유전자 알고리즘을 이용한 경유횟수 제약이 있는 최단 경로 탐색

이면섭 한국지식정보기술학회 2017 한국지식정보기술학회 논문지 Vol.12 No.2

The major role of intelligent character in an artificial intelligence game is to be smarter and to add more interest on character behaviors. Moreover, to apply artificial intelligence on behaviors and movements of intelligent characters, the time to process artificial intelligence should be shortened because even without the process of artificial intelligence, it takes time to visualize and print out a character with intelligence. In this study, the method that searching a spanning tree with the lowest cost which is limited in a number of transfer by a using genetic algorithm is suggested. In genetic algorithm, a roulette selection was chosen to be the selection method, and hybridization method of uniform crossover was applied. The elite conservation strategy for throwing back the optimal parameter to the following generation was applied to shorten the searching time. The Dandelion Code, an effective method to encode a tree in the evolution algorithm, was used. As a result of experiment, when the definite number was 50, 60, 70, and 80, the number of transfer limitation was 3 and 5 to compare the cost and time of spanning tree. When the definite number was 80 and the number of transfer was 5, the route was searched 0.161 seconds earlier.

대전형 액션 게임에서 에너지 점수를 도입한 지능 캐릭터

이면섭,조병헌,정성훈,성영락,오하령 한국정보처리학회 2006 정보처리학회논문지. 소프트웨어 및 데이터 공학 Vol.13 No.4

This paper proposes intelligent characters for fighting action games to which energy points are applied for more realistic implementation than those of previous researches. The intelligent characters decide their actions in consideration of their energy level as well as a current action, the step of the action, the distance, and past actions of opponent characters that were used in existing intelligent ones. We used two types of energy, HP(Health Point) and MP(Mana Point) that were frequently employed in recent on-line games. We experimented with proposed intelligent characters to investigate whether the intelligent characters learn proper actions and cope with opponent characters in consideration of their energy levels. Experimental results showed that the intelligent characters reacted with the best actions to obtain high score if their energy is sufficient, Otherwise, they did the actions to(that?) recharge their energy. From this observation, we could conclude that the proposed intelligent characters worked well and did effective actions in consideration of the their energy. 본 논문에서는 대전형 액션 게임을 위한 지능 캐릭터 구현에 있어서 기존 연구보다 실제적인 구현을 위하여 에너지 점수를 적용한 지능 캐릭터를 제안한다. 제안한 지능 캐릭터는 기존연구에서 사용한 상대 캐릭터의 현재 행동, 행동의 단계, 유효거리, 과거 행동 뿐 만 아니라 자신의 에너지 양을 고려하여 행동을 결정하게 된다. 에너지는 현재 온라인 게임 등에서 많이 사용하는 것으로 HP (Health Point)와 MP(Mana Point) 두 가지를 사용하였다. 에너지 점수를 적용한 지능캐릭터가 학습하는지 그리고 에너지점수를 적용했을 때 어떻게 행동하는지를 알아보기 위한 여러 가지 실험을 하였다. 실험 결과 에너지가 양이 많은 경우에는 상대방에게 피해를 크게 줄 수 있는 행동으로 대응하고 그렇지 않은 경우에는 에너지를 재충전하기 위한 행동들을 하는 것을 관찰하였다. 이런 결과로 볼 때 본 논문에서 제안한 에너지를 적용한 지능캐릭터가 잘 동작하며 에너지를 고려하여 효과적인 반응을 보임을 알 수 있다.

Effect of Parameters of Genetic Algorithm on Results in Action Games

이면섭 한국지식정보기술학회 2015 한국지식정보기술학회 논문지 Vol.10 No.6

In this paper, it was intended to identify effect of parameter values of genetic algorithm on results. As the study subject, an action game with 10 actions was used. In the action game, intelligent characters were set to find optimal solutions by evolving by themselves along with repetition of generations. Among several parameters of the genetic algorithm, population size, cross rate, and mutation rate were changed and applied. The range of population size was set to be from 50 to 950, and the cross rate and the mutation rate were applied, changing from 0.7% to 1.0% and from 0.005% to 0.24% respectively. As the results, it was achieved to find optimal solution after 90 generations. It was found that the combination of parameters to identify optimal solution within the most rapid time were 300 in population size, 1 in cross rate, and 0.01 in mutation rate. Because this combination of parameters cannot be found with some trials, there are some difficulties that a number of repeated experiments with various combinations of parameters are required

