• Journal of the Korea Society for Simulation
• /
• v.19 no.2
• /
• pp.91-98
• /
• 2010

Conventional path-finding has focused on finding short collision-free paths. However, as computer games become more sophisticated, it is required to take tactical information like ambush points or lines of enemy sight into account. One way to make this information have an effect on path-finding is to represent a heuristic function of a search algorithm as a weighted sum of tactics. In this paper we consider the problem of learning heuristic to optimize path-finding based on given tactical information. What is meant by learning is to produce a good weight vector for a heuristic function. Training examples for learning are given by a game level-designer and will be compared with search results in every search level to update weights. This paper proposes a learning algorithm integrated with search for tactical path-finding. The perceptron-like method for updating weights is described and a simulation tool for implementing these is presented. A level-designer can mark desired paths according to characters' properties in the heuristic learning tool and then it uses them as training examples to learn weights and shows traces of paths changing along with weight learning.

• Journal of Korea Multimedia Society
• /
• v.12 no.9
• /
• pp.1333-1341
• /
• 2009

Tactical path-finding in computer games is path-finding where a path is selected by considering not only basic elements such as the shortest distance or the minimum time spend but also tactical information of surroundings when deciding character's moving trajectory. One way to include tactical information in path-finding is to represent a heuristic function as a sum of tactical quality multiplied by a weighting factor which is.. determined based on the degree of its importance. The choice of weighting factors for tactics is very important because it controls search performance and the characteristic of paths found. In this paper. we propose a method for improving a heuristic function by adjusting weights based on the difference between paths on examples given by a level designer and paths found during the search process based on the CUITent weighting factors. The proposed method includes the search algorithm modified to detect search errors and learn heuristics and the perceptron-like weight updating formular. Through simulations it is demonstrated how different paths found by tactical path-finding are from those by traditional path-finding. We analyze the factors that affect the performance of learning and show the example applied to the real game environments.

• Journal of Gifted/Talented Education
• /
• v.22 no.3
• /
• pp.729-755
• /
• 2012

As an effort to understand alienated gifted students, we investigated learning characteristics and learning tactics of a scientifically gifted student with economic difficulty and physical disadvantage. The student we studied is attending the Saturday Physics Class which is an after school science activity offered by our university. We adopted techniques of qualitative case study. Participant observation was carried out at the field and the interview was done with the participant, his mother, and his teacher of 5th grade. Field documents and self-reports were used to understand the student synthetically. As a result, learning characteristics of the participant could be summarized as a spontaneous learning which originated from the internal motivation and struggle for learning to overcome the sense of inferiority and isolation from the peers. The participant adopted a strategic method for learning to satisfy his learning desire given the circumstance of socioeconomic and physical disadvantage: the three tactics we found were various learning routes, meta-cognitive ability and fervent response.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.3-6
• /
• 2004

본 논문에서는 퍼지 로직을 이용하여 로봇과 공의 상태에 따른 로봇 행동의 선택 알고리즘을 제시한다. 전략 및 전술 알고리즘으로 많이 알려진 Modular Q-학습 알고리즘은 개체의 수에 따른 상태수를 지수 함수적으로 증가시킬 뿐만 아니라, 로봇이 협력하기 위해 중재자 모듈이라는 별도의 알고리즘을 필요로 한다. 그러나 앞으로 제시하는 퍼지 로직을 적용한 로봇축구 전략 및 전술 알고리즘은 퍼지 로직을 이용하여 로봇의 주행 알고리즘을 선택하는 과정과 로봇의 행동을 협력하는 과정을 동시에 구현함으로써, 계산 양을 줄여 로봇 축구에 보다 적합하게 해준다.

• Journal of Convergence for Information Technology
• /
• v.11 no.1
• /
• pp.12-19
• /
• 2021

For strengthening the national defense, the function of tactical network is essential. tactics and strategies in wartime situations are based on numerous information. Therefore, various reconnaissance devices and resources are used to collect a huge amount of information, and they transmit the information through tactical networks. In tactical networks that which use contention based channel access scheme, high-speed nodes such as recon aircraft may have performance degradation problems due to unnecessary channel occupation. In this paper, we propose a learning-backoff method, which empirically learns the size of the contention window to determine channel access time. The proposed method shows that the network throughput can be increased up to 25% as the number of high-speed mobility nodes are increases.

