• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.2
• /
• pp.126-135
• /
• 2002

The study is concerned with an approach to the design of new architectures of fuzzy neural networks and the discussion of comprehensive design methodology supporting their development. We propose an Adaptive Fuzzy Polynomial Neural Networks(APFNN) based on Fuzzy Neural Networks(FNN) and Self-organizing Networks(SON) for model identification of complex and nonlinear systems. The proposed AFPNN is generated from the mutually combined structure of both FNN and SON. The one and the other are considered as the premise and the consequence part of AFPNN, respectively. As the premise structure of AFPNN, FNN uses both the simplified fuzzy inference and error back-propagation teaming rule. The parameters of FNN are refined(optimized) using genetic algorithms(GAs). As the consequence structure of AFPNN, SON is realized by a polynomial type of mapping(linear, quadratic and modified quadratic) between input and output variables. In this study, we introduce two kinds of AFPNN architectures, namely the basic and the modified one. The basic and the modified architectures depend on the number of input variables and the order of polynomial in each layer of consequence structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the AFPNN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed AFPNN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.10
• /
• pp.562-574
• /
• 2004

We address the problem of energy-optimal voltage scheduling for fixed-priority hard real-time systems. First, we prove that the problem is NP-hard. Then, we present a fully polynomial time approximation scheme (FPTAS) for the problem. for any $\varepsilon$>0, the proposed approximation scheme computes a voltage schedule whose energy consumption is at most (1+$\varepsilon$) times that of the optimal voltage schedule. Furthermore, the running time of the proposed approximation scheme is bounded by a polynomial function of the number of input jobs and 1/$\varepsilon$. Experimental results show that the approximation scheme finds more efficient voltage schedules faster than the best existing heuristic.

• International Journal of Aeronautical and Space Sciences
• /
• v.8 no.1
• /
• pp.46-53
• /
• 2007

In this paper, a guidance law for vision-based automatic landing of unmanned aerial vehicles (UAVs) is proposed. Automatic landing is a challenging but crucial capability for UAVs to achieve a fully autonomous flight. In an autonomous landing maneuver of UAVs, the decision of where to landing and the generation of guidance command to achieve a successful landing are very significant problem. This paper is focused on the design of guidance law applicable to automatic landing problem of fixed-wing UAV and rotary-wing UAV, simultaneously. The proposed guidance law generates acceleration command as a control input which derived from a specified time-to-go ($t_go$) polynomial function. The coefficient of $t_go$-polynomial function are determined to satisfy some terminal constraints. Nonlinear simulation results using a fixed-wing and rotary-wing UAV models are presented.

• Asia pacific journal of information systems
• /
• v.16 no.2
• /
• pp.47-67
• /
• 2006

This study takes a quantitative approach to the influence of TTF (Task-Technology Fit) on the individual's use and performance of GSS (Group Support Systems), while traditional studies on TTF have taken the experimental approach to explore the characteristic fit between diverse tasks and technologies. We have the following two research inquires: Are the IS use and performance maximized when information technologies are provided by the exact amount of demand?; and, Does TTF at the high level between task and IT produce better IS use (or performance) than at the low level? To investigate these issues, we use the polynomial regression analysis and response surface methodology of Edwards (1993) instead of traditional direct measure of TTF. This method measures the degree of desired and actual level of information technologies in conducting tasks, and traces the dynamic changes of dependent variables (IS use and performance) according to the variances of each independent variable. Our results conclude that user's IS use and performance are maximized when information technologies are actually provided by no more or less than the desired level. We also found that TTF at the high level promotes better IS use and performance than TTF at the low level.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1448-1451
• /
• 1997

In this paper, a speed control of nin-servo induction motor is considered. In this case, it is difficult to satisfy precise control performance. SO H.inf. robust controller is designed for this problem by usign polynomial approach and Youla parameterization.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.3
• /
• pp.253-258
• /
• 2005

In this study, we introduce a new category of fuzzy inference systems based on information granulation to carry out the model identification of complex and nonlinear systems. Informal speaking, information granules are viewed as linked collections of objects (data, in particular) drawn together by the criteria of proximity, similarity, or functionality. To identify the structure of fuzzy rules we use genetic algorithms (GAs). Granulation of information with the aid of Hard C-Means (HCM) clustering algorithm help determine the initial parameters of fuzzy model such as the initial apexes of the membership functions and the initial values of polynomial functions being used in the premise and consequence part of the fuzzy rules. And the initial parameters are tuned effectively with the aid of the genetic algorithms and the least square method (LSM). The proposed model is contrasted with the performance of the conventional fuzzy models in the literature.

