Yong-Jun You

Many-to-Many Warship Combat Tactics Generation Methodology Using the Evolutionary Simulation

In most existing warships combat simulation system, the tactics of a warship is manipulated by human operators. For this reason, the simulation results are restricted due to the stereotype of human operators. To deal with this, we have employed the genetic algorithm for supporting the evolutionary simulation environment. In which, the tactical decision by human operators is replaced by the human model with a rule-based chromosome for representing tactics so that the population of simulations are created and hundreds of simulation runs are continued on the basis of the genetic algorithm without any human intervention until to find emergent tactics which shows the best performance throughout the simulation. This paper proposes an evolutionary tactics generation methodology for the emergent tactics in many-to-many warship combat simulation. To do this, 3:3 warship combat simulation tests are performed.

The Rule-based Agent Modeling and Simulation considering the Evacuation Behavior Characteristics on the Passenger Ship Fire

ABSTRACTThis paper suggests the passenger model considered evacuation behavioral characteristics on the passenger ship fire using a rules-based agent technique. The existing evacuation simulation system was modeled only passenger speed. The speed-based model considered passenger’s physical characteristics, so it couldn’t consider evacuation behavioral characteristics. For solving this problem, we modeled the passenger model using a rule-based agent applied evacuation behavioral characteristics. The rule-based agent consists of knowledge base and inference engine. In knowledge base, we represented evacuation behavioral characteristics, and chose the examples of the evacuation behavioral characteristics to show various patterns of behavior. And we simulated in the IMO MSC/Circ.1238 example 8 and we proved the simulation results could represent variety patterns of human behavior. Key words : Rule-based Agent, Modeling and Simulation, Evacuation Behavior Characteristics요 약본 논문은 규칙기반 에이전트를 이용하여 여객선 화재시 피난행동특성을 고려한 승객행동모델을 제안한다 . 기존 시스템들은 승객의 속도만을 반영한 속도기반모델을 사용하였다 . 속도기반모델은 승객의 신체적 특징만 고려한 모델로 피난행동특성을 반영하기 어렵다. 이 문제점을 해결하기 위해 규칙기반 에이전트로 피난행동특성을 반영한 승객모델을 모델링하였다 . 규칙기반 에이전트는 지식베이스와 추론엔진으로 구성된다. 지식베이스에는 피난행동특성들 중에서 다양한 행동 패턴을 보여줄 수 있는 예제를 선택하여 규칙형태의 지식으로 표현하고 , 추론엔진은 지식을 전방추론하여 승객의 행위를 결정하도록 구현하였다 . 이 승객모델을 IMO MSC/Circ.1238의 문제 8에 적용하여 시뮬레이션을 진행한 결과 규칙기반 에이전트로 모델링한 승객이 다양한 인간의 행동 패턴을 표현할 수 있음을 검증하였다 .주요어 : 규칙기반 에이전트., 모델링 & 시뮬레이션, 피난행동특성

A Study on the Requirements for Designing Agent-Based Computer Generated Force

CGF (Computer Generated Forces) is known as a representation of constructive forces in a simulation environment which attempts to model human behavior. Currently, CGF deals with some extent on lower levels of command as well as weapon system, however, their behavior is stereotyped so as to support only simple level of decision-making. In this paper, we propose the level of agent-based CGF by integrating various viewpoints of requirements for supporting the systematic design.