• 센서학회지
• /
• 제31권6호
• /
• pp.423-427
• /
• 2022

A pressure sensor that mimics the sensing ability of human skin has emerged as high-profile technology because it shows remarkable applications in numerous fields such as robotics, human health monitoring, and artificial prosthetics. Whereas recent pressure sensors have achieved high sensitivity similar to that of human skin, they still show limited detection bandwidth. Moreover, once these e-skin are fabricated, their sensitivity and stiffness are fixed; therefore, they can be used for only limited applications. Our study proposes a new adaptive pressure sensor built with uniform gallium microdroplet-elastomer composite. Based on the phase transition of gallium microdroplets, the proposed sensor undergoes mode transformation, enabling it to have a higher sensitivity and wider detection bandwidth compared with those of human skin. In addition, we succeeded in extending a single adaptive pressure sensor to sensor arrays based on its high uniformity, reproducibility, and large-scale manufacturability. Finally, we designed an adaptive e-skin with the sensor array and demonstrated its applications on health monitoring tasks including blood pulse and body weight measurements.

• 대한의용생체공학회:의공학회지
• /
• 제29권3호
• /
• pp.239-248
• /
• 2008

In this study, a new method using a capacitive sensor and an adaptive filter was proposed to deal with artifacts contaminating an oscillation signal in oscilometric blood pressure measurement. The proposed method makes use of a variation of the capacitance between an electrode fixed to a cuff and an external object to detect artifacts caused by the external object bumping into the cuff. The proposed method utilizes the adaptive filter based on linear prediction to remove the detected artifacts. The conventional method using linear interpolation and the proposed method using the adaptive filter were applied to three types of the artifact-contaminated oscillation signals(no overlap, non-consecutive overlap, and consecutive overlap between artifacts and oscillations) to compare them in terms of the artifact reduction performance. The proposed method was more robust than the conventional method in the case of consecutive overlap between artifacts and oscillations. The proposed method could be useful for measuring blood pressure in such a noisy environment that the subject is being transported.

• 한국음향학회지
• /
• 제35권5호
• /
• pp.395-402
• /
• 2016

본 논문에서는 음향벡터 선배열 센서 기반에서의 광대역 적응빔형성기법을 다룬다. 적응 빔형성을 위하여 안정적인 공분산행렬추정은 매우 중요한 문제이다. 기존의 코히어런트 신호부공간기반의 적응 빔형성기법은 초점조정행렬(focusing matrix) 추정으로 인해 방위각 추정에 오차가 발생하며 또한 공분산행렬 추정을 위하여 많은 데이터 단편을 필요로 한다. 방위각 추정오차 및 공분산 행렬 추정시 필요한 데이터 단편의 수 문제를 완화하기 위하여 음압센서 선배열에 적용된 조향공분산 행렬 기법을 음향벡터 선배열 센서에 확장하여 적용한다. 그리고 부배열 기법을 통하여 공분산행렬의 차원을 줄임으로써 적은 수의 데이터 단편으로 안정적인 공분산행렬 추정이 가능하고 방위각 추정성능을 향상시킨다. 모의 실험을 통하여 기존의 코히어런트 신호 부공간 전처리 기반 광대역 빔형성기법과 제안한 기법의 방위각 추정 성능을 분석한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(4)
• /
• pp.263-266
• /
• 2000

We proposes a new fingerprint minutia matching algorithm which matches the fingerprint minutiae by using adaptive distance. In general, fingerprint is deformed by pressure and orientation when a user press his fingerprint to sensor. These nonlinear deformations change the distance between minutiae and reduce verification rate. We define the adaptive distance using ridge frequency. Adaptive distance normalizes the distance between minutiae and compensates for nonlinear deformation. Our algorithm can distinguish two different fingerprints better and is more robust. Experimental results show that the performance of the proposed algorithm is superior to using Euclidean distance.

• International Journal of Korean Welding Society
• /
• 제1권1호
• /
• pp.1-6
• /
• 2001

This paper describes several advances in sensor and process control techniques for applications in Submerged Arc Welding (SAW), which combine to give a fully automatic system capable of controlling and adapting the overall welding process. This technology has been applied in longitudinal and spiral pipe mills and in pressure vessel production.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2002년도 Proceedings of the International Welding/Joining Conference-Korea
• /
• pp.336-341
• /
• 2002

Recent developments in sensor hardware and in advanced software have made it feasible to consider automating some of the most difficult welding operations. This paper describes some techniques used to automate successfully multipass submerged arc welding operations typically used in pressure vessel manufacture, shipbuilding, production of offshore structures and in pipe mills.

