• Journal of the Korea Institute of Building Construction
• /
• v.23 no.5
• /
• pp.629-638
• /
• 2023

This study introduces an innovative approach to monitor the whereabouts of workers within indoor construction settings. While traditional modalities such as GPS and NTRIP have demonstrated efficacy for outdoor localizations, their precision dwindles in indoor environments. In response, this research advocates for the adoption of Aruco markers. Leveraging computer vision technology, these markers facilitate the quantification of the distance between a worker and the marker, subsequently pinpointing the worker's instantaneous location with heightened accuracy. The methodology's efficacy was rigorously evaluated in a real-world construction scenario. Parameters including system stability, the influence of lighting conditions, the extremity of measurable distances, and the breadth of recognition angles were methodically appraised. System stability was ascertained by maneuvering the camera at a uniform velocity, gauging its marker recognition prowess. The impact of varying luminosity on marker discernibility was scrutinized by modulating the ambient lighting. Furthermore, the camera's spatial movement ascertained both the upper threshold of distance until marker recognition waned and the maximal angle at which markers remained discernible.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.05a
• /
• pp.205-206
• /
• 2023

This paper presents a new method for indoor localization track workers in construction sites. While GPS and NTRIP are effective for outdoor positioning, they are less accurate when used indoors. To address this issue, the proposed method utilizes Aruco markers to measure the distance between workers and the markers. By collecting data values, the location of each worker can be determined in real-time with high accuracy. This approach has the potential to enhance work efficiency and safety at construction sites, as it provides more precise indoor positioning compared to conventional methods, leading to improved work efficiency.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.250-252
• /
• 2018

본 논문은 4차 산업 혁명 핵심 기술인 자율주행에 대하여 기술하였으며 그 중 Hector Slam을 사용 하였다. Hector slam 같은 경우 RAM이 4G 이상 되어야 제대로 동작하지만 SBC(Odroid xu4) 같은 경우 RAM의 크기가 2G이므로 최적화할 필요성이 있다. SBC(Odroid xu4)에서도 사용 가능하도록 Hector slam 구현 최적화를 하였으며, 향후에 Aruco Marker를 이용하여 위치를 좀 더 섬세히 보정 해볼 것이며 또한 Aruco Marker의 ID를 통해 사물 인식을 하여 사람에게 사물에 대한 정보를 알려줌으로써 사람과 협업을 할 수 있는 로봇이 될 것이다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.131-132
• /
• 2020

본 논문에서는 OpenCV의 Marker Detection 기술을 이용하여 특정지점의 마커를 영상처리기술로 인식하여 드론의 자동 이착륙 및 주변 위기상황, 미션수행 등을 마커를 통해서 드론에게 전달하여 비행 제어할 수 있는 체계를 개발한다. 드론은 OpenCV Aruco모듈을 이용하여 Marker ID별로 특정 명령어를 데이터 베이스와 비교하여 비행제어 명령을 수행한다. 지상에서는 마커의 변경을 통해서 실시간으로 미션변경을 할 수 있다. 이를 통해 드론은 제어용 송수신 채널을 통해서 통신을 하고는 있으나, 주파수 채널수가 제한이 되어 있으므로 구체적인 비행 제어 명령을 마커를 통해 이착륙시 추가적이며, 자동적인 진행이 가능하다.