• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.6 no.4
• /
• pp.238-242
• /
• 2013

We consider the problem of constructing observers for nonlinear systems with unknown inputs. Connectionist networks, also called neural networks, have been broadly applied to solve many different problems since McCulloch and Pitts had shown mathematically their information processing ability in 1943. In this thesis, we present a genetic neuro-control scheme for nonlinear systems. Our method is different from those using supervised learning algorithms, such as the backpropagation (BP) algorithm, that needs training information in each step. The contributions of this thesis are the new approach to constructing neural network architecture and its training.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2158-2160
• /
• 2004

본 논문에서는 불확실한 비선형 시스템에 대하여 새로운 고장진단 방법을 이용한 고장 허용 제어기를 설계한다. 잔류 신호는 비선형 관측기 구조를 이용하여 얻을 수 있다. 고장 성분은 neuro-fuzzy 근사기로 추정한다. 제안된 고장 허용 제어기는 강인 제어기와 고장 성분을 보상할 수 있는 보상제어기로 구성된다. 여기서 제안된 고장진단 방법은 고장으로 인해 발생되는 보상제어기의 크기로 고장을 진단함으로써 고장 전후의 강인 제어기의 특성을 계속유지 할 수 있게 설계하였다. 본 논문에서 제안한 고장 허용 제어기의 성능은 컴퓨터 모의실험을 통하여 증명하였다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.9
• /
• pp.1356-1362
• /
• 2015

Modular snake-like robots are robust for failure and have flexible locomotions for obstacle environment than of walking robot. This requires an adaptation capability which is obtained from a learning approach, but has not been analysed as well. In order to investigate the property of adaptation of locomotion for different terrains, NEAT controllers are trained for a flat terrain and tested for obstacle terrains. The input and output characteristics of the adaptation for the neural network controller are analyzed for different terrains in frequency domain.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.866-871
• /
• 1991

The multi-layer neural network possesses the desirable characteristics of parallel distributed processing and learning capacity, by which the uncertain variation of the parameters in the dynamically complex system can be handled adoptively. However the error back propagation algorithm that has been utilized popularly in the learning procedure of the mulfi-Jayer neural network has the significant limitations in the real application because of its slow convergence speed. In this paper, an approach to improve the convergence speed is proposed using the fuzzy logic that can effectively handle the uncertain and fuzzy informations by linguistic level. The effectiveness of the proposed algorithm is demonstrated by computer simulation of PUMA 560 robot manipulator.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1995.10a
• /
• pp.249-253
• /
• 1995

본 연구에서는 잠재적 응용성이 큰 ER유체와 압전필름을 액튜에이터로 하는 하이브리드형 지능구조물을 제안한 후 능동 진동제어를 실시하였다. 먼저 중공(hollow)의 샌드위치 형태 복합재료(glass/epoxy)보에 ER유체와 압전필름을 각각 삽입과 접착을 하여 하이브리드형 지능구조물을 제작하였다. 그리고 각 매체의 액튜에이팅 특성을 고려하여, ER유체 액튜에이터(ERFA)는 전장부하 함수로 도출되는 구조물의 주파수응답을 특징으로 하였고, 압전필름 액튜에이터(PFA)는 신경 슬라이딩 모드 제어기 (neuro sliding mode controller : NSC)를 적용하였다. 이 두 액튜에이터가 동시에 작동하는 능동 진동제어계를 실험적으로 구현한 후 과도응답과 강제 응답에 대한 진동제어 성능을 단일 액튜에이터 작동시와 비교 고찰하여 제시된 하이브리드 액튜에이팅의 효과를 입증하였다.

• Journal of Power Electronics
• /
• v.16 no.2
• /
• pp.564-571
• /
• 2016

This study proposes an adaptive neuro-fuzzy inference system (ANFIS)-based rotor position controller for brushless direct current (BLDC) motors to improve the control performance of the drive under transient and steady-state conditions. The dynamic response of a BLDC motor to the proposed ANFIS controller is considered as standard reference input. The effectiveness of the proposed controller is compared with that of the proportional integral derivative (PID) controller and fuzzy PID controller. The proposed controller solves the problem of nonlinearities and uncertainties caused by the reference input changes of BLDC motors and guarantees a fast and accurate dynamic response with an outstanding steady-state performance. Furthermore, the ANFIS controller provides low torque ripples and high starting torque. The detailed study includes a MATLAB-based simulation and an experimental prototype to illustrate the feasibility of the proposed topology.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.13 no.1
• /
• pp.76-82
• /
• 2005

In this paper, validation of Driver Steering Model has been conducted. The comparison between the simulation model and vehicle test results shows that the model is very feasible for describing combined human driver and actual vehicle dynamic behaviors. The 3D vehicle model is consisted of 6-DOF sprung mass and 4-quarter car model for vehicle body dynamics. Powertrain model including differential gear and Pacejka tire model are applied. The driver steering model is also validated with vehicle test result. The driver steering model is based on angle and displacement error from the desired path, recognized by driver.

• The Korean Journal of Pain
• /
• v.10 no.2
• /
• pp.262-265
• /
• 1997

We experienced a case of osteoid osteoma in thoracic vertebra accompanied with herniated nucleus pulposus during the management of back pain. A 32 year old male patient with herniated nucleus pulposus complained of back pain and radiation to right leg. Lumbar epidural block with 1% mepivacaine 5 ml was performed for pain control and it relieved the radiating pain. However patient continued to experience severe exacerated back pain at night which responded to aspirin. Therefore we performed further examination for existence of disease of the spine and diagnosed osteoid osteoma in the right pedicle of T12 vertebra. In conclusion, we recommend physicians when evaluating patients with back pain to be congnizant of possible existence of neoplastic disease of the spine and incorporate it in differential diagnosis.

• Proceedings of the KIEE Conference
• /
• 1997.07b
• /
• pp.739-742
• /
• 1997

This paper proposes the structure and parameter optimization technique of fuzzy neural networks using genetic algorithm. Fuzzy neural network has advantages of both the fuzzy inference system and neural network. The determination of the optimal parameters and structure of the fuzzy neural networks, however, requires special efforts. To solve these problems, we propose a new learning method for optimization of fuzzy neural networks using genetic algorithm. It can optimize the structure and parameters of the entire fuzzy neural network globally. Numerical example is provided to show the advantages of the proposed method.

• Smart Structures and Systems
• /
• v.5 no.2
• /
• pp.139-152
• /
• 2009

Globally, civil infrastructures are deteriorating at an alarming rate caused by overuse, overloading, aging, damage or failure due to natural or man-made hazards. With such a vast network of deteriorating infrastructure, there is a growing interest in continuous monitoring technologies. In order to provide a true distributed sensor and control system for civil structures, we are developing a Structural Nervous System that mimics key attributes of a human nervous system. This nervous system is made up of building blocks that are designed based on mechanoreceptors as a fundamentally new approach for the development of a structural health monitoring and diagnostic system that utilizes the recently developed piezo-fibers capable of sensing and actuation. In particular, our research has been focused on producing a sensory nervous system for civil structures by using piezo-fibers as sensory receptors, nerve fibers, neuronal pools, and spinocervical tract to the nodal and central processing units. This paper presents up to date results of our research, including the design and analysis of the structural nervous system.