• 센서학회지
• /
• 제19권5호
• /
• pp.375-384
• /
• 2010

In this paper, we proposed a new shaft distortion correction system having an adaptive PID controller using displacement sensors, which is adaptively reflecting variations of shaft strength owing to irregular heat treatment during an annealing process and sensitivity to the seasonal temperature changes. Generally, the shafts are annealed by heat treatment in order to enlarge the strength of the shaft, which causes an distortion of a shaft such as irregular bending of the shaft. In order to correct such a distortion of the shaft, a mechanical pressure is properly impacted to the distorted shaft. However, the strength of every shaft is different from each other owing to irregular annealing and seasonal temperature changes. Especially, the strength of a thin shaft such as a car transmission shaft is much more sensitive than that of a thick shaft. Therefore, it is very important for considering the strength of each shaft during correction of the car transmission shaft distortion in order to generate proper mechanical pressure. The conventional PID controller for the shaft distortion correction system does not consider each different strength of each shaft, which causes low productivity. Therefore, we proposed a new PID controller considering variations of shaft strength caused by seasonal temperature changes as well as irregular heat treatment and different cooling time. Three displacement sensors are used to measure a degree of distortion of the shaft at three different location. The proposed PID controller generates adaptively different coefficients according to different strength of each shaft using appropriately obtained pressure times from long-term experiments. Consequently, the proposed shaft distortion correction system increases the productivity about 30 % more than the conventional correction system in the real factory.

• 대한기계학회논문집A
• /
• 제40권1호
• /
• pp.113-120
• /
• 2016

본 연구에서는 좌/우 편제동을 통해 차량의 요 모션을 제어하는 제동 기반 요모멘트 제어 시스템에서의 가상 제동 압력 센서를 개발하였다. 제동 압력을 추정하기 위해 유압시스템을 경험적 방법으로 모델링하였고 이를 기반으로 요레이트 피드백 제동 압력 관측기를 설계하였다. 차량 요레이트 동역학에 존재하는 외란의 영향을 최소화 하기 위해 외란 적응 기법, 외란 축소 기법 및 최적 이득 기법을 관측기 설계에 적용하였고 그 방법들 간의 성능 비교 및 검증을 HILS 를 통해 수행하였다. 그 결과 외란 축소 방식의 견실 관측기의 압력 추정 성능이 일반적인 Luenberger 관측기 대비 가장 우수하였으며 그 원인에 대해 분석하였다.

• Journal of information and communication convergence engineering
• /
• 제7권3호
• /
• pp.412-417
• /
• 2009

With ubiquitous computing aid, it can improve human being's life quality if all people have more convenient medical service under pervasive computing environment. In this paper, for a pervasive health care application for diabetes patient, we've implemented a health care system, which is composed of three parts. Various sensors monitor both outer and inner environment of human such as temperature, blood pressure, pulse, and glycemic index, etc. These sensors form zigbee-based sensor network. And as a backend, medical information server accumulates sensing data and performs back-end processing. To simply transfer these sensing values to a medical team may be a low level's medical service. So, we've designed a model with context awareness for more improved medical service which is based on ART(adaptive resonance theory) neural network. Our experiments show that a proposed healthcare system can provide improved medical service because it can recognize current context of patient more concretely.

• 소음진동
• /
• 제7권3호
• /
• pp.427-437
• /
• 1997

Mean active intensity based active control for the cancellation of radiated noise out of the duct exit is studied. The active intensity control strategy is drerived based on the relation of the exterior sound field out of the duct termination and interior sound field of the duct. One of the characteristics of this control strategy is that the control performance can be maintained regardless of the sensor loction, compared with the conventional local pressure control methods at either interior downstream or exterior field positions. It is also suggested that the digital filtering for the active intensity control can be achieved by time-domain filtered-x LMP (Lest-Mean-Product) adaptive algorithm. Experiments for an open-ended duct are performed to compare the active intensity control performance with conventional pressure control one. Active control experiment of local sound pressure is conducted by widely used filtered-x LMS adaptive Algorithm and active intensity control implementaion uses the derived filter d-x LMP algorithm. It is shown that the exterior sound fileds was much better observable by sensing of the active intensity than by just sound pressure. It is also demonstrated that the global control performance of external field by acoustic intensity is superior to the conventional sound pressure control performance.