• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.45-49
• /
• 2006

홈네트워크 시스템의 본격적인 보급과 함께 가정용서비스 로봇의 최근 연구 성과들은 인간과 지능로봇이 가정에 공존하며 서로 의사소통을 할 수 있는 시대가 가까운 미래에 현실화 될 수 있음을 보여주고 있다. 그러나 가정의 환경적인 특징은 open되어 있기 때문에 그러한 환경에 적응하고 주어진 임무를 수행하는 데는 단일 로봇 또는 단일 홈서버 보다는 로봇을 포함하는 홈네트워크 시스템 내의 여러 장치들이 어울려 분산처리를 수행하는 multi-agent 시스템이 일반적으로 더 좋다고 알려져 있다. 따라서 본 논문은 홈네트워크 시스템 환경에서 가정에서 필요한 agent들을 정의하기 위한 framework 모델을 구축하고 각 agent 간의 통신 protocol architecture를 제시한다. 또한 로봇 또는 홈서버의 단일 지능이나 기능보다는 그 안에 존재하는 복수개의 agent instance들의 집합으로 agent를 정의하고 각 agent 내외에서 agent들 사이의 협력(cooperation)과 (타협)negotiation을 통해 환경과 적응하는 방법 및 사람과 교감(interactive)하는 방법을 제시한다.

• Journal of Korean Association for Spatial Structures
• /
• v.23 no.2
• /
• pp.29-36
• /
• 2023

Reinforcement learning (RL) is widely applied to various engineering fields. Especially, RL has shown successful performance for control problems, such as vehicles, robotics, and active structural control system. However, little research on application of RL to optimal structural design has conducted to date. In this study, the possibility of application of RL to structural design of reinforced concrete (RC) beam was investigated. The example of RC beam structural design problem introduced in previous study was used for comparative study. Deep q-network (DQN) is a famous RL algorithm presenting good performance in the discrete action space and thus it was used in this study. The action of DQN agent is required to represent design variables of RC beam. However, the number of design variables of RC beam is too many to represent by the action of conventional DQN. To solve this problem, multi-agent DQN was used in this study. For more effective reinforcement learning process, DDQN (Double Q-Learning) that is an advanced version of a conventional DQN was employed. The multi-agent of DDQN was trained for optimal structural design of RC beam to satisfy American Concrete Institute (318) without any hand-labeled dataset. Five agents of DDQN provides actions for beam with, beam depth, main rebar size, number of main rebar, and shear stirrup size, respectively. Five agents of DDQN were trained for 10,000 episodes and the performance of the multi-agent of DDQN was evaluated with 100 test design cases. This study shows that the multi-agent DDQN algorithm can provide successfully structural design results of RC beam.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.896-902
• /
• 2004

The purpose of the study of multi-robot system is to realize multi-robot system easy for the control of robot system in case robot is adapted in the complicated environment of task structure. The purpose of the study of multi-robot system is to realize multi-robot system easy for the control of robot system in case robot is adapted in the complicated environment of task structure. To make real time control possible by making effective use of recognized information in this dynamic environment, suitable distribution of tasks should be made in consideration of function and role of each performing robots. In this paper, IHMA (Intelligent Hybrid Modular Architecture) of Intelligent combined control architecture which utilizes the merits of deliberative and reactive controllers will be suggested and its efficiency will be evaluated through the adaptation of control architecture to representative multi-robot system.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.703-704
• /
• 2008

Multi-agent technologies are essential in realizing the upcoming ubiquitous environment. In the multi-agent environment, each agent has its own set of services and stores these services in the service repository of the multi-agent system. By using this repository, the user can retrieve the most appropriate service. In this paper, we propose an efficient service repository architecture that can improve the existing agent-based service discovery.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10a
• /
• pp.355-357
• /
• 2000

실세계에서 발생하는 복잡한 문제들을 해결하기 위한 노력으로, 다중 에이전트 시스템)Multi-Agent System) 구축에 대한 관심이 높아지고 있다. 다양한 종류의 분산 인공지능 문제들을 에이전트라는 추상적 단위와 에이전트간의 상호작용을 토대로 해결하는 시스템을 개발하기 위하여, 본 연구에서는 다중 에이전트 지향의 소프트웨어를 개발함에 있어 중요한 요소인 조정(Coordination)을 지원하는 아키텍쳐를 제안한다. 문제영역을 분석하고, 다중 에이전트 시스템의 특성을 파악하여 시스템 요소들의 조정을 지원하는 아키텍쳐 공정을 제안한다. 또한, 이를 지능형 교통정보 시스템에 적용하여 본다.

