• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.39-47
• /
• 2020

The fire protection system for high-rise buildings is currently confined to the preparation of sprinklers, emergency stairs, and exit and monitoring systems. On the other hand, an integrated system, including the model with scenario-based actions, is required for effective fire protection. An integrated fire protection system is needed to operate and manage the total cycle of the fire protection. In this study, an integrated fire protection system, which included sensing and consequent processes related to fire emergencies, was designed and implemented. The designed scheme can gather and analyze the data of the production, operation, and consumption patterns as it integrates fire protection systems for fire fighters and evacuating people. The integrated fire protection technology and system, which has target performance with satisfied 1/2 sec transaction response time and 1.2 transactions per second, is expected to contribute to market creation in converged technology-based fire protection fields.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.55-57
• /
• 2001

강화학습은 한 에이전트가 자신이 놓여진 환경으로부터의 보상을 최대화할 수 있는 최적의 행동 전략을 학습하는 것이다. 따라서 강화학습은 입력(상태)과 출력(행동)의 쌍으로 명확한 훈련 예들이 제공되는 교사 학습과는 다르다. 특히 Q-학습과 같은 비 모델 기반(model-free)의 강화학습은 사전에 환경에 대한 별다른 모델을 설정하거나 학습할 필요가 없으며 다양한 상태와 행동들을 충분히 자주 경험할 수만 있으면 최적의 행동전략에 도달할 수 있어 다양한 응용분야에 적용되고 있다. 하지만 실제 응용분야에서 Q-학습과 같은 강화학습이 겪는 최대의 문제는 큰 상태 공간을 갖는 문제의 경우에는 적절한 시간 내에 각 상태와 행동들에 대한 최적의 Q값에 수렴할 수 없어 효과를 거두기 어렵다는 점이다. 이런 문제점을 고려하여 본 논문에서는 로봇 축구 시뮬레이션 환경에서 각 선수 에이전트의 동적 위치 결정을 위해 효과적인 새로운 Q-학습 방법을 제안한다. 이 방법은 원래 문제의 상태공간을 몇 개의 작은 모듈들로 나누고 이들의 개별적인 Q-학습 결과를 단순히 결합하는 종래의 모듈화 Q-학습(Modular Q-Learning)을 개선하여, 보상에 끼친 각 모듈의 기여도에 따라 모듈들의 학습결과를 적응적으로 결합하는 방법이다. 이와 같은 적응적 중재에 기초한 모듈화 Q-학습법(Adaptive Mediation based Modular Q-Learning, AMMQL)은 종래의 모듈화 Q-학습법의 장점과 마찬가지로 큰 상태공간의 문제를 해결할 수 있을 뿐 아니라 보다 동적인 환경변화에 유연하게 적응하여 새로운 행동 전략을 학습할 수 있다는 장점을 추가로 가질 수 있다. 이러한 특성을 지닌 AMMQL 학습법은 로봇축구와 같이 끊임없이 실시간적으로 변화가 일어나는 다중 에이전트 환경에서 특히 높은 효과를 볼 수 있다. 본 논문에서는 AMMQL 학습방법의 개념을 소개하고, 로봇축구 에이전트의 동적 위치 결정을 위한 학습에 어떻게 이 학습방법을 적용할 수 있는지 세부 설계를 제시한다.

• Journal of Microbiology and Biotechnology
• /
• v.17 no.11
• /
• pp.1868-1874
• /
• 2007

We have developed a robotic system for an automated parallel cell cultivation process that enables screening of induction parameters for the soluble expression of recombinant protein. The system is designed for parallelized and simultaneous cultivation of up to 24 different types of cells or a single type of cell at 24 different conditions. Twenty-four culture vessels of about 200 ml are arranged in four columns${\times}$six rows. The system is equipped with four independent thermostated waterbaths, each of which accommodates six culture vessels. A two-channel liquid handler is attached in order to distribute medium from the reservoir to the culture vessels, to transfer seed or other reagents, and to take an aliquot from the growing cells. Cells in each vessel are agitated and aerated by sparging filtered air. We tested the system by growing Escherichia coli BL21(DE3) cells harboring a plasmid for a model protein, and used it in optimizing protein expression conditions by varying the induction temperature and the inducer concentration. The results revealed the usefulness of our custom-made cell cultivation robot in screening optimal conditions for the expression of soluble proteins.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.9 no.6
• /
• pp.612-619
• /
• 2004

Servo motor systems with ball-screw and timing-belt are widely used in NC, robot, FA and industrial applications. However, the nonlinear friction torque and damping effect in machine elements reduce the control performance. Especially tracking errors in trajectory control and very low velocity control range are serious due to the break-away friction and Stribeck effects. In this paper, a new double speed controller is proposed for compensation of the nonlinear friction torque. The proposed double speed controller has outer speed controller and inner friction torque compensator. The proposed friction torque compensator compensates the nonlinear friction torque with actual speed and speed error information. Due to the actual information for friction torque compensator without parameters and mathematical model of motor, proposed compensator is very simple structure and the stability is very high. The proposed compensator is verified by simulation and experimental results.

