• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.253-258
• /
• 2008

Recently, game players want new game requiring more various tactics and strategies in the complex environment beyond simple and repetitive play. Various artificial intelligence techniques have been suggested to make the game characters learn within this environment, and the recent researches include the neural network and the genetic algorithm. The Genetic programming(GP) has been used in this study for learning strategy of the agent in the pursuit-evasion problem which is used widely in the game theories. The suggested GP algorithm is faster than the existing algorithm such as neural network, it can be understood instinctively, and it has high adaptability since the evolving chromosomes can be transformed to the reasoning rules.

• Journal of Korea Game Society
• /
• v.13 no.6
• /
• pp.27-34
• /
• 2013

This paper analyzes the behaviour of a prey to avoid the pursuit of a predator at predator-prey relationship to be appeared in the collective behavior of animals. One of the methods to avoid the pursuit of a predator is to rotate quickly when a predator arrives near to it. At that moment, a critical distance and a rotating angular are very important for the prey in order to survive from the pursuit, where the critical distance is the distance between the predator and the prey just before rotation. In order to analyze the critical distance and the rotating angular, this paper introduces the energy for a predator which it has at starting point of the chase and consumes during the chase. Through simulations, we can know that the rotating angle for a prey to survive from the pursuit is increased when the critical distance is shorter and when the ratio of predator's mass and prey's mass is also decreased. The results of simulations are the similar phenomenon in nature and therefore it means that the method to analyze in this paper is correct.

• Journal of the Korea Computer Graphics Society
• /
• v.15 no.2
• /
• pp.9-17
• /
• 2009

Animating multiple characters to compete with each other is an important problem in computer games and animation films. However, it remains difficult to simulate strategic competition among characters because of its inherent complex decision process that should be able to cope with often unpredictable behavior of opponents. We adopt a reinforcement learning method in Markov games to action models built from captured motion data. This enables two characters to perform globally optimal counter-strategies with respect to each other. We also extend this method to simulate competition between two teams, each of which can consist of an arbitrary number of characters. We demonstrate the usefulness of our approach through various competitive scenarios, including playing-tag, keeping-distance, and shooting.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.23 no.4
• /
• pp.399-406
• /
• 2020

In this paper, we propose a time-varying proportional navigation guidance law that determines the proportional navigation gain in real-time according to the operating situation. When intercepting a target, an unidentified evasion strategy causes a loss of optimality. To compensate for this problem, proper proportional navigation gain is derived at every time step by solving an optimal control problem with the inferred evader's strategy. Recently, deep reinforcement learning algorithms are introduced to deal with complex optimal control problem efficiently. We adapt the actor-critic method to build a proportional navigation gain network and the network is trained by the Proximal Policy Optimization(PPO) algorithm to learn an evasion strategy of the target. Numerical experiments show the effectiveness and optimality of the proposed method.