• Transactions on Control, Automation and Systems Engineering
• /
• v.4 no.3
• /
• pp.217-223
• /
• 2002

The paper presents a single constraint equation of the direct kinematic solution of 6-dof (Stewart-Gough) platforms. Many research works have presented a single polynomial of the direct kinematics for several 6-dof platforms. However, the formulation of the polynomial has potential problems such as complicated formulation procedures and discrimination of the actual solution from all roots. This results in heavy computational burden and time-consuming task. Thus, to overcome these problems, we use a new formulation approach, called the Tetrahedron Approach, to easily derive a single constraint equation, not a polynomial one, of the direct kinematics and use two well-known numerical iterative methods to find the solution of the single constraint equation. Their performance and characteristics are investigated through a series of simulation.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.36 no.9
• /
• pp.660-669
• /
• 1987

Routh approximation method is the most computationally attractive. But this method may cause time-response error because this method does not match the time-response directly. In this paper a new mixed method for obtaining stable reduced-order models for high-order continuous-time systems is proposed. It makes use of the advantages of the Routh approximation method and the Minimization of Integral Squared Error(MISE) criterion approach. In this mixed method the characteristic polynomial of the reduced-order model is first obtained from that of original system by using the Auxiliary Denominator Polynomial(ADP). The numerator polynomial is then determined so as to minimize the intergral squared-error of unit step responses. The advantages of the propsed method are that the reduced models are always stable if the original system are stable and the frequency domain and time domain characteristic of the original system will be preserved in the reduced models.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.113-116
• /
• 2006

본 논문은 소프트웨어 공정에 대하여 기호코팅을 이용한 유전자 알고리즘 기반 퍼지 다항식 뉴럴 네트워크 (Genetic Algorithms-based Fuzzy Polynomial Neural Networks ; gFPNN)의 모델을 제안한다. 유전자 알고리즘에는 이진코딩, 기호코팅, 실수코딩이 있다. 제안된 모델은 스트링의 길이에 따른 해밍절벽을 기호코딩으로 극복하였다. gFPNN에 전반부 멤버쉽 함수는 삼각형과 가우시안형의 멤버쉽 함수가 사용된다. 그리고 규칙의 후반부는 간략, 선형, 이차식 그리고 변형된 이차식 함수에 의해 설계된다. 실험적 예제를 통하여 제안된 모델의 성능이 근사화 능력과 일반화 능력이 우수함을 보인다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.429-432
• /
• 2004

본 논문은 기존의 퍼지 다항식 뉴럴 네트워크 (Fuzzy Polynomial Neural Networks ; FPNN) 모델을 이용하여 비선형성 데이터에 대한 추론을 제안한다. 복잡한 비선형 시스템의 모델동정을 위하여 생성된 GMDH 방법에 기초한 FPNN의 각 노드는 퍼지 규칙을 기반으로 구축되었으며, 층이 진행되는 동안 모델 스스로 노드의 선택과 제거를 통해 최적의 네트워크 구조를 생성할 수 있는 유연성을 가지고 있다. FPNN 각각의 활성노드를 퍼지다항식 뉴론(Fuzzy Polynomial Neuron ; FPN)이라고 표현한다. FPNN의 후반부 구조는 입출력 변수 사이 의 간략과 회귀다항식 (1차, 2차, 변형된 2차식) 함수에 의해 구현된다. 규칙의 전반부 멤버쉽 함수는 삼각형과 가우시안형의 멤버쉽 함수가 사용된다. 또한 유전자 알고리즘을 사용하여 각노드의 부분표현식을 구성하는 입력변수의 수, 입력변수와 차수의 선택 동조를 통하여 최적의 Genetic Algorithms(GAs)을 이용한 FPNN모델을 설계하는 것이 유용하고 효과적임을 보인다.