• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.27 no.1
• /
• pp.64-73
• /
• 2004

This paper addresses a problem of machining determination of input quantity in a multi-stage flexible flow system with non-identical parallel machines considers a subcontracting, machining restraint, and machine yield. We develop a nonlinear programing with the objective of minimizing the sum of in-house processing cost and subcontracting cost. To solve this model, we introduce a single-processor parallel genetic algorithm(SPGA) to improve a weak point for the declined robustness of simple algorithm(SGA). The efficiency of the SPGA is examined in comparison with the SGA for the same problem. In of examination the SPGA is to provide the excellent solution than the solution of the SGA.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.3 no.1
• /
• pp.115-125
• /
• 1999

Simple Genetic Algorithm(SGA) proposed by J. H. Holland is a population-based optimization method based on the principle of the Darwinian natural selection. The theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. Although GA does well in many applications as an optimization method, still it does not guarantee the convergence to a global optimum in some problems. In designing intelligent systems, specially, since there is no deterministic solution, a heuristic trial-and error procedure is usually used to determine the systems' parameters. As an alternative scheme, therefore, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve. In this paper we review the existing co-evolutionary algorithms and propose co-evolutionary schemes designing intelligent systems according to the relation between the system's components.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.292-297
• /
• 1999

Evolving in artificial agent is an extremely difficult problem, but on the other hand, a challenging task. At present the studies mainly centered on single agent learning problem. In our case, we use simulated soccer to investigate multi-agent cooperative learning. Consider the fundamental differences in learning mechanism, existing reinforcement learning algorithms can be roughly classified into two types-that based on evaluation functions and that of searching policy space directly. Genetic Programming developed from Genetic Algorithms is one of the most well known approaches belonging to the latter. In this paper, we give detailed algorithm description as well as data construction that are necessary for learning single agent strategies at first. In following step moreover, we will extend developed methods into multiple robot domains. game. We investigate and contrast two different methods-simple team learning and sub-group loaming and conclude the paper with some experimental results.

• Journal of Bio-Environment Control
• /
• v.13 no.4
• /
• pp.209-216
• /
• 2004

This study was aimed to assess the effects of microclimate factors on lettuce chlorophyll fluorescent responses and to develop an environment control system for plant growth by adopting a simple genetic algorithm. The photosynthetic responses measurements were repeated by changing one factor among six climatic factors at a time. The maximum Fv'/Fm' resulted when the ambient temperature was $21^{\circ}C,\;CO_2$ concentration range of 1,200 to 1,400 ppm, relative humidity of $68\%$, air current speed of $1.4m{\cdot}s^{-1}$, and the temperature of nutrient solution of $20^{\circ}C$. In PPF greater than $140{\mu}mol{\cdot}m^{-2}{\cdot}s^{-1}$, Fv'/Fm' values were decreased. To estimate the effects of combined microclimate factors on plant growth, a photosynthesis efficiency model was developed using principle component analysis for six microclimate factors. Predicted Fv'/Fm' values showed a good agreement to measured ones with an average error of $2.5\%$. In this study, a simple genetic algorithm was applied to the photosynthesis efficiency model for optimal environmental condition for lettuce growth. Air emperature of $22^{\circ}C$, root zone temperature of $19^{\circ}C,\;CO_2$ concentration of 1,400 ppm, air current speed of $1.0m{\cdot}s^{-1}$, PPF of $430{\mu}mol{\cdot}m^{-2}{\cdot}s^{-1}$, and relative humidity of $65\%$ were obtained. It is feasible to control plant environment optimally in response to microclimate changes by using photosynthesis efficiency model combined with genetic algorithm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.2
• /
• pp.228-233
• /
• 2004

Nowadays, the robot with various and complex functions is required. previous algorithms, however, cannot satisfy the requirement. In order to solve these problems, we introduce the 2-Layer Fuzzy Controller, which has a small number of fuzzy rules corresponding to various inputs and outputs. Also, it controls robustly and effectively an object. The main problem in the fuzzy controller is how to design the fuzzy rule. This paper designs the optimal 2-layer fuzzy controller using the Schema Co-Evolutionary Algorithm. The schema co-evolutionary algorithm can find more rapidly and excellently than simple genetic algorithm does.

