• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.11
• /
• pp.1034-1037
• /
• 2015

In recent years in the field of robotics, various methods have been developed to create an intimate relationship between people and robots. These methods include speech, vision, and biometrics recognition as well as gesture-based interaction. These recognition technologies are used in various wearable devices, smartphones and other electric devices for convenience. Among these technologies, gesture recognition is the most commonly used and appropriate technology for wearable devices. Gesture recognition can be classified as contact or noncontact gesture recognition. This paper proposes contact gesture recognition with IMU and EMG sensors by using the hidden Markov model (HMM) twice. Several simple behaviors make main gestures through the one-stage HMM. It is equal to the Hidden Markov model process, which is well known for pattern recognition. Additionally, the sequence of the main gestures, which comes from the one-stage HMM, creates some higher-order gestures through the two-stage HMM. In this way, more natural and intelligent gestures can be implemented through simple gestures. This advanced process can play a larger role in gesture recognition-based UX for many wearable and smart devices.

• Journal of Sensor Science and Technology
• /
• v.29 no.6
• /
• pp.440-446
• /
• 2020

A well-known difficulty in attitude estimation based on inertial measurement unit (IMU) signals is the occurrence of external acceleration under dynamic motion conditions, as the acceleration significantly degrades the estimation accuracy. Lee et al. (2012) designed a Kalman filter (KF) that could effectively deal with the acceleration issue. Ahmed and Tahir (2017) modified this method by adjusting the acceleration-related covariance matrix because they considered covariance modeling as a pivotal factor in the estimation accuracy. This study investigates the effects of covariance modeling on estimation accuracy in an IMU-based attitude estimation KF. The method proposed by Ahmed and Tahir can be divided into two: one uses the covariance including only diagonal components and the other uses the covariance including both diagonal and off-diagonal components. This paper compares these three methods with respect to the motion condition and the window size, which is required for the methods by Ahmed and Tahir. Experimental results showed that the method proposed by Lee et al. performed the best among the three methods under relatively slow motion conditions, whereas the modified method using the diagonal covariance with a high window size performed the best under relatively fast motion conditions.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.23 no.6
• /
• pp.633-641
• /
• 2015

The ROM (roll over mitigation) system is a next-generation suspension system that can improve vehicle-driving stability and ride comfort. Currently, mass-produced safety systems, such as ESC (electronic stability control) and ECS (electronic control suspension), enable measurements of longitudinal and lateral acceleration as well as yaw rate through inertial sensor clusters, but they lack direct measurements of the roll angle. Therefore, in this paper, a roll angle estimation algorithm from ESC system sensors and tire normal force has been proposed. Furthermore, this study presents a method for roll over mitigation force distribution between the front and rear of a ROM system. Performance and reliability of the roll angle estimation and roll over mitigation force distribution were investigated through simulations. The simulation results showed that the proposed control algorithm and strategy are reliable during vehicle rollovers.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.37 no.10
• /
• pp.992-997
• /
• 2009

This paper focuses on the calculation of the missile angular velocity under the reduced sensor condition and its verification using the Flight Motion Simulator(FMS). The missile angular velocity is usually measured by the body gyroscopes, but we assume that the inertial sensors on the missile body are in the absence of pitch and yaw gyroscopes. Under this reduced sensor condition, this paper shows the missile angular velocity can be calculated by using the gimbal seeker gyroscope, the roll body gyroscope, the gimbal angle and its rate. The FMS experiment was carried out to verify the proposed algorithm.

• Journal of Mechanical Science and Technology
• /
• v.19 no.6
• /
• pp.1253-1267
• /
• 2005

Misalignment can be an important problem in the integration of GPS/INS. Observability analysis of the alignment errors in the integration of low-grade inertial sensors and multi-antenna GPS is presented in this paper. A control-theoretic approach is adopted to study the observability of time-varying error dynamics models. The relationship between vehicle motions and the observability of the errors in the lever arm and relative attitude between GPS antenna array and IMU is given. It is shown that alignment errors can be made observable through maneuvering. The change of acceleration makes the components of the relative attitude error that are orthogonal to the direction of the acceleration change observable. The change of angular velocity makes the components of the lever arm error that are orthogonal to the direction of the angular velocity observable. The motion of constant angular velocity has no influence on the estimation of the lever arm.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.709-711
• /
• 2004

본 논문은 3차원 상에서 사용자의 동작을 관성센서로 입력받아 제스처를 인식하는 시스템을 소개한다. 사용자가 취한 제스처 동작은 관성 센서를 통하여 각속도 및 가속도 신호열로 변환된다. 궤적 추정 알고리즘은 이를 2차원 상의 동작 궤적으로 변환한다. 인식 알고리즘은 이 동작 궤적을 입력받아 베이지안 네트웍에 기반한 제스처 모델들로부터의 likelihood를 계산한 후, 최대 likelihood를 갖는 모델을 선택하여 인식을 수행한다. 16명의 필자로부터 13개의 제스처 동작을 각 24회씩 수집하여 실험한 결과 평균 99.4%의 인식률을 얻었다.

