• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2433-2435
• /
• 2000

Inertial Navigation System(INS) provides short-term accurate navigation solution but its error grows with time due to integration characteristics. Meanwhile, Global Positioning System(GPS) provides long-term stable solution but it has poor error characteristics in high dynamic region. So for its synergistic relationship, an integrated INS/GPS systems has been widely used as an advanced navigation system. Generally, two kinds of integration method are used. One is loosely coupled mode which uses GPS-derived position and velocity as measurements in an integrated Kalman filter. The other is tightly coupled one which uses pseudorange and pseudorange rate as Kalman filter measurements. In this paper the system error models and observation models for two kinds of integrated systems are derived, respectively, and their performance are compared through Monte-Carlo simulations.

• 한국해안해양공학회지
• /
• 제8권3호
• /
• pp.221-230
• /
• 1996

기존의 전유공 케이슨 방파제로부터의 파 반사를 계산하기 위하여 개발된 Suh and Park의 이론 모형을, 케이슨 전면 하부의 연직벽을 경사가 매우 급한 경사벽으로 가정함으로써, 부분 유공 케이슨 방파제에 적용하였다. 또한, 이 모형에서, Kano and Liu가 제안한 차단계수를 이용하여 유공벽에서의 관성저항항을 수정하였다. 이 모형을 1993년도에 보고된 Park et al.의 수리실험 자료와 비교해 본 결과, 실험 자료 및 이론 모형 결과에서 모두 관성저항의 영향이 중요하여 B/L$_{c}$가 약 0.2일 때 반사율이 최소가 됨을 보이는데 (여기서 B=유수실의 폭, L$_{c}$=유수설 내에서의 파장), 이 값은 찬성저항의 영향을 무시했을 때 얻어지는 값 0.25보다 약간 작은 값이다. 또한 선형파 이론에 근거한 이 모형은 파의 비선형성이 증가함에 따라 반사율을 크게 계산하는 경향이 있음을 보이며, 따라서 이 모형은 파형경사가 작은 통상파에 적용하는 것이 바람직함을 알 수 있었다.있었다.

• 한국전자통신학회논문지
• /
• 제15권2호
• /
• pp.291-298
• /
• 2020

수화는 청각 장애인이 다른 사람들과 의사소통할 수 있도록 설계된 것이다. 그러나 수화는 충분히 대중화되어 있지 않기 때문에 청각 장애인이 수화를 통해서 일반 사람들과 원활하게 의사소통하는 것은 쉽지 않은 문제이다. 이러한 문제점에 착안하여 본 논문에서는 웨어러블 컴퓨팅 및 딥러닝 기반 미국식 수화인식 시스템을 설계하고 구현하였다. 이를 위해서 본 연구에서는 손등과 손가락에 장착되는 총 6개의 IMUs(Inertial Measurement Unit) 센서로 구성된 시스템을 구현하고 이를 이용한 실험을 수행하여 156개 특징이 수집된 데이터 추출을 통해서 총 28개 단어에 대한 미국식 수화 인식 방법을 제안하였다. 특히 LSTM (Long Short-Term Memory) 알고리즘을 사용하여 최대 99.89%의 정확도를 달성할 수 있었고 향후 청각 장애인들의 의사소통에 큰 도움이 될 것으로 예상된다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 E
• /
• pp.1840-1842
• /
• 1998

Gyroscope is a very important core sensor, as a rotation sensor in inertial space, in inertial guidance and navigation system on aeronautics, plane, vessel and so on for civilian and millitary applications. Mechnical gyroscopes, adopting a principle of spinning a top, have been used in many application system. These mechnical gyroscopes need high power consumption, long warming time and complicated peripheral devices. But fiber-optic gyroscopes, based on the Sagnac effect, have novel advantages as small volume. simple scheme, low power consumption and high reliability. So we have developed a Intermediate grade All-fiber Optic Gyroscope, which has open-loop and minimum reciprocal configuration scheme. We have designed feedback circuits for stability of amplitude and phase using four lock-in amplifier(LIA) circuits and also used for noise limitation. This paper describes the scheme of optical part and electronic part and also test results of this all-fiber optic gyroscope. The performance have been achieved as long-term bias drift of $9.54^{\circ}/h$, random walk of $0.0317^{\circ}/\sqrt{h}$ and dynamic range of ${\pm}150\;deg/s$.

• 로봇학회논문지
• /
• 제18권4호
• /
• pp.427-435
• /
• 2023

This paper proposes a method of estimating the pose of a mobile robot by using a learning model. When estimating the pose of a mobile robot, wheel encoder and inertial measurement unit (IMU) data are generally utilized. However, depending on the condition of the ground surface, slip occurs due to interaction between the wheel and the floor. In this case, it is hard to predict pose accurately by using only encoder and IMU. Thus, in order to reduce pose error even in such conditions, this paper introduces a pose estimation method based on a learning model using data of the wheel encoder and IMU. As the learning model, long short-term memory (LSTM) network is adopted. The inputs to LSTM are velocity and acceleration data from the wheel encoder and IMU. Outputs from network are corrected linear and angular velocity. Estimated pose is calculated through numerically integrating output velocities. Dataset used as ground truth of learning model is collected in various ground conditions. Experimental results demonstrate that proposed learning model has higher accuracy of pose estimation than extended Kalman filter (EKF) and other learning models using the same data under various ground conditions.