• Journal of the Korea Society for Simulation
• /
• v.30 no.3
• /
• pp.1-9
• /
• 2021

Commander and staffs on the battlefield are aware of the situation and, based on the results, they perform military activities through their military decisions. Recently, with the development of information technology, the demand for artificial intelligence to support military decisions has increased. It is essential to identify, collect, and pre-process the data set for reinforcement learning to utilize artificial intelligence. However, data on enemies lacking in terms of accuracy, timeliness, and abundance is not suitable for use as AI learning data, so a training model is needed to collect AI learning data. In this paper, a methodology for learning artificial intelligence was presented using the constructive wargame model exercise data. First, the role and scope of artificial intelligence to support the commander and staff in the military decision-making process were specified, and to train artificial intelligence according to the role, learning data was identified in the Chang-Jo 21 model exercise data and the learning results were simulated. The simulation data set was created as imaginary sample data, and the doctrine of ROK Army, which is restricted to disclosure, was utilized with US Army's doctrine that can be collected on the Internet.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.317-320
• /
• 2005

본 논문에서는 퍼지 로직을 이용하여 로봇의 여러 환경변수에 따라 로봇들의 행동을 적절히 선택하는 알고리즘을 제시한다. 전략 및 전술 알고리즘으로 많이 알려진 Modular Q-학습 알고리즘은 개체의 수에 따른 상태수를 지수 함수적으로 증가시킬 뿐만 아니라, 로봇이 협력하기 위해 중재자모듈이라는 별도의 알고리즘을 필요로 한다. 그러나 앞으로 제시하는 로봇 행동의 퍼지 적정성을 고려한 로봇축구 전략 및 전술 알고리즘은 환경 변수에 따라 로봇 행동의 적절성을 퍼지 로직을 통하여 얻어내게 하였으며, 이를 이용함으로써 다수 로봇의 상호작용도 고려할 수 있게 하였다.

• Journal of Korea Game Society
• /
• v.6 no.2
• /
• pp.13-22
• /
• 2006

As the development of games continues, the intelligence of NPC is becoming more and more important. Nowadays, the NPCs of MMORPGS are not only capable of simple actions like moving and attacking players, but also utilizing variety of skills and tactics as human-players do. This study suggests a method that grants characters used in RPG(Role-Playing Game) an ability of training and adaptation using Neural network and Genetic Algorithm. In this study, a simple game-play model is constructed to test how suggested intellect characters could train and adapt themselves to game rules and tactics. In the game-play model, three types of characters(Tanker, Dealer, Healer) are used. Intellect character group constructed by NN and GA, and trained by combats against enemy character group constructed by FSM. As the result of test, the proposed intellect characters group acquire an appropriate combat tactics by themselves according to their abilities and those of enemies, and adapt change of game rule.

• Journal of the Korea Society for Simulation
• /
• v.21 no.4
• /
• pp.65-73
• /
• 2012

Due to the development of technology related to a weapon system and the info-communication, the battle system of a warship has to manage many kinds of human intervention tactics according to the complicated battlefield environment. Therefore, many kinds of studies about M&S(Modeling & Simulation) have been carried out recently. The previous M&S system based on an agent, however, has simply used non-flexible(or fixed) tactics. In this paper, we propose an agent modeling methodology which has reinforcement learning function for spontaneous(active) reaction and generation evolution learning Function using Genetic Algorithm for more proper reaction for warship battle. We experiment with virtual 1:1 warship combat simulation on the west sea so as to test validity of our proposed methodology. We consequently show the possibility of both reinforcement and evolution learning in a warship battle.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.1
• /
• pp.79-85
• /
• 2006

This paper presents a strategy and tactic for robot soccer using furry logic mediator that determines robot action depending on the positions and the roles of adjacent two robots. Conventional Q-learning algorithm, where the number of states increases exponentially with the number of robots, is not suitable for a robot soccer system, because it needs so much calculation that processing cannot be accomplished in real time. A modular Q-teaming algorithm reduces a number of states by partitioning the concerned area, where mediator algorithm for cooperation of robots is used additionally. The proposed scheme implements the mediator algorithm among robots by fuzzy logic system, where simple fuzzy rules make the calculation easy and hence proper for robot soccer system. The simulation of MiroSot shows the feasibility of the proposed scheme.