• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.1
• /
• pp.23-34
• /
• 2002

This paper proposes a new neuro-fuzzy system that is fast trained by evolution strategy and describes application results of the proposed system to nonlinear system identification to show its usefulness. As training methods of neuro-fuzzy systems, modified error back-propagation algorithms and genetic algorithms have been used so far. However, the former has some drawbacks such as long training time, falling to local optimum, and experimental selecting of learning rates and the latter has difficulty in precise searching solutions because genetic algorithms represents solutions as genotype individuals. The evolution strategy we used can do precise search because its individuals are represented as phenotype real values, it seldom falls into a local optimum, and its training speed is faster than error back-propagation algorithms. We apply our neuro-fuzzy systems to nonlinear system identification. It was found from experiments that training speed is fast and the training results were considerably good.

• Journal of the Korea Society for Simulation
• /
• v.26 no.3
• /
• pp.73-86
• /
• 2017

Environmental data plays an important role to enhance the reliability of experimental results in M&S(Modeling and Simulation). Especially in Military M&S, extreme weather events can be used for virtual training and simulation importantly. However, the environmental data is huge and it is dispersed among multiple organizations. It is difficult for M&S operators to select the date and area where the weather phenomenon occurs in the real environmental data and to acquire them. Environmental data retrieval technology based on Fuzzy criteria is one of the important technologies for developing Environmental Scenario Generator. As a result of this study, a fuzzy retrieval algorithm composed of four main parameters(RV, MF, FRA, and MRV) was presented. This study suggests that the RV can be used as 14 m/s for wind speed and 80 mm/d for precipitation to search the date of storm accompanied by high wind. The MF, the FRA, and MRV can be used sigmoid, 0.2, and 1 respectively. The algorithm proposed in this study is expected to be very useful for searching the date on which weather phenomena necessary for simulation occurred.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.5
• /
• pp.399-405
• /
• 2007

This paper deals with the controller for a stable walking of a humanoid robot using genetic algorithm. A humanoid robot has instability during walking because it isn't fixed on the ground, and its nonlinearities of the joints increase its instability. If controller isn't robust, the robot may fall down at the ground during walking because of its nonlinearities. To solve this problem, robust controller is required to reduce the effect of nonlinearities and to gain the good tracking performance. In this paper, motion controller that is based on fuzzy-sliding mode controller is proposed. This controller can remove the effect of the saturation by limitation of the input voltage. It also includes compensator for reducing the effect of the nonlinearity by backlash and PI controller improving the tracking performance. In here, genetic algorithm is used for searching the optimal gains of the controller. From the given controller, a humanoid robot can moved more preciously. All the processes are investigated through simulations and are verified experimentally in a real joint system for a humanoid robot.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.5
• /
• pp.539-546
• /
• 2007

The important field of research in ship operation is related to the high efficiency of transportation, the convenience of maneuvering ships and the safety of navigation. For these purposes, many intelligent technologies for ship automation have been required and studied. In this paper, we propose an intelligent voice instruction-based learning (VIBL) method and discuss the building of a ship's steering control system based on this method. The VIBL system concretely consists of two functions: a text conversion function where an instructor's inputted voice is recognized and converted to text, and a linguistic instruction based learning function where the text instruction is understood through a searching process of given meaning elements. As a study method, the fuzzy theory is adopted to build maneuvering models of steersmen and then the existing LIBL is improved and combined with the voice recognition technology to propose the VIBL. The ship steering control system combined with VIBL is tested in a ship maneuvering simulator and its validity is shown.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.7
• /
• pp.38-50
• /
• 1997