• International Journal of Fluid Machinery and Systems
• /
• 제2권4호
• /
• pp.431-438
• /
• 2009

A Large Eddy Simulation (LES) of the flow in an inducer is carried out under flow rate oscillations. The present study focuses on the dynamic response of the backflow and the unsteady pressure performance to the flow rate fluctuations under non-cavitation conditions. The amplitude of angular momentum fluctuation evaluated by LES is larger than that evaluated by RANS. However, the phase delay of backflow is nearly the same as RANS calculation. The pressure performance curve exhibits a closed curve caused by the inertia effect associated with the flow rate fluctuations. Compared with simplified one dimensional evaluation of the inertia component, the component obtained by LES is smaller. The negative slope of averaged performance curve becomes larger under unsteady conditions. From the conservations of angular momentum and energy, an expression useful for the evaluation of unsteady pressure rise was obtained. The examination of each term of this expression show that the apparent decrease of inertia effects is caused by the response delay of Euler's head and that the increase of negative slope is caused by the delay of inertial term associated with the delay of backflow response. These results are qualitatively confirmed by experiments.

• Smart Structures and Systems
• /
• 제2권3호
• /
• pp.209-224
• /
• 2006

The use of smart dampers to optimally control the response of structures is on the increase. To maximize the potential use of such damper systems, their accurate modeling and assessment of their performance is of vital interest. In this study, the performance of a controllable fluid dashpot damper, in terms of damper forces, damper dynamic range and damping force hysteretic loops, respectively, is studied mathematically. The study employs a damper Bingham-Maxwell (BingMax) model whose mathematical formulation is developed using a Fourier series technique. The technique treats this one-dimensional Navier-Stokes's momentum equation as a linear superposition of initial-boundary value problems (IBVPs): boundary conditions, viscous term, constant Direct Current (DC) induced fluid plug and fluid inertial term. To hold the formulation applicable, the DC current level to the damper is supplied as discrete constants. The formulation and subsequent simulation are validated with experimental results of a commercially available magneto rheological (MR) dashpot damper (Lord model No's RD-1005-3) subjected to a sinusoidal stroke motion using a 'SCHENK' material testing machine in the Materials Laboratory at the University of Technology, Sydney.

• 한국군사과학기술학회지
• /
• 제2권2호
• /
• pp.186-196
• /
• 1999

The CPS provides data with long-term stability independent of passed time and the INS provides high-rate data with short-term stability. By integrating these complementary systems, a highly accurate navigation system can be achieved. In this paper, a loosely-coupled GPS/INS integration system is designed. It is a simple structure and is easy to implement and preserves independent navigation capability of GPS and INS. The integration system consists of a NCU, an IMU, a GPS receiver, and a monitoring system. The navigation algorithm in the NCU is designed under the multi-tasking environment based on a real-time kernel system and the monitoring system is designed using the Visual C++. The integrated Kalman filter is designed as a feedback formed 15-state filter, in which the states are position errors, velocity errors, attitude errors and sensor bias errors. The van test result shows that the integrated system provides more accurate navigation solution then the inertial or the GPS-alone navigation system.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2007년도 춘계학술대회B
• /
• pp.2261-2266
• /
• 2007

In this study, the numerical analysis to estimate condensation heat and mass transfer of the condenser was carried out using the PMA (porous medium approach). In the PMA, the details of tube bundle in the condenser are replaced by the porous medium, and the flow resistance term is added in the momentum equation. In this regard, the PMA is quite helpful for the study of tube bundle in the large condenser. The pressure loss through tube bundle can be compensated by viscous and inertial momentum sink terms, which was validated numerically. Value of the pressure drop was compared to that of Butterworth correlation. Three dimensional analysis of condensation for McAllister condenser with the PMA was conducted using Fluent 6.2 and UDFs (use-defined functions). The result of condensation rate was analogous to previous results (experimental and numerical data).

• Structural Engineering and Mechanics
• /
• 제66권5호
• /
• pp.649-663
• /
• 2018

The scaled boundary finite element method is extended to evaluate the dynamic stress intensity factors and T-stress with a numerical procedure based on the improved continued-fraction. The improved continued-fraction approach for the dynamic stiffness matrix is introduced to represent the inertial effect at high frequencies, which leads to numerically better conditioned matrices. After separating the singular stress term from other high order terms, the internal displacements can be obtained by numerical integration and no mesh refinement is needed around the crack tip. The condition numbers of coefficient matrix of the improved method are much smaller than that of the original method, which shows that the improved algorithm can obtain well-conditioned coefficient matrices, and the efficiency of the solution process and its stability can be significantly improved. Several numerical examples are presented to demonstrate the increased robustness and efficiency of the proposed method in both homogeneous and bimaterial crack problems.