• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.72-75
• /
• 2004

In this paper, a finger-click recognition method is proposed to improve the recognition performance for finger-clicking of a wearable input device, called $SCURRY^{TM}$. The proposed method is composed of three parts including feature extraction part, valid click discrimination part, and cross-talk avoidance part. Two types of MEMS inertial sensors are embedded into the wearable input device to measure the angular velocity of a hand (hand movement) and the acceleration rates at the ends of fingers (finger-click motion). The experiment applied to the $SCURRY^{TM}$ device shows the improved stability and performance.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.141.6-141
• /
• 2001

This paper presents the attitude algorithm using Fuzzy technique to mix gyro information with accelerometer. The attitude angle calculated by the low-cost gyros only increases its error with time rapidly because of the integration process of the algorithm and large sensor error. It is known that the accelerometer output includes the attitude information of a vehicle and its information is more effective during low dynamic maneuver. Therefore it is needed to combine two information appropriately for obtaining the attitude information from low-cost MEMS inertial sensors. Because Fuzzy logic is very effective to make a decision of maneuvering state, it is applied to the mixed algorithm. It is shown by experiment ...

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.439-439
• /
• 2000

In this paper, the relationship among the vertical acceleration, measuring points and walking patterns is analyzed. To measure acceleration, the acceleration measurement unit and communication board is constructed. It uses MEMS accelerometer ADXL-202 that detects 2-axis acceleration simultaneously. It is shown by the experiment test that the walking pattern is recognized and walking step is detected at easy when acceleration measurement unit is mounted on leg.. This results can be directly utilized in designing the personal navigation system with low-cost inertial sensor.

• 한국전자통신학회논문지
• /
• 제10권4호
• /
• pp.469-475
• /
• 2015

본 논문은 반도체 센서를 이용하여 공간상 시스템의 자세를 정확히 측정할 수 있도록 확장 칼만 필터를 설계하는 방법에 관한 연구이다. 공간상 자세는 관성좌표계(고정 좌표계)로부터 몸체에 부착된 회전좌표계의 상호 관계로 표현한다. 자세를 표현하는데 있어서 간결한 방법인 쿼터니언을 상태변수로 이용하며, 속도 센서로부터 계측된 값을 입력으로 가정하고, 상태 변화를 추정하였다. 그리고 가속도 센서로부터 획득된 값을 관측 데이터로 하여 추정한 값과의 정합과정을 통해 최적의 추정치를 얻어낸다. 이때 추정의 정밀도를 높이기 위해 추정 주기를 센서특성에 맞춰 조절하도록 확장 칼만 필터를 설계하였다. 그 결과, 3축 속도 센서와 3축 가속도 센서를 이용하여 설계된 추정기의 RMS(: Root Mean Square) 추정오차가 시뮬레이션에서 약 1.7 [$^{\circ}$] 이하로 유지되었고, 실험에서 100 [ms] 의 주기로 상태추정을 함으로 추정기가 유용함을 입증하였다.

• Journal of Acupuncture Research
• /
• 제30권4호
• /
• pp.25-33
• /
• 2013

Objectives : To assess the test-retest reliability and the intratest repeatability in measuring the cervical range of motion of healthy subjects with wireless microelectromechanical system inertial measurement unit(MEMS-IMU) system and to discuss the feasibility of this system in the clinical setting to evaluate the cervical spine musculoskeletal. Methods : 12 healthy people who were evaluated as no- or mild-disability with neck disability index were participated. Their cervical motion were measured with IMU twice in consecutive two days for the test-retest reliability study. Intratest repeatability was calculated in the two tests separately. The calculated intraclass correlation coefficients(ICC) were discussed and compared with the those of the previous studies. Results : Cervical range of motion data were acquired and statistically processed: left rotation($61.64^{\circ}$), right rotation($65.12^{\circ}$), extension($61.98^{\circ}$), flexion($52.81^{\circ}$), left bending($39.31^{\circ}$), right bending($41.08^{\circ}$). ICCs were 0.77~0.98(intratest repeatability) and 0.74~0.93 (test-retest reliability) in the primary motion. In the coupling motion, intratest repeatability ICCs were 0.93~ 0.99(transverse primary plane), 0.88~0.97(saggital primay plane), and 0.77~0.93(coronal primary plane). Test-retest reliability of coupling motion were 0.90~0.97(transverse primary plane), 0.00~0.72(saggital primary plane), and 0.04~0.76(coronal primary plane). Conclusions : Several types of range-of-motion devices are now on use in many fields including medicine, but the practicality of the devices in clinical use is questionable for the convenient and economical aspects. In this study, we presented the reliability of cervical range of motion test with the developed wireless MEMS-IMU system and discussed its potential utility in clinical use.

• 한국항공우주학회지
• /
• 제43권7호
• /
• pp.609-618
• /
• 2015

CNUSAIL-1은 태양돛을 탑재한 3U 크기의 큐브위성이다. 저궤도에서 태양돛을 전개하고, 이에 따른 자세와 궤도에 대한 영향을 확인하는 임무를 수행한다. 본 논문에서는 CNUSAIL-1을 위한 자세결정 알고리즘의 구현 가능성을 제시하였다. 위성의 기준센서는 태양센서, 3축 지자기센서를 이용하며, 관성센서는 MEMS 자이로센서를 사용한다. 큐브위성용 센서는 상대적으로 저가이며, 성능 및 잡음특성이 좋지 않은 단점이 있다. 따라서 자세결정 알고리즘으로 노이즈 특성을 고려할 수 있는 확장칼만필터를 적용하였다. 또한 자세결정의 결정론적 방법인 QUEST 알고리즘과 비교하여 그 타당성을 검증하였다.

• 센서학회지
• /
• 제26권4호
• /
• pp.235-238
• /
• 2017

In this study, the design of a tri-axis micromachined gyroscope is proposed and the vibration characteristic of the structure is analyzed. Tri-axis vibratory gyroscopes that utilize Coriolis effect are the most commonly used micromachined inertial sensors because of their advantages, such as low cost, small packaging size, and low power consumption. The proposed design is a single structure with four proof masses, which are coupled to their adjacent ones. The coupling springs of the proof masses orthogonally transfer the driving vibrational motion. The resonant frequencies of the gyroscope are analyzed by finite element method (FEM) simulation. The suspension beam spring design of proof masses limits the resonance frequencies of four modes, viz., drive mode, pitch, roll and yaw sensing mode in the range of 110 Hz near 21 kHz, 21173 Hz, 21239 Hz, 21244 Hz, and 21280 Hz, respectively. The unwanted modes are separated from the drive and sense modes by more than 700 Hz. Thereafter the drive and the sense mode vibrations are calculated and simulated to confirm the driving feasibility and estimate the sensitivity of the gyroscope. The cross-axis sensitivities caused by driving motion are 1.5 deg/s for both x- and y-axis, and 0.2 deg/s for z-axis.