This paper concerns a VHDL design and simulation of an accurate and cost-effective fuzzy logic controller (FLC). The accurcy of the proposed FLC is obtained by using the center of gravity (COG) defuzzifier that considers both membership values and spans of membership functions in calculating a crisp value. The cost-effectiveness of the proposed FLC is obtained by restructuring the conventional FLC in the following ways: Firstly, the MAX-MIN inference is inference is replaced by a read-modify-write operation that can be implemented economically in the structure of register files. Secondly, the division in the COG defuzzifier is avoided by finding the moment equilibrium point. The proposed COG defuzzifier has two disadvantages that it requires additional multipliers and it takes a lot of computation time to find the moment equilibrium point. The first disadvantage is overcome by replacing the mulitpliers with stochastic AND operations and the second disadvantage is alleviated by using a coarse-to-fine searching algorithm. The proposed FLC is described in VHDL structurally and behaviorally and whether it is working well or not is checked on SYNOPSYS VHDL simulator by using the truck backer-upper control problem.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.5
• /
• pp.423-428
• /
• 2005

This paper proposes a new edge detection method using a $3{\times}3$ ideal binary pattern and lookup table (LUT) for the mobile robot localization without any parameter adjustments. We take the mean of the pixels within the $3{\times}3$ block as a threshold by which the pixels are divided into two groups. The edge magnitude and orientation are calculated by taking the difference of average intensities of the two groups and by searching directional code in the LUT, respectively. And also the input image is not only partitioned into multiple groups according to their intensity similarities by the histogram, but also the threshold of each group is determined by fuzzy reasoning automatically. Finally, the edges are determined through non-maximum suppression using edge confidence measure and edge linking. Applying this edge detection method to the mobile robot localization using projective invariance of the cross ratio. we demonstrate the robustness of the proposed method to the illumination changes in a corridor environment.

• The Journal of Information Technology
• /
• v.4 no.2
• /
• pp.117-127
• /
• 2001

A self-structured neural network algorithm that finds optimal fuzzy membership functions and nile base to fuzzy model is proposed and a fuzzy-neural network controller is designed to get more accurate position and velocity control of wheeled mobile robot. This procedure that is composed of three steps has its own unique process at each step. The elements of output term set are increased at first step and then the rule base Is varied according to increase of the elements. The adjusted controller is in competition with controller which doesn't include any increased elements. The adjusted controller will be removed if the control-law lost. Otherwise, the controller is replaced with the adjusted system. After finished regulation of output term set and rule base, searching for input membership functions is processed with constraints and fine tuning of output membership functions is done.

• 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.

• Journal of Distribution Science
• /
• v.11 no.3
• /
• pp.13-22
• /
• 2013

Purpose - Product purpose for optimal values solution for synthesize evaluative criteria and optimize product design values. In addition, product designer has to consider the product design to conform to project, laws and regulations, authentication, from the product design stage. Research design, data, methodology - How to use an evaluative criteria model's imprecise market data by evaluative criteria research design; product mapping relationships between design parameters and customer requirements using product predicted value method. An evaluative criteria model and their associated criteria status, product evaluative criteria model of results. Results - Therefore, after the enterprise product design project analysis, effectiveness and the customer degree of satisfaction must be appraised to obtain the maximum value for the benefit on behalf of the implementation goals, the promotion product level and market competition strength. Conclusions - In multi criterion decision making (MCDM), using its searching software capacity to obtain the optimal solution.

• Journal of Industrial Technology
• /
• v.29 no.B
• /
• pp.9-15
• /
• 2009

In the case of traditional binary encoding technique, it takes long time to converge the optimal solutions and brings about complexity of the systems due to encoding and decoding procedures. However, the ROGAs (real-coded genetic algorithms) do not require these procedures, and the k-means clustering algorithm can avoid global searching space. Thus, this paper proposes a new approach by using their advantages. The proposed method constructs the multiple predictors using the optimal differences that can reveal the patterns better and properties concealed in non-stationary time series where the k-means clustering algorithm is used for data classification to each predictor, then selects the best predictor. After selecting the best predictor, the cluster centers of the predictor are tuned finely via RCGKA in secondary tuning procedure. Therefore, performance of the predictor can be more enhanced. Finally, we verifies the prediction performance of the proposed system via simulating typical time series examples.