• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2008.10a
• /
• pp.359-363
• /
• 2008

In recent warfare, importance of counter-fire is increased. New weapon system and counter-fire have critical impact on defeating of the enemy. At an initial battle counter-fire, we need to study about striking plan against the enemy mine artillery. In this paper, we studied, using MANA(agent based model), how much cannon's hit probability and UAV-based target acquisition have influence on striking the mine artillery. We constructed MANA model based on the characteristics of a regional database of Korean peninsula. If we develop detail database of the ROK Army, the proposed MANA model can be used as a war-game model for the regimental level of the Army.

• Proceedings of the KIEE Conference
• /
• 2006.11a
• /
• pp.219-221
• /
• 2006

This paper represent the algorism for analysis effect of the long-term contract to reduce market power of pivotal supplier using Agent Based model.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.329-332
• /
• 2003

현재의 일반적인 다중에이전트 시스템 환경은 실시간이며 복잡한 환경을 제공한다. 또한 제한적인 통신 환경을 제공한다. 본 논문의 테스트 환경인 Unreal Tournament 의 환경은 일반적인 다중 에이전트 시스템 환경을 제공한다. UT 게임의 GameBots 시스템에서 실시간 다중 에이전트 협상 시스템의 구현을 위한 ACL (Agent Communication Language) 을 정의하였으며 그에 따른 다중 에이전트 협상 프로토콜을 정의하였다. 통신 환경은 단일 채널 환경이며 제한적인 통신을 제공한다. 에이전트들은 게임 시작과 동시에 인지 정보를 기반으로 맴을 작성하게 된다. UT 게임은 환경이 실시간으로 급변하기 때문에 최단의 협상 과정을 가져야 한다. 협상 시스템의 구성은 협상 과정에서 모든 것을 정하기엔 시간이 부족하기 때문에 빠른 협상 과정을 유도하기 위하여 협상과정의 일부분을 사전에 정의함으로써 협상과정을 단순화 시켰으며, 나머지는 실시간 협상과정을 통하여 동적으로 역할 분담을 하였다 협상 방법으로는 각 에이전트의 의견이 반영될 수 있는 투표 (voting) 방법을 사용하였다.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.3024-3026
• /
• 1999

This paper presents a cooperative action controller of a multi-agent system. To achieve an object, i.e. win a game, it is necessary that a robot has its own roles, actions and work with each other. The presented incorporated action controller consists of the role selection, action selection and execution layer. In the first layer, a fuzzy logic controller is used. Each robot selects its own action and makes its own path trajectory in the second layer. In the third layer, each robot performs their own action based on the velocity information which is sent from main computer. Finally, simulation shows that each robot selects proper roles and incorporates actions by the proposed controller.

• Journal of Information Processing Systems
• /
• v.1 no.1 s.1
• /
• pp.107-111
• /
• 2005

When a pedagogical agent system aims to provide students with interactive help, it needs to know what knowledge the student has and what goals the student is currently trying to achieve. That is, it must do both assessment and plan recognition. These modeling tasks involve a high level of uncertainty when students are allowed to follow various lines of reasoning and are not required to show all their reasoning explicitly. In this paper, the student model for interactive edutainment applications is proposed. This model is based on Bayesian Networks to expose constructs and parameters of rules and propositions pertaining to game and problem solving activities. This student model could be utilized as the emotion generation model for student and agent as well.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2003.11a
• /
• pp.329-333
• /
• 2003

This study presents supplier buyer models representing the interactions between supplier and buyer under a long-term replenishment contract in a supply chain system. We established the models according to the economic power of each party. Analysis based on Stackelberg game theoretic approach is tried for each model. We develop methods for each agent to follow to complete a contract for the best interest of each participant.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1999.05a
• /
• pp.154-162
• /
• 1999

This paper introduces the research and development of automatic generation technology to develop the character agent. The R&D of this technology includes three major elements-body model generation, automatic motion generation and synthetic human generation. Main areas of application would be cyber space- 3D game, animation, virtual shopping, on line chatting, virtual education system, simulation and security system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.636-639
• /
• 2016

최근 모바일 게임에서의 인공지능과 관련된 연구가 활발히 진행되고 있다. 본 논문에서는 모바일 축구 시뮬레이션 게임에서 활용할 수 있는 인공지능 에이전트를 Hierarchical FSM 기반으로 설계하고 구현하여 실제 축구경기 결과와 비슷한 결과 도출하였다. 이러한 Hierarchical FSM을 기반으로 한 지능형 에이전트는 코드의 재활용성이 높고 개념적으로 간단하여 인공지능 에이전트를 설계 및 구현하기에 적합하다.

• 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.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.321-324
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement Beaming is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement loaming is different from supervised teaming in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement loaming algorithms like Q-learning do not require defining or loaming any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state-action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem, we suggest Adaptive Mediation-based Modular Q-Learning(AMMQL) as an improvement of the existing Modular Q-Learning(MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state space effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.