• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.6_1
• /
• pp.953-959
• /
• 2022

In this paper, the design result of robotic tools and the development of robot control system for chemical coupler assembly are presented. This research aims to eliminate the risk of chemicals exposed to human operators by developing the robotic tools and robot automation system for chemical tank lorry unloading that were done manually. Due to tight tolerance between couplers, even small pose error may result in very large internal force. In order to resolve the problem, the 6-axis compliance device is employed, which can provide not only enough compliance between couplers but also F/T sensing. The 6-axis compliance device having large force and moment capacity is designed. A simple linear gripper with rack-and-pinion is designed to grasp two sizes of couplers. The proposed robot automation system consists of 6-DOF collaborative robot with offset wrist, 6-axis compliance device with F/T sensing, linear gripper, and two robot visions.

• Architectural research
• /
• v.12 no.2
• /
• pp.85-92
• /
• 2010

In recent years, construction projects have been forced to cope with lack of skilled labor and increasing hazard circumstance of human operations. A construction robotic system has been frequently accomplished as one alterative for overcoming these difficulties in increasing construction quality, enhancing productivity, and improving safety. However, while the complexity of such a system increases, there are few ways to carry out an assessment of the system. This paper introduces a knowledge-based multi-criteria decision-making process to assist decision makers in systematically evaluating an automated system for a given project and quantifying its system performance index. The model employs linguistic terms and fuzzy numbers in attempts to deal with the vagueness inherent in experts' or decision makers' subjective opinions, considering the contribution resulted from their knowledge on a decision problem. As an illustrative case, the system, called Robotic-based Construction Automation, for constructing steel erection of high-rise buildings was applied into this model. The results show the model's capacities and imply the application to other extended types of construction robotic systems.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2009.05a
• /
• pp.561-562
• /
• 2009

The needs for inspection automation for more systematic and efficient maintenance were gradually increased by several inspectors and researchers. With the robotic and digital image processing technologies, in this paper, new inspection automation system were introduced and tested in the real PSC box crack inspection procedures The configuration and scheme of robotic inspection and digital image processing algorithms were represented. The designed robotic sensors and image processing system were tested and the feasibility and possibility of the robot based automatic inspection were approved in the real PSC box bridges.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1511-1515
• /
• 2004

Machine vision system plays an important role in factory automation. Its many applications are found in automobile manufacturing industries, as an eye for robotic automation system. In this paper, an automobile position measuring vision system(APMVS) applicable to manufacturing line for under body painting of a car is introduced. The APMVS measures position and orientation of the car body to be sealed or painted by the robots. The configuration of the overall robotic sealing/painting system, design and application procedure, and application examples are described.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.8
• /
• pp.930-935
• /
• 2019

Our proposed system, MobiAutoBot, is a conceptual model of robotics process automation software at a cognitive automation level that can support mobile devices. MobiAutoBot consists of two parts: MobiAutoBot controller and MobiAutoBot runner. The MobiAutoBot controller directs, monitors, and interacts the Job on the mobile device, and the MobiAutoBot runner installed on the mobile device performs the commanded operation. If we provide automation for mobile devices to small businesses through our proposed MobiAutoBot, we can build low-cost robotic process automation functions that can be linked to existing information systems with mobile devices such as simple smartphones, even without an expensive information-based infrastructure. Our proposal is expected to serve as an opportunity to spread automation of robotics processes even to small and medium-sized companies and individual users who are difficult to equip with all information system infrastructure.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.3
• /
• pp.398-405
• /
• 2004

A human-friendly interactive system that is based on the harmonious symbiotic coexistence of humans and robots is explored. Based on the interactive technology paradigm, a robotic cane is proposed for blind or visually impaired pedestrians to navigate safely and quickly through obstacles and other hazards. Robotic aids, such as robotic canes, require cooperation between humans and robots. Various methods for implementing the appropriate cooperative recognition, planning, and acting, have been investigated. The issues discussed include the interaction between humans and robots, design issues of an interactive robotic cane, and behavior arbitration methodologies for navigation planning.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2008.05a
• /
• pp.111-115
• /
• 2008

The productivity is the one of the primary elements to evaluate the new systems performance for building construction along with the economic efficiency, the safety, and the quality improvement. At present the on-site construction automation systems(we will call it as Construction Factory or CF hereafter) is under the development by research group for construction automation of high-rise building. The system includes many sub-systems such as a robotic crane, a construction factory, bolting robots, a building material management system using RFID and so on. In this study we discuss and propose the method to evaluate for these hybrid on-site automation system fundamentally. In future we devise a framework of evaluation modules for the on-site building automation system on the basis of this discuss.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.295-296
• /
• 2020

본 논문에서는 RPA(Robotic Process Automation) Tool 개발 과정 중 머신 러닝, 딥러닝에 필요한 이미지 크롤링 및 전처리 기능을 이용한 가공된 데이터 셋 처리 과정을 기술한다. 개발된 RPA 툴에서 머신러닝 및 딥러닝에 사용될 데이터 확보 기능을 제공하며, 세부적으로 이미지 전처리(Convert Gray, Histogram Equalization, Binary, Resize)등 반복적으로 사용되는 기능들을 제공한다. 개발된 툴을 통해 RPA의 자동화 기능과, 전처리 기능의 융합을 통해 업무의 효율성을 제공한다.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2010.05a
• /
• pp.33-36
• /
• 2010

The attention in construction Automation is getting higher since it could be the answer to the lack of skilled labor by decrease in construction population and aging which adversely affects productivity and quality in the construction site. We are on the way to develop a construction automation system adequate for domestic circumstances in Korea; it is called RCA(Robotic-crane based Construction Automation)system. Climbing hydraulic robots system is a part of RCA system and makes Construction factory(CF) climb through the guide rail on the core wall. The safety of climbing hydraulic robots system is at issue due to the overloaded weight of CF. Preventing this issue, present study did the design verification through the structural analysis and the simulation. Mock-up test also was done to analyze the drive performance of climbing hydraulic robots system.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.543-545
• /
• 1998

Robotic soccer is multi-agent system playing soccer game under given rule. This system consists of three mobile robots, vision sensor, action decision module, action execution module and communication module. This paper presents new action decision method using multi-layer neural networks.