• Journal of KIISE:Software and Applications
• /
• v.32 no.11
• /
• pp.1047-1058
• /
• 2005

In this paper we propose an agent architecture to improve the flexibility of videoconference systems with strategic-INTER knowledge. The proposed architecture achieves more flexibility by selecting dynamically the strategy of QoS (Quality of Service). To select the strategies, the systems have to take into the consideration of the properties of problems occurred on QoS and the status of problem solving process. This architecture is introduced as a part of knowledge base of agent dealing with cooperation between software modules of videoconference systems. We have implemented the architecture and our prototype system shows its capability of flexible problem solving against the QoS degradation, along with other possible problems within the time limitation. We confirmed that the proposed architecture can improve its flexibility of a videoconference system compared to conventional systems.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.12
• /
• pp.3477-3489
• /
• 1999

Our study is focused on a multi-agent system which provides the efficient teleconferencing facility with the help of the intelligent agents who allow the participants to minimize the effort to call meeting. Our system has multi-agent architecture and use the KQML to communicate among agents. Communicating through the KQML allows us to overcome the heterogeneity when we add new agents into the system. In this paper, we propose a multi-agent structure for the teleconferencing system based on the web and describe in detail the now or the KQML messages among agents.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.05a
• /
• pp.1124-1128
• /
• 2005

In this paper, a plan-coordination architecture is proposed for multi-agent control in the fractal manufacturing system (FrMS). A fractal in FrMS is a set of distributed agents whose goal can be achieved through cooperation, coordination, and negotiation with other agents. Since each agent in the FrMS generates, achieves, and modifies its own plan fragments autonomously during the coordination process with other agents, it is necessary to develop a systematic methodology for the achievement of global plan in the manufacturing system. The heterarchical structure of the FrMS provides a compromised plan-coordination approach, it compromise a centralized plan-generation/execution (which mainly focuses on the maximization of throughput) with a distributed one (which focuses on the autonomy of each module and flexibility of the whole system). Plan-coordinators in lower level fractal independently generate plan fragments according to the global plan of higher level fractal, and plan-coordinators in higher level fractal mediate/coordinate the plan fragments to enhance the global performance of the system. This paper assumes that generation method of the plan fragments and the negotiation policy of the fractal is achieved by a simple process, and we mainly focuses on the information exchanging and distributed decision making process to coordinate the combinations of plan fragments within a limited exchange of information.

• Journal of the Korea Society for Simulation
• /
• v.18 no.4
• /
• pp.117-125
• /
• 2009

Recently the intelligent agent systems have been emerged as one of key issues for developing the defense M&S systems. However, most conventional agent architecture of M&S systems utilize the script-based models and can only deal with the individual behaviors so that they cannot suitably describe the precise tactical/strategic behavior and/or complex warfare environment. To overcome these problems, we have proposed the hierarchical multi-agent system architecture that is able to intelligently cope with the complex missions based on the functional role of each agent on the hierarchy such as an intelligence officer, captain, warship commander. Several simulation tests performed on 2:2 warship warfare models will illustrate our techniques.

• Journal of Internet Computing and Services
• /
• v.9 no.3
• /
• pp.9-24
• /
• 2008

In this paper, we describes the design of software architecture supporting for healthcare context information service platform based on multi agent in home environment. In this platform, the DOGF supports the execute object and healthcare sensors and device's logical services grouping. JADE framework can support mobility in heterogeneous environment. The multi agents on platform order to support a healthcare context information service it will be able to divide. An agent collects an environment information from distributed devices. Another an agent follows mobile-device specific and it does a different service. And an agent where it manages like this. The mobile-proxy&agent is an interface part between DOGF and JADE, support data interchange or mobility pattern. For DOGF and JADE to provide healthcare context information service, we describes the design of multi agent software platform and multi agent classification by services. Finally we showed the system environments which is physical environments and prototype based on platfonn for healthcare context information services.