• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2011년도 추계학술대회
• /
• pp.219-222
• /
• 2011

어떠한 객체의 움직임을 추적하거나 상태변화를 추정하기 위해서 사용하는 방법으로는 칼만필터, 파티클 필터, 동적 클러스터링 등이 있다. 이 중 동적클러스터링 기법은 여러 프레임에 걸쳐 클러스터를 추적하고 변화 경향을 분석하는데 유용한 방법이다. 본 논문에서는 유사성 기반의 동적 클러스터링 방법을 제안하고 시뮬레이션 하여 검증하였다. 제안한 동적 클러스터링 방법은 연속된 각 프레임에 대해 유사한 특성을 가지는 클러스터를 인접한 프레임에 걸쳐 동일한 클러스터로 판단하는 방법이다. 각 정지 프레임에서의 클러스터의 특성을 이용하여 프레임의 변화를 분석하고 유사성이 높은 클러스터들을 동일 클러스터로 지정하였다. 유사성 판단 방법은 Mamdani방식의 퍼지 모델을 제안하였다. 제안한 알고리즘은 시간에 대해 연속성을 가진 레이더 반사도 데이터에 적용하였고 시간의 흐름에 따른 클러스터의 변화를 관측할 수 있었다.

• Journal of Power Electronics
• /
• 제13권4호
• /
• pp.516-527
• /
• 2013

This paper presents a new overall system for state-of-available-power (SoAP) prediction for a lithium-ion battery pack. The essential part of this method is based on an adaptive network architecture which utilizes both fuzzy model (FIS) and artificial neural network (ANN) into the framework of adaptive neuro-fuzzy inference system (ANFIS). While battery aging proceeds, the system is capable of delivering accurate power prediction not only for room temperature, but also at lower temperatures at which power prediction is most challenging. Due to design property of ANN, the network parameters are adapted on-line to the current battery states (state-of-charge (SoC), state-of-health (SoH), temperature). SoC is required as an input parameter to SoAP module and high accuracy is crucial for a reliable on-line adaptation. Therefore, a reasonable way to determine the battery state variables is proposed applying a combination of several partly different algorithms. Among other SoC boundary estimation methods, robust extended Kalman filter (REKF) for recalibration of amp hour counters was implemented. ANFIS then achieves the SoAP estimation by means of time forward voltage prognosis (TFVP) before a power pulse occurs. The trade-off between computational cost of batch-learning and accuracy during on-line adaptation was optimized resulting in a real-time system with TFVP absolute error less than 1%. The verification was performed on a software-in-the-loop test bench setup using a 53 Ah lithium-ion cell.

• 한국지능시스템학회논문지
• /
• 제21권4호
• /
• pp.469-474
• /
• 2011

본 논문에서는 기동표적의 위치 오차값 보상 기법을 이용한 지능형 기동표적 추적 기법을 제안한다. 기동표적의 관측값과 예상위치와의 차이를 가속도와 순수 잡음으로 분리한다. 최적의 수준으로 가속도를 추출하기 위하여 K-means 클러스터링 기법과 TS 퍼지 시스템을 이용한다. K-means 클러스터링에 의해 분리된 가속도와 잡음에 대한 소속함수를 설정하고 퍼지 모델화하여 기동표적의 특성을 파악한다. 계산상의 오차를 보상하기 위하여 분리된 가속도와 잡음은 추적 알고리즘의 계산과정에 적절히 이용된다. 추정값 계산시, 가속도를 분리 하므로써 필터링 과정은 표적의 비선형 기동을 선형기동으로 인식하여 칼만필터의 성능을 유지시킨다. 기동표적의 비선형성에 대한 오차는 추정된 가속도를 통해 보상된다. 제안된 시스템의 소속함수에 사용되는 파라미터값을 조종하여 상황에 따라 적응성과 강인성을 향상시킨다. 제안된 시스템은 실시간 추적이 가능하도록 구성하였으며, 몇 가지 예를 통하여 본 논문에서 제안한 방법의 우수성을 증명한다.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1991년도 춘계학술대회
• /
• pp.34-38
• /
• 1991

This paper describes and discusses the employment of EMG pattern analysis to provide upper-motor-neuron paraplegics with patient-responsive control of FES (functional electrical stimulation) for the purpose of walker-supported walking. The use of above - lesion EMG signals as a solution to the control problem is considered. The AR (autoregressive) parameters are identified by Kalman filter algorithm using DSP chip and classified by fuzzy theory. The control and stimuli part of the below-lesion are based on microprocessor(8031). The designed stimulator is a 4-channel version. The experiments described above have only attempted to discriminate between standing function and sit-down function. A further advantage of the this system is applied for motor rehabilitation of social readaption of paralyzed humans.