• Korean Journal of Construction Engineering and Management
• /
• v.10 no.1
• /
• pp.136-145
• /
• 2009

The objective of the Intelligent Excavation System (IES) is to recognize the work environment and produce work plan and automatically control the excavator through integrating sensor and robot technologies. This paper discusses one of the core technologies of IES development project, development of 3D work environment modeling. 3D laser scanner is used for 3-dimensional mathematical model that can be visualized in virtual space in 3D. This paper describes (1) how the most appropriate 3D imaging system has been chosen; (2) the development of user interface and customization of the s/w to control the scanner for IES project; (3) the development of the mobile station for the scanner; (4) and the algorithm for the automatic registration of laser scan segments for IES project. The development system has been tested on the construction field and lessons learned and future development requirements are suggested.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.4
• /
• pp.37-45
• /
• 2000

In this paper we suggest a method of detecting the two eyes position of moving viewer by using images obtained through a color CCD camera, and also a method of rendering view-dependent 3D image which consists of depth estimation, image-based 3D object modeling and stereoscopic display process. Through the experiment of applying the suggested methods, we could find the accurate two-eyes position with the success rate of 97.5% within the processing time of 0.39 second using personal computer, and display the view-dependent 3D image using Fl6 flight model. And through the similarity measurement of stereo image rendered at z-buffer by Open Inventor and captured by stereo camera using robot, we could find that view-dependent 3D picture obtained by our proposed method is optimal to viewer.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.817-822
• /
• 2005

Recently tile study that exploits visual information for tile services of robot in indoor environments is active. Conventional image processing approaches are based on the pre-defined geometric models, so their performances are likely to decrease when they are applied to the uncertain and dynamic environments. For this, diverse researches to manage the uncertainty based on the knowledge for improving image recognition performance have been doing. In this paper we propose a Bayesian network modeling method for predicting the existence of target objects when they are occluded by other ones for improving the object detection performance of the service robots. The proposed method makes object relationship, so that it allows to predict the target object through observed ones. For this, we define the design method for small size Bayesian networks (primitive Bayesian netqork), and allow to integrate them following to the situations. The experiments are performed for verifying the performance of constructed model, and they shows $82.8\%$ of accuracy in 5 places.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.4
• /
• pp.437-442
• /
• 2012

To analyze a multibody system, this paper proposes an implicit numerical integration method for joint coordinates subsystem synthesis method. To verify the proposed method, a multibody model for an unmanned robot vehicle, which consists of six identical independent suspension systems, is developed. The symbolic method is applied to compute the system Jacobian matrix for the implicit integration method. The proposed method is also verified by performing rough terrain run-over simulation in comparison with the conventional implicit integration method. In addition, to evaluate the efficiency of the proposed method, the CPU time obtained by using this method is compared with that obtained by using the conventional implicit method.

• Health Policy and Management
• /
• v.25 no.2
• /
• pp.67-70
• /
• 2015

To establish an appropriate policy for robotic surgery in Korea, the National Evidence-based Collaborating Agency (NECA) and the Korean Society of Health Policy and Administration held a round-table conference (RTC) to gather opinions through a comprehensive discussion of scientific information in gastric cancer. The NECA RTC is a public discussion forum wherein experts from diverse fields and members of the lay public conduct in-depth discussions on a selected social issue in the health and medical field. For this study, representatives from the medical field, patient groups, industry, the press, and policy makers participated in a discussion focused on the medical and scientific evidence for the use of robotic surgery in gastric cancer. According to the RTC results, robotic surgery showed more favorable results in safety and efficacy than open surgery and it is similar to laparoscopy. When the cost-effectiveness of robotic surgery and laparoscopy is compared, robotic surgery costs are higher but there was no difference between the two of them in terms of effectiveness (pain, quality of life, complications, etc.). In order to resolve the high cost issue of the robotic surgery, a proper policy should be implemented to facilitate the development of a cost-effective model of the robotic surgery equipment. The higher cost of robotic surgery require more evidence of its safety and efficacy as well as the cost-effectiveness issues of this method. Discussions on the national insurance coverage of robotic surgery seems to be necessary in the near future.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.1
• /
• pp.55-60
• /
• 2011

The most basic algorithm in SLAM(Simultaneous Localization And Mapping) technique of mobile robots is EKF(Extended Kalman Filter) SLAM. However, it requires prior information of characteristics of the system and the noise model which cannot be estimated in accurate. By this limit, Kalman Filter shows the following behaviors in a highly uncertain environment: becomes too sensitive to internal parameters, mathematical consistency is not kept, or yields a wrong estimation result. In contrast, $H_{\infty}$ filter does not requires a prior information in detail. Thus, based on a idea that $H_{\infty}$ filter based SLAM will be more robust than the EKF-SLAM, we propose a framework of $H_{\infty}$ filter based SLAM and show that suggested algorithm shows slightly better result man me EKF-SLAM in a highly uncertain environment.