• Journal of Sensor Science and Technology
• /
• v.28 no.2
• /
• pp.81-87
• /
• 2019

One of the major issues in inertial and magnetic measurement unit (IMMU)-based 3D orientation estimation is compensation for magnetic disturbances in magnetometer signals, as the magnetic disturbance is a major cause of inaccurate yaw estimation. In the proposed approach, a kinematic constraint is used to provide a measurement equation in addition to the accelerometer and magnetometer signals to mitigate the disturbance effect on the orientation estimation. Although a Kalman filter (KF) is the most popular framework for IMMU-based orientation estimation, a complementary filter (CF) has its own advantages over KF in terms of mathematical simplicity and ease of implementation. Accordingly, this paper introduces a quaternion-based CF with a constraint-combined correction equation. Furthermore, the weight of the constraint relative to the magnetometer signal is adjusted to adapt to magnetic environments to optimally deal with the magnetic disturbance. In the results of our validation experiments, the average and maximum of yaw errors were $1.17^{\circ}$ and $1.65^{\circ}$ from the proposed CF, respectively, and $8.88^{\circ}$ and $14.73^{\circ}$ from the conventional CF, respectively, showing the superiority of the proposed approach.

• Korean Journal of Applied Biomechanics
• /
• v.29 no.1
• /
• pp.1-8
• /
• 2019

Objective: The purpose of this study was to confirm that the cross-country ski sprint course in PyeongChang, where the 2018 Winter Olympics course was to utilize wearable devices equipped with inertial measurement unit (IMU), global positioning system (GPS) and heart rates sensor. Method: For the data collection, two national level cross-country (XC) skiers performed classic technique on the entire sprint course. We analyzed cycle characteristics, range of motion on double poling (DP) technique, average velocity, and displacement of 3 points according to the terrain. Results: The absolute cycle time gradually decreased during starting, middle and finish sections. While the length of the DP increased and the heart rates tended to increase for men skier. In addition, the results indicated that range of motion of knee joint during starting and finish section decreased more than middle section. The errors of latitude and longitude data collected through GPS were within 3 m from 3 points. Conclusion: Through the first case study in Korea, which analyzed the location and condition of XC skiers in the entire sprint course in real time, confirmed that feedback was available in the field using various wearable sensors.

• Journal of Positioning, Navigation, and Timing
• /
• v.8 no.2
• /
• pp.49-58
• /
• 2019

A number of techniques have been studied to estimate the position of pedestrians in indoor space. Among them, the technique of estimating the position using only the sensors attached to the body of the pedestrian without using the infrastructure is regarded as a very important technology for special purpose pedestrians such as the firefighters. In particular, it forms a research field under the name of Pedestrian Dead Reckoning (PDR). In this paper, we focus on a method for step detection which is essential when performing PDR using Inertial Measurement Unit (IMU) mounted on a shoe. Many researches have been done to detect the stance phase where the foot contacts the ground. Most of these methods, however, have a way to detect the specific size of the sensor signal and require thresholds for these methods. This has the difficulty of changing these thresholds if the user is different. To solve this problem, we propose a stance phase detection method that does not require any threshold value. It is expected that this result will make it easier to commercialize the technology because PDR can be implemented without user-dependent parameter setting.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.5
• /
• pp.289-295
• /
• 2021

Recently, owing to the development of autonomous vehicles, research on precisely determining the position of a moving object has been actively conducted. Previous research mainly used the fusion of GNSS/IMU (Global Positioning System / Inertial Navigation System) and sensors attached to the vehicle through a Kalman filter. However, in recent years, new technologies have been used to determine the location of a moving object owing to the improvement in computing power and the advent of deep learning. Various techniques using RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), and NARX (Nonlinear Auto-Regressive eXogenous model) exist for such learning-based positioning methods. The purpose of this study is to compare the precision of existing filter-based sensor fusion technology and the NARX-based method in case of GNSS signal blockages using simulation data. When the filter-based sensor integration technology was used, an average horizontal position error of 112.8 m occurred during 60 seconds of GNSS signal outages. The same experiment was performed 100 times using the NARX. Among them, an improvement in precision was confirmed in approximately 20% of the experimental results. The horizontal position accuracy was 22.65 m, which was confirmed to be better than that of the filter-based fusion technique.