유전자 알고리즘을 이용한 대전형 액션게임의 지능캐릭터

이면섭,조병헌,성영락,정성훈,오하령 한국정보처리학회 2005 정보처리학회논문지. 소프트웨어 및 데이터 공학 Vol.12 No.3

This paper proposes a method to provide intelligence for characters in fighting action games by using genetic algorithm. The proposed characters without any knowledge on the rules of the game learn the rules and techniques for generations, and have the capability of evolving. To evaluate adaptability for varying circumstances, we changed the rules and compared the results. The experimental results show that the intelligent characters can adapt to the new rules. An advantage of the proposed method is that it can be easily applied to characters for other category of games such as PC games and internet online games. 본 논문에서는 유전자 알고리즘을 이용하여 대전형 액션 게임의 캐릭터를 지능화 하는 방법을 제안한다. 초기에 게임 규칙을 전혀 모르는 캐릭터가 학습이 진행되면 세대를 거듭하면서 스스로 게임규칙을 학습하고 기술을 습득하며 진화할 수 있는 지능을 갖추게 된다. 제안한 캐릭터가 환경변화에 잘 적응하는지를 평가하기 위하여 게임 도중에 게임규칙을 바꾸어 규칙 변경 전후의 결과를 비교 검토하였다. 실험결과 게임 규칙이 변경된 후에도 빠르게 새로운 규칙에 적응하였다. 본 논문에서 제안한 방법은 대전형 액션게임뿐만 아니라 PC게임이나 인터넷 온라인 게임 등의 캐릭터 구현에도 쉽게 적용가능한 장점이 있다.

가온요법을 위한 암조직 세포의 무침습 온도분포 추정에 관한 기초연구

이면섭,김낙환,홍승홍,Lee, Myeon-Seop,Kim, Nak-Hwan,Hong, Seung-Hong 대한의용생체공학회 1986 의공학회지 Vol.7 No.1

In the clincal application of Hyperthermia, tissue temperature measurements are made at only a few selecte4 locations because of patient tolerance and practical clinical limitation. Since it is necessary to know the complete carcinoma cell temperature field in order to treat effectively, The difficulty of making such estimates from only a few point compounded because of the lack of knowledge of the carcinomoma cell blood perfusion characteristics. To solve the temperature on carcinoma cell, A noninvasive method (the finite element method) is used. The simulation results show that the finite element method is promising for estimating the complete corcinoma cell temerature distribution, if some knowledge of the blood perfusion pattern is available.

암세포 조직의 온도분포 추정에 관한 연구 ( 1 ) ( A Study on the Estimation of Temperature Distribution in Tumor ( 1 ) )

이면섭,김기욱,이은실,홍승홍 대한전자공학회 1985 대한전자공학회 학술대회 Vol.1985 No.1

Hyperthermia의 온도 측정시스템 선형화 ( The Linearization of Temperature Measurement System in Hyperthermia )

이면섭,김낙훈,홍승홍 대한전자공학회 1986 대한전자공학회 학술대회 Vol.1986 No.1

유전자 알고리즘을 이용한 대전형 액션게임의 지능캐릭터

이면섭,조병헌,성영락,정성훈,오하령,Lee Myun-sub,Cho Byeong-heon,Seong Yeong-rak,Jung Sung-hoon,Oh Ha-ryoung 한국정보처리학회 2005 정보처리학회논문지B Vol.12 No.3

