• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.63-68
• /
• 2000

In this paper, a control algorithm suitable for multi-objective control problems is proposed based on the orthogonal array which is normally used in statistics and industrial engineering. And a newly defined Nth-certainty factor is suggested, which can effectively exclude the less confident rules. The Nth-certainty factor is defined by the F-values of the ANOVA(analysis of variance) table. It is shown that the algorithm can be successfully adopted to the design of controller for an active magnetic bearing system.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.1
• /
• pp.89-93
• /
• 2010

In this paper a new fuzzy clustering approach to the color clustering problem has been proposed. To deal with the limitations of the traditional FCM algorithm, we propose a spatial homogeneity-based FCM algorithm. Moreover, the cluster validity index is employed to automatically determine the number of clusters for a given image. We refer to this method as VS-FCM algorithm. The effectiveness of the proposed method is demonstrated through various clustering examples.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.98-103
• /
• 1995

In this paper, a method is presented for finding an optimal temperature control pattern in microwaveoven using genetic algorithm. Power spectrum of temperature variance of charcoal is obtained and oven system modeling with fuzzy-neural-network is explained. Fan on/off timing is converted to strings in gene pool and then genetic iterations make the power spectrum of simmulated temperature variance of microwave oven closer to that o charcoal.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.218-223
• /
• 1995

A new design scheme of a binary decision tree is proposed. In this scheme a binary decision tree is constructed by using genetic algorithm and FCM algorithm. At each node optimal or near-optimal feature or feature subset among all the available features is selected based on fitness function in genetic algorithm which is inversely proportional to classification error, balance between cluster, number of feature used. The proposed design scheme is applied to the handwtitten alphabetic characters. Experimental results show the usefulness of the proposed scheme.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.375-378
• /
• 2002

In this paper we propose parallel GA to optimize mutation rate and crossover rate using server-client model. The performance of GA depend on the good choice of crossover and mutation rates. Although many researcher has been study about the good choice, it is still unsolved problem. proposed GA optimize crossover and mutation rates trough evolving subpopulation. In virtue of the server-client model, these parameters can be evolved rapidly with relatively low-grade

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.305-309
• /
• 2006

In this paper, a face recognition algorithm system using Principle Component Analysis is proposed. The algorithm recognized a person by comparing characteristics (features) of the face to those of known individuals which is a face database of Intelligence Control Laboratory(ICONL). Experiments were simulated in order to demonstrate the performance of this algorithm due to face recognition which presented for the classification of face and non-face and the classification of known and unknown.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.451-456
• /
• 2005

In this paper, we propose the navigation control method for intelligent robot using messy genetic algorithm. The fuzzy controller design for navigation of the intelligent robot was dependant on expert's knowledge. But, the parameters of the fuzzy logic controller obtained from expert's control action may not be outimal. In this paper, to solve the above problem, we propose the identification method to automatically tune the number of fuzzy rule and parameters of memberships of fuzzy controller using mGA. Finally, to show and evaluate the generality and feasibility of the proposed method, we provides some simulations for wall following navigation of intelligent robot.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.4
• /
• pp.518-523
• /
• 2008

In this paper, we present the development of an intelligent diagnosis system for detecting faults of the transmission line. Based on the TFDR (Time-Frequency Domain Reflectometry), the fault detecting performs to measure the location of fault line. We analyze the reflected signal which is sent from the wire detecting system and classify the fault type of the wires by using intelligent diagnosis system. In order to analyze effectively, we construct the intelligent diagnosis system which is based on the fuzzy-bayesian algorithm. Finally, we provide the simulation results which are performed at transmission line to evaluate the feasibility and generality of the proposed method in this paper.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.270-275
• /
• 2016

We present an approach for acceleration actor-critic algorithm for reinforcement learning with continuous action space. Actor-critic algorithm has already proved its robustness to the infinitely large action spaces in various high dimensional environments. Despite that success, the main problem of the actor-critic algorithm remains the same-speed of convergence to the optimal policy. In high dimensional state and action space, a searching for the correct action in each state takes enormously long time. Therefore, in this paper we suggest a search accelerating function that allows to leverage speed of algorithm convergence and reach optimal policy faster. In our method, we assume that actions may have their own distribution of preference, that independent on the state. Since in the beginning of learning agent act randomly in the environment, it would be more efficient if actions were taken according to the some heuristic function. We demonstrate that heuristically-accelerated actor-critic algorithm learns optimal policy faster, using Educational Process Mining dataset with records of students' course learning process and their grades.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.392-394
• /
• 2003

There is a well-known simple, stable standard merge algorithm, which uses only linear time but for the price of double space. This extra space consumption has been often remarked as lack of the standard merge sort algorithm that covers a merge process as central operation. In-place merging is a way to overcome this lack and so is a topic with a long tradition of inspection in the area of theoretical computer science. We present an in-place merging algorithm that rear-ranges the elements to be merged by rotation, a special form of block interchanging. Our algorithm is novel, due to its technique of determination of the rotation-areas. Further it has a short and transparent definition. We will give a presentation of our algorithm and prove that it needs in the worst case not more than twice as much comparisons as the standard merge algorithm. Experimental work has shown that our algorithm is efficient and so might be of high practical interest.