• 한국지능시스템학회논문지
• /
• 제7권5호
• /
• pp.46-50
• /
• 1997

퍼지논리 제어에서 정적인 퍼지규칙은 플랜트나 환경 파라메터의 중대한 변화에 대처할 수 없다. 이러한 문제를 해결하기 위하여 지금까지 스스로 조직화하는 퍼지제어 및 신경망에 기초한 뉴로퍼지등의 기법이 도입되었다.그러나 이러한 기존 방법들은 동적으로 변화된 퍼지 규칙이 완전하지 않거나 모순될 수 있음으로 해서 퍼지 제어기를 위험한 상황에 처하게 할수도 있다. 본 논문에서는 예측 신경망을 사용하여 새로운 적응퍼지 제어기법을 제안한다.제안된 퍼지제어기는 비록 제어 플랜트나 환경 파라메터가 변화할지라도 초기의 완전하고 모순되지 않은 퍼지 규칙과 계속해서 학습하는 예측 신경망의 예측에러를 이용하여 제어출력을 안전하게 적응적으로 변화시켜간다. 직류 서보모터의 위치제어문제를 이용하여 실험해본 결과 제안한 방법이 적응면에서 매우 유용함을 보였다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 춘계학술대회 학술발표 논문집
• /
• pp.136-139
• /
• 2001

This paper develops a sensor based navigation method that utilizes fuzzy logic and the Dempster-Shafer evidence theory for mobile robot in uncertain environment. The proposed navigator consists of two behaviors: obstacle avoidance and goal seeking. To navigate reliably in the environment, we make a map building process before the robot finds a goal position and create a robust fuzzy controller. In this paper, the map is constructed on a two-dimensional occupancy grid. The sensor readings are fused into the map using D-S inference rule. Whenever the robot moves, it catches new information about the environment and replaces the old map with new one. With that process the robot can go wandering and finding the goal position. The usefulness of the proposed method is verified by a series of simulations. This paper deals with the fuzzy modeling for the complex and uncertain nonlinear systems, in which conventional and mathematical models may fail to give satisfactory results. Finally, we provide numerical examples to evaluate the feasibility and generality of the proposed method in this paper.

• 전기학회논문지
• /
• 제61권2호
• /
• pp.291-297
• /
• 2012

In this paper, potential field algorithm was used for path planning in dynamic environment. This algorithm is used to plan a robot path because of its elegant mathematical analysis and simplicity. However, there are some problems. The problems are problem of collision risk, problem of avoidance path, problem of time consumption. In order to solve these problems, we fused potential field with fuzzy system. The input of the fuzzy system is set using relative velocity and location of robot and obstacle. The output of the fuzzy system is set using the weighting factor of repulsive potential function. The potential field algorithm is improved by using fuzzy potential field algorithm and, path planning in various environment has been done.

• 한국시뮬레이션학회논문지
• /
• 제9권1호
• /
• pp.21-38
• /
• 2000

A fuzzy dispatching algorithm with adaptable control scheme is proposed for more flexible and adaptable operation of AGV system. The basic idea of the algorithm is prioritization of all move requests based on the fuzzy urgency. The fuzzy urgency is measured by the fuzzy multi-criteria decision-making method, utilizing the relevant information such as incoming and outgoing buffer status, elapsed time of move request, and AGV traveling distance. At every dispatching decision point, the algorithm prioritizes all move requests based on the fuzzy urgency. The performance of the proposed algorithm is compared with several dispatching algorithms in terms of system throughput in a hypothetical job shop environment. Simulation experiments are carried out varying the level of criticality ratio of AGVs , the numbers of AGVs, and the buffer capacities. The rule presented in this study appears to be more effective for dispatching AGVs than the other rules.

• 해양환경안전학회지
• /
• 제2권1호
• /
• pp.89-95
• /
• 1996

This paper describes a fuzzy controlled pre-chlorination technique for purifying the pulluted raw water in water purification lants. For the purpose of obtaining the high quality water, the appropriate pre-chlorine dosage rate has to be continuously adjusted according to a change in quality of a intake raw water, weather, solar nergy mount, temperature and etc. Therefore, the method of expressing an expert's empirical knowledge cumulated from his past carrier by fuzzy reasoning and the fuzzy controller design technique is necessary.In this paper fuzzy membership functions and rules accordingto emprircal knowledge and experimental field data were obtained, And also fuzzy cintriller design using four feedforward components for the determination of pre-chlorine dosage rate and four feedback ones for the compensation of its dosage rate with residual chlorine and its change rate, was executed.