• 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 Institute of Control, Robotics and Systems
• /
• v.17 no.10
• /
• pp.1021-1028
• /
• 2011

The objective of this paper is to optimize the design parameters of a novel mechanism for a robotic knee orthosis. The feature of the proposed knee othosis is to drive a knee joint with independent actuation during swing and stance phases, which can allow an actuator with fast rotation to control swing motions and an actuator with high torque to control stance motions, respectively. The quadriceps device operates in five-bar links with 2-DOF motions during swing phase and is changed to six-bar links during stance phase by the contact motion to the patella device. The hamstring device operates in a slider-crank mechanism for entire gait cycle. The suggested kinematic model will allow a robotic knee orthosis to use compact and light actuators with full support during walking. However, the proposed orthosis must use additional linkages than a simple four-bar mechanism. To maximize the benefit of reducing the actuators power by using the developed kinematic design, it is necessary to minimize total weight of the device, while keeping necessary actuator performances of torques and angular velocities for support. In this paper, we use a SGA (Simple Genetic Algorithm) to minimize sum of total link lengths and motor power by reducing the weight of the novel knee orthosis. To find feasible parameters, kinematic constraints of the hamstring and quadriceps mechanisms have been applied to the algorithm. The proposed optimization scheme could reduce sum of total link lengths to half of the initial value. The proposed optimization scheme can be applied to reduce total weight of general multi-linkages while keeping necessary actuator specifications.

• Smart Structures and Systems
• /
• v.11 no.1
• /
• pp.53-67
• /
• 2013

In this paper we apply Monte Carlo Filter to identifying dynamic parameters of structural systems and improve the efficiency of this algorithm. The algorithms using Monte Carlo Filter so far has not been practical to apply to structural identification for large scale structural systems because computation time increases exponentially as the degrees of freedom of the system increase. To overcome this problem, we developed a method being able to reduce number of particles which express possible structural response state vector. In MCF there are two steps which are the prediction and filtering processes. The idea is very simple. The prediction process remains intact but the filtering process is conducted at each node of structural system in the proposed method. We named this algorithm as relaxation Monte Carlo Filter (RMCF) and demonstrate its efficiency to identify large degree of freedom systems. Moreover to increase searching field and speed up convergence time of structural parameters we proposed an algorithm combining the Genetic Algorithm with RMCF and named GARMCF. Using shaking table test data of a model structure we also demonstrate the efficiency of proposed algorithm.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.2 no.1
• /
• pp.95-110
• /
• 1998

Evolutionary Algorithms (EAs) are population-based optimization methods based on the principle of Darwinian natural selection. The representative methodology in EAs is genetic algorithm (GA) proposed by J. H. Holland, and the theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. In the meaning of these foundational concepts, simple genetic algorithm (SGA) allocate more trials to the schemata whose average fitness remains above average. Although SGA does well in many applications as an optimization method, still it does not guarantee the convergence of a global optimum in GA-hard problems and deceptive problems. Therefore as an alternative scheme, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve in contrast with traditional single population evolutionary algorithm. In this paper we show why the co-evolutionary algorithm works better than SGA in terms of an extended schema theorem. And predator-prey co-evolution and symbiotic co-evolution, typical approaching methods to co-evolution, are reviewed, and dynamic fitness landscape associated with co-evolution is explained. And the experimental results show a co-evolutionary algorithm works well in optimization problems even though in deceptive functions.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.2
• /
• pp.406-417
• /
• 2014

This paper presents a numerical optimization method to improve the performance of the propeller with Turbo-Ring using real-coded genetic algorithm. In the presented method, Unimodal Normal Distribution Crossover (UNDX) and Minimal Generation Gap (MGG) model are used as crossover operator and generation-alternation model, respectively. Propeller characteristics are evaluated by a simple surface panel method "SQCM" in the optimization process. Blade sections of the original Turbo-Ring and propeller are replaced by the NACA66 a = 0.8 section. However, original chord, skew, rake and maximum blade thickness distributions in the radial direction are unchanged. Pitch and maximum camber distributions in the radial direction are selected as the design variables. Optimization is conducted to maximize the efficiency of the propeller with Turbo-Ring. The experimental result shows that the efficiency of the optimized propeller with Turbo-Ring is higher than that of the original propeller with Turbo-Ring.