus suitable in a large-scale environment. To achieve it, we build up a three-dimensional space by adding a virtual axis to the robot's two-dimensional coordinate system and extract a plane by using a detected line on the two-dimensional space and the virtual axis. Since Jacobian matrix with respect to the landmark position has small value, we can estimate the position of landmarks better than the conventional line model. The simulation results verify that the new model yields less linearization errors than the conventional line model.

• 한국전자통신학회논문지
• /
• 제14권4호
• /
• pp.637-644
• /
• 2019

본 논문은 일반 그래프 최적화(g2o, General Graph Optimization)를 사용하여 그래프 기반 SLAM을 구현한 결과를 기술한다. 일반 그래프 최적화는 SLAM을 노드와 엣지의 그래프를 통하여 표현한다. 노드는 시간에 따른 로봇의 위치를 나타내며, 엣지는 노드들 사이의 구속 조건을 나타낸다. 구속 조건은 센서에 의한 측정값에 의해 결정된다. 일반 그래프 최적화는 구속 조건에 의해 결정되는 성능지표를 최적화하여 SLAM 문제를 해결한다. 실현된 일반 그래프 최적화 방법을 SLAM 방법의 성능 시험용으로 공개된 실험 데이터를 사용하여 검증하였다.

• 한국기계가공학회지
• /
• 제18권8호
• /
• pp.98-103
• /
• 2019

A more accurate method of feature point extraction and matching for three-dimensional reconstruction using low-resolution images of crops is proposed herein. This method is important in basic computer vision. In addition to three-dimensional reconstruction from exact matching, map-making and camera location information such as simultaneous localization and mapping can be calculated. The results of this study suggest applicable methods for low-resolution images that produce accurate results. This is expected to contribute to a system that measures crop growth condition.

• 드라이브 ㆍ 컨트롤
• /
• 제20권4호
• /
• pp.1-8
• /
• 2023

In recent years, robots have been utilized in various industries to reduce workload and enhance work efficiency. The following mobility offers users convenience by autonomously tracking specific locations and targets without the need for additional equipment such as forklifts or carts. In this paper, deep learning techniques were employed to recognize individuals and assign each of them a unique identifier to enable the recognition of a specific person even among multiple individuals. To achieve this, the distance and angle between the robot and the targeted individual are transmitted to respective controllers. Furthermore, this study explored the control methodology for mobility that tracks a specific person, utilizing Simultaneous Localization and Mapping (SLAM) and Proportional-Integral-Derivative (PID) control techniques. In the PID control method, a genetic algorithm is employed to extract the optimal gain value, subsequently evaluating PID performance through simulation. The SLAM method involves generating a map by synchronizing data from a 2D LiDAR and a depth camera using Real-Time Appearance-Based Mapping (RTAB-MAP). Experiments are conducted to compare and analyze the performance of the two control methods, visualizing the paths of both the human and the following mobility.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2022년도 제65차 동계학술대회논문집 30권1호
• /
• pp.323-324
• /
• 2022

본 논문에서는 사람의 개입 없이 실내 환경에서 마스크 미 착용자를 스스로 발견한 후 방역수칙위반 사실에 대한 경고와 함께 마스크 착용을 권고하는 인공지능 기반의 자율주행 서비스 로봇을 개발한다. 제안한 시스템에서 로봇은 동시적 위치 추적 지도 작성 기법인 SLAM(Simultaneous Localization and Mapping)기술을 이용하여 지도를 작성한 후 사용자가 제공한 웨이포인트(Waypoint)를 기반으로 자율주행한다. 또한, YOLO(You Only Look Once) 알고리즘을 이용한 실시간 객체 인식 기술을 활용하여 보행자의 마스크 착용 여부를 판단한다. 실험을 통해 사전에 작성된 지도에 지정된 웨이포인트를 따라 로봇이 자율주행하는 것을 확인하였다. 또한, 충전소로 이동할 경우, 영상 처리 기법을 활용하여 충전소에 부착된 표식에 근접하도록 이동하여 충전이 진행됨을 확인하였다.

• 한국건설관리학회논문집
• /
• 제17권3호
• /
• pp.32-42
• /
• 2016

현재 국내 대부분의 토목 건축 구조물이 BIM 정보가 부재한 상황에서 준공 BIM(as-built BIM)의 수요가 점차 증가하고 있다. 준공 BIM 구축을 위한 공간자료 취득에는 고밀도의 포인트 클라우드를 생성할 수 있는 레이저 스캐너가 주로 활용되고 있다. 하지만 기존의 고정식 스캔 시스템은 이동이 번거롭고, 정밀한 위치 선정이 필요 하며, 스캔 자료 정합을 위해 별도의 표지를 설치하거나 공액점을 추출하는 과정이 필요하다. 본 연구에서는 수작업을 최소화하기 위해 기존의 고정식 스캔 시스템을 대체할 수 있는 이동식 스캔 시스템을 제안하고자 하며, 기반 기술로 graph-based SLAM을 적용하였다. 테스트 장비는 총 세 개의 2차원 스캐너를 탑재하고 있으며, 중앙의 한 개는 수평으로 설치되어 graph 구축을 통한 이동경로취득에 사용되었고, 좌우 두 개는 수직으로 설치되어 시스템 진행의 연직 방향으로 주변 구조물에 대한 3차원 스캔 정보 취득에 사용되었다. 개발된 graph-based SLAM은 이동경로 상에 누적된 위치오차를 해소하기 위한 loop closure 처리 방법으로 Adaboost 기계학습을 적용하였다. 이는 특히 본 연구에서 사용한 장비와 같이 기계학습을 위한 다수의 feature 정보를 제공할 수 있는 멀티 스캐너 시스템에 적합한 방식이며, 두 실내공간을 대상으로 한 테스트에서 단일 스캐너 대비 false positive rate를 각각 7.9% 및 13.6%까지 줄일 수 있었다. 최종적으로 연구대상지역의 2차원 및 3차원 지도 구축을 통해 개발된 graph-based SLAM의 효용성을 확인하였다.