• 지능정보연구
• /
• 제14권1호
• /
• pp.35-45
• /
• 2008

본 논문에서는 유비쿼터스 환경에서 사용자 Context를 표현하는 과제를 해결하기 위하여 시맨틱 웹 기술 및 퍼지 개념을이용하여 사용자 Context를 언어와 기종에 독립적이면서 사람이 생각하는 것과 최대한 유사한 형태로 기술하는 것을 제안한다. 재래의 방법으로 사용되어온 일반 집합으로 Context를 표현하는 방법은 실세계의 환경을 표현하는데 한계가 있기 때문에 본 논문에서는 퍼지 개념과 표준 웹 온톨로지 언어 OWL이 융합된 Fuzzy OWL 언어를 사용하였다. 본 논문에서 제안하는 방법은 사용자가 접한 환경정보들을 수치로 표현하며 이를 OWL로 기술한다. 그리고 OWL로 변환된 Context를 Fuzzy OWL로 변환한다. 마지막으로 자동적인 상황인지가 가능한지 여부를 퍼지 추론 엔진인 FiRE를 사용하여 검증한다. 본 논문에서 제시하는 방법을 사용하면 유비쿼터스 컴퓨팅 환경에서도 사용할 수 있는 형태로 Context를 기술할 수 있으며 퍼지의 개념을 사용하여 Context를 표현하기 때문에 상태나 정도를 표현함에 있어 좀 더 효과적이다. 뿐만 아니라, 기술된 Context를 기반으로 현재 사용자가 접한 환경의 상태를 추론할 수 있으며 추론된 상태에 따라 시스템이 자동적으로 작동하게 하는 것이 가능하다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1300-1304
• /
• 2003

The following paper deals with a control problem of a mobile robot in home network environment. The home network causes the mobile robot to communicate with sensors to get the sensor measurements and to be adapted to the environment changes. To get the improved performance of control of a mobile robot in spite of the change in home network environment, we use the fuzzy inference system with multiple reward reinforcement learning. The multiple reward reinforcement learning enables the mobile robot to consider the multiple control objectives and adapt itself to the change in home network environment. Multiple reward fuzzy Q-learning method is proposed for the multiple reward reinforcement learning. Multiple Q-values are considered and max-min optimization is applied to get the improved fuzzy rule. To show the effectiveness of the proposed method, some simulation results are given, which are performed in home network environment, i.e., LAN, wireless LAN, etc.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1993년도 Fifth International Fuzzy Systems Association World Congress 93
• /
• pp.1211-1214
• /
• 1993

In this paper, we present a fuzzy-number-oriented methodology to model uncertain geometric robot environment and to manipulate geometric uncertainty between robot coordinate frames. We describe any geometric primitive of robot environment as a parameter vector in parameter space. Not only ill-known values of the parameterized geometric primitive but the uncertain quantities of coordinate transformation are represented by means of fuzzy numbers restricted to appropriate membership functions. For consistent interpretation about geometric primitives between different coordinate frames, we manipulate these uncertain quantities using fuzzy arithmetic.

• 한국지능시스템학회논문지
• /
• 제7권5호
• /
• pp.67-76
• /
• 1997

In this paper, a theoretical study is presented for the force control of a miniature robotic manipulator which is driven by a pair of piezo-electric bimorph cells. In the theoretical analysis, one finger is modeled as a flexible cantilevers with a force sensor at the tip and the finger is a solid beam. The robotic finger is used to hold the objects with different stiffness such as an iron block and a living insect and a moving objcet. So it is very important to develop an adequate controller for the holding operation of the finger. The main problems in force controlling are overdamping, overshoot and unknown environment(such as the stiffness of object and unknown plant parameters). So, the main target is propose the new fuzzy compensation for unknown environment and incease the system performance. The fuzzy compensation is implemented by using PI-type fuzzy approach to identified unknown environment. And the result of proposed controller was compared with the conventaional PID and optimal controller.

• 대한전기학회논문지
• /
• 제41권5호
• /
• pp.551-561
• /
• 1992

A neural optimization network and fuzzy rules are proposed to control the redundant robot manipulators in an environment with obstacle. A neural optimization network is employed to solve the optimization problem for resolved motion control of redundant robot manipulators in an environment with obstacle. The fuzzy rules are proposed to determine the weights of neural optimization networks to avoid the collision between robot manipulators and obstacle. The inputs of fuzzy rules are the resultant distance and change of the distance and sum of the changes by differential motion of each joint. And the output of fuzzy rules is defined as the capability of collision avoidance of joint differential motion. The weightings of neural optimization networks are adjusted according to the capability of collision aboidance of each joint. To show the validities of the proposed method, computer simulation results are illustrated for the redundant robot of the planar type with three degrees of freedom.