본 논문에서는 유전자 알고리즘을 이용하여 대전형 액션 게임의 캐릭터를 지능화 하는 방법을 제안한다. 초기에 게임 규칙을 전혀 모르는 캐릭터가 학습이 진행되면 세대를 거듭하면서 스스로 게임규칙을 학습하고 기술을 습득하며 진화할 수 있는 지능을 갖추게 된다. 제안한 캐릭터가 환경변화에 잘 적응하는지를 평가하기 위하여 게임 도중에 게임규칙을 바꾸어 규칙 변경 전후의 결과를 비교 검토하였다. 실험결과 게임규칙이 변경된 후에도 빠르게 새로운 규칙에 적응하였다. 본 논문에서 제안한 방법은 대전형 액션게임뿐만 아니라 PC게임이나 인터넷 온라인 게임 등의 캐릭터 구현에도 쉽게 적용가능한 장점이 있다. This paper proposes a method to provide intelligence for characters in fighting action games by using genetic algorithm. The proposed characters without any knowledge on the rules of the game learn the rules and techniques for generations, and have the capability of evolving. To evaluate adaptability for varying circumstances, we changed the rules and compared the results. The experimental results show that the intelligent characters can adapt to the new rules. An advantage of the proposed method is that it can be easily applied to characters for other category of games such as PC games and internet online games.

상대캐릭터의 행동패턴에 적응하는 지능캐릭터의 구현

이면섭,조병헌,정성훈,성영락,오하령,Lee, Myun-Sub,Cho, Byeong-Heon,Jung, Sung-Hoon,Seong, Yeong-Rak,Oh, Ha-Ryoung 대한전자공학회 2005 電子工學會論文誌 IE (Industry electronics) Vol.42 No.3

본 논문에서는 유전자 알고리즘을 이용하여 특정 행동 패턴을 보이는 상대 캐릭터에 적절히 적응할 수 있는 대전 액션 게임용 지능캐릭터의 구현 방법을 제안한다. 이를 위하여 현재 행동과 단계뿐 만 아니라 과거행동까지 고려하여 학습한다. 제안한 방법이 얼마나 효과가 있는지 알아보기 위하여 현재의 행동과 단계만을 적용한 실험(실험-1)과 과거 행동을 추가한 행동패턴을 적용한 실험(실험-2) 결과를 비교 분석하였다. 실험 평가는 두 캐릭터가 획득한 점수를 측정하여 그 비로써 평가하였다. 실험 결과 초기에는 실험-1에서 높은 점수비로 시작하지만 일정 세대 이후부터는 실험-2의 점수비가 좋아지며, 실험-2에서는 모두 최적해를 찾을 수 있었다. 또한 실험-2에서 지능 캐릭터는 행동이 완료되는 시점에서 최대의 점수를 얻기 위해 이동(전진,후진)이나 시간 지연 동작을 하여 스스로 진화하면서 게임 규칙을 학습함을 보았다. This paper proposes an implementation method of intelligent characters that can properly adapt to action patterns of opponent characters in fighting games by using genetic algorithm. For this intelligent characters, past actions patterns of opponent characters should be included in the learning process. To verify the effectiveness of the proposed method, two types of experiments are performed and their results are compared. In first experiment(exp-1), intelligent characters consider current action and its step of an opponent character. In second experiment (exp-2), on the other hands, they take past actions of an opponent characters into account additionally. As a performance index, the ratio of score obtained by an intelligent character to that of an opponent character is adopted. Experimental results shows that even if the performance index of exp-1 is better than that of exp-2 at the beginning of stages, but the performance index of exp-2 outperforms that of exp-1 as stages go on. Moreover, optimum solutions are always found in all experimental cases in exp-2. Futhermore, intelligent characters in exp-2 could learn moving actions (forward and backward) and waiting actions for getting more scores through self evolution.

대전 액션 게임에서 유전자 알고리즘을 이용한 지능 캐릭터의 복수 경로 탐색

이면섭 한국지식정보기술학회 2011 한국지식정보기술학회 논문지 Vol.6 No.1

This study applied genetic algorithm in order to search an intelligence character's multiple paths in fighting action game. There arises a problem when searching several adjoining routes in genetic algorithm. This study introduced proximity density to solve such a problem. In addition, this study applied interference density in order to exclude the interference between the searched routes. In addition, this research carried out simulation by applying the elite preservation strategy to prevent the lowering of searching capability. As a result of conducting simulation, this research searched interference-free multiple paths, and in case the elite preservation strategy was applied, the more the number of routes, there came out an increase in capabilities.