• 한국지능시스템학회논문지
• /
• 제12권6호
• /
• pp.497-502
• /
• 2002

기동표적의 추적문제는 상태추정의 분야에서 수 십 년에 걸쳐 연구되어 왔다. 칼만 필터는 표적의 상태를 추정하기 위해 널리 사용되어 왔으나, 기동이 발생할 경우, 그 성능은 현저히 저하될 수 있다. 본 논문에서는 이러한 문제점을 해결하고, 기동표적을 효과적으로 추적하기 위해, DNA 코딩에 기반한 상호작용 다중모델 기법을 제안한다. 제안된 기법은 DNA 코딩에 기반한 퍼지 논리를 이용함으로써, 기존의 기법들의 수학적 한계를 극복할 수 있다. 컴퓨터 모의실험을 통하여, 제안된 기법의 추적 성능은 적응 상호작용 다중모델 기법 및 유전 알고리즘 기반 상호작용 다중모델 기법과 비교된다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제11권4호
• /
• pp.280-285
• /
• 2011

Recently, with the development of service robots and the concept of ubiquitous, the position estimation of mobile objects has received great interest. Some of the localization schemes are introduced, which provide the relative location of the moving objects subjected to accumulated errors. To implement a real time localization system, a new absolute position estimation method for a mobile robot in indoor environment is proposed. Design and implementation of the localization system comes from the usage of active beacon systems (based upon RFID technology). The active beacon system is composed of an RFID receiver and an ultra-sonic transmitter. The RFID receiver gets the synchronization signal from the mobile robot and the ultra-sonic transmitter sends out the traveling signal to be used for measuring the distance. Position of a mobile robot in a three dimensional space can be calculated basically from the distance information from three beacons and the absolute position information of the beacons themselves. In some case, the mobile robot can acquire the ultrasonic signals from only one or two beacons, due to the obstacles located along the moving path. In this paper, a position estimation scheme using fewer than three sensors is developed. Also, the extended Kalman filter algorithm is applied for the improvement of position estimation accuracy of the mobile robot.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.2161-2166
• /
• 2009

It is important for 0-6 hour nowcasting to provide for a high-quality initial condition in a meso-scale atmospheric model by a data assimilation of several observation data. The polarimetric radar data is expected to be assimilated into the forecast model, because the radar has a possibility of measurements of the types, the shapes, and the size distributions of hydrometeors. In this paper, an impact on rainfall prediction of the data assimilation of hydrometeor types (i.e. raindrop, graupel, snowflake, etc.) is evaluated. The observed information of hydrometeor types is estimated using the fuzzy logic algorism. As an implementation, the cloud-resolving nonhydrostatic atmospheric model, CReSS, which has detail microphysical processes, is employed as a forecast model. The local ensemble transform Kalman filter, LETKF, is used as a data assimilation method, which uses an ensemble of short-term forecasts to estimate the flowdependent background error covariance required in data assimilation. A heavy rainfall event occurred in Okinawa in 2008 is chosen as an application. As a result, the rainfall prediction accuracy in the assimilation case of both hydrometeor types and the Doppler velocity and the radar echo is improved by a comparison of the no assimilation case. The effects on rainfall prediction of the assimilation of hydrometeor types appear in longer prediction lead time compared with the effects of the assimilation of radar echo only.

• 조명전기설비학회논문지
• /
• 제28권9호
• /
• pp.1-7
• /
• 2014

Brushless Direct Current(BLDC) Motor is essential to measure a rotor position because of that this motor type needs to synchronize the rotor's position and changeover phase current instead of a brush and commutator used on the existing dc motor. Recently, many researches have studied on sensorless control drive for BLDC motor. The conventional control methods are a compensation value dq, Kalman filter, Fuzzy logic, Neurons neural network, and the like. These methods has difficulties of detecting BEMF accurately at low speed because of low BEMF voltage and switching noise. And also, the operation is long and complex. So, it is required a high-performance microprocessor. Therefore, it is not suitable for a small BLDC motor sensorless drive. This paper presents control methods suitable for economic small BLDC motor sensorless drive which are an improved design of the BEMF detection circuit, simplifying a complex algorithm and computation time reduction. The improved motor sensorless drive is verified stability and validity through being designed, manufactured and analyzed.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제10권2호
• /
• pp.101-106
• /
• 2010

We propose an approach to estimate the real-time moving trajectory of an object in this paper. The object's position is obtained from the image data of a CCD camera, while a state estimator predicts the linear and angular velocities of the moving object. To overcome the uncertainties and noises residing in the input data, a Extended Kalman Filter(EKF) and neural networks are utilized cooperatively. Since the EKF needs to approximate a nonlinear system into a linear model in order to estimate the states, there still exist errors as well as uncertainties. To resolve this problem, in this approach the Kohonen networks, which have a high adaptability to the memory of the inputoutput relationship, are utilized for the nonlinear region. In addition to this, the Kohonen network, as a sort of neural network, can effectively adapt to the dynamic variations and become robust against noises. This approach is derived from the observation that the Kohonen network is a type of self-organized map and is spatially oriented, which makes it suitable for determining the trajectories of moving objects. The superiority of the proposed algorithm compared with the EKF is demonstrated through real experiments.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제16권3호
• /
• pp.216-223
• /
• 2016

This paper proposes a new controller design method for a fork-type lifter (FTL) of a transportation mobile robot. The transportation robot needs to pick up a package from a stack on a storage shelf and move on by a planned path in a logistics center environment. The position of the storage shelf is recognized by reading a QR code on the floor, and using this position, the robot can move to reach the storage shelf and pick up the package. PID controllers and an adaptive controller are designed to control the velocity of two wheels and the position of the FTL. An adaptive controller for the lifter is designed to elevate up and down on a slideway to the correct height position of the package on the stack of the storage shelf. The simulation results show that the PID controllers can respond smoothly to the desired angular velocity and the adaptive controller can adapt quickly and correctly to the desired height.