지자기센서를 이용하지 않는 6축 IMU 기반의 3차원 관절각 추정용 순환 신경망

이창준,김우재,이정근 한국정밀공학회 2023 한국정밀공학회지 Vol.40 No.4

Inertial measurement unit (IMU)-based 3D joint angle estimation have a wide range of important applications, among them, in gait analysis and exoskeleton robot control. Conventionally, the joint angle was determined via the estimation of 3D orientation of each body segment using 9-axis IMUs including 3-axis magnetometers. However, a magnetometer is limited by magnetic disturbance in the vicinity of the sensor, which highly affects the accuracy of the joint angle. Accordingly, this study aims to estimate the joint angle using the 6-axis IMU signals composed of a 3-axis accelerometer and a 3-axis gyroscope without a magnetometer. This paper proposes a recurrent neural network (RNN) model, which indirectly utilizes the joint kinematic constraint and thus estimates joint angles based on 6-axis IMUs without using a magnetometer signal. The performance of the proposed model was validated for a mechanical joint and human elbow joint, under magnetically disturbed environments. Experimental results showed that the proposed RNN approach outperformed the conventional approach based on a Kalman filter (KF), i.e., RNN 3.48o vs. KF 10.01o for the mechanical joint and RNN 7.39o vs. KF 21.27o for the elbow joint.

FMCW MIMO 레이다를 이용한 거리-각도 동시 추정 기법

김정훈,송성찬,전주환 한국전자파학회 2019 한국전자파학회논문지 Vol.30 No.2

Frequency-modulated continuous wave(FMCW) radars with array antennas are widely used because of their light weight and relatively high resolution. A usual approach for the joint range and angle estimation of a target using an array FMCW radar is to create a range-angle matrix with the deramped received signal, and subsequently apply two-dimensional(2D) frequency estimation methods such as 2D fast Fourier transform on the range-angle matrix. However, such frequency estimation approaches cause bias errors since the frequencies in the range-angle matrix are not independent. Therefore, we propose a new maximum likelihood-based algorithm for joint range and angle estimation of targets using array FMCW radar, and demonstrate that the proposed algorithm achieves the Cramér-Rao bounds, both for range as well as angle estimation.

Kalman Randomized Joint UKF Algorithm for Dual Estimation of States and Parameters in a Nonlinear System

Behrouz Safarinejadian,Navid Vafamand 대한전기학회 2015 Journal of Electrical Engineering & Technology Vol.10 No.3

This article presents a new nonlinear joint (state and parameter) estimation algorithm based on fusion of Kalman filter and randomized unscented Kalman filter (UKF), called Kalman randomized joint UKF (KR-JUKF). It is assumed that the measurement equation is linear. The KRJUKF is suitable for time varying and severe nonlinear dynamics and does not have any systematic error. Finally, joint-EKF, dual-EKF, joint-UKF and KR-JUKF are applied to a CSTR with cooling jacket, in which production of propylene glycol happens and performance of KR-JUKF is evaluated.

유연관절로봇을 위한 정확한 외부토크 측정시스템 개발: 랜덤워크모델을 이용한 칼만필터 기반 추정

박영진,정완균 한국로봇학회 2014 로봇학회 논문지 Vol.9 No.1

In this paper, an external torque estimation problem in one-degree-of-freedom (1-DOF) flexible-joint robot equipped with a joint-torque sensor is revisited. Since a sensor torque from the jointtorque sensor is distorted by two dynamics having a spring connection, i.e., motor dynamics and link dynamics of a flexible-joint robot, a model-based estimation, rather than a simple linear spring model, should be required to extract external torques accurately. In this paper, an external torque estimation algorithm for a 1-DOF flexible-joint robot is proposed. This algorithm estimates both an actuating motor torque from the motor dynamics and an external link torque from the link dynamics simultaneously by utilizing the flexible-joint robot model and the Kalman filter estimation based on random-walk model. The basic structure of the proposed algorithm is explained, and the performance is investigated through a custom-designed experimental testbed for a vertical situation under gravity.

Lithium-ion battery state of charge and parameters joint estimation using cubature Kalman filter and particle filter

Xu, Wei,Xu, Jinli,Yan, Xiaofeng The Korean Institute of Power Electronics 2020 JOURNAL OF POWER ELECTRONICS Vol.20 No.1

Accurate estimation of the state of charge (SOC) of a lithium-ion battery is one of the most crucial issues of battery management system (BMS). Existing methods can achieve accurate estimation of the SOC under stable working conditions. However, they may result in inaccuracy under unstable working conditions such as dynamic cycles and different temperature conditions. This is due to the fact that the dynamic behaviors of battery states have not been considered by the parameter identification methods. In this paper, a SOC and parameter joint estimation method is put forward, where the battery model parameters are identified in real time by a particle filter (PF) with consideration of the battery states. Meanwhile, a cubature Kalman filter (CKF) is used to estimate SOC. Then, experiments under dynamic cycles and different temperature conditions are undertaken to assess the performance of the proposed algorithm when compared with the existing joint estimations. The results show that the proposed joint method can achieve a high accuracy and robustness for SOC estimation.

Robust 2D human upper-body pose estimation with fully convolutional network

Lee, Seunghee,Koo, Jungmo,Kim, Jinki,Myung, Hyun Techno-Press 2018 Advances in robotics research Vol.2 No.2

With the increasing demand for the development of human pose estimation, such as human-computer interaction and human activity recognition, there have been numerous approaches to detect the 2D poses of people in images more efficiently. Despite many years of human pose estimation research, the estimation of human poses with images remains difficult to produce satisfactory results. In this study, we propose a robust 2D human body pose estimation method using an RGB camera sensor. Our pose estimation method is efficient and cost-effective since the use of RGB camera sensor is economically beneficial compared to more commonly used high-priced sensors. For the estimation of upper-body joint positions, semantic segmentation with a fully convolutional network was exploited. From acquired RGB images, joint heatmaps accurately estimate the coordinates of the location of each joint. The network architecture was designed to learn and detect the locations of joints via the sequential prediction processing method. Our proposed method was tested and validated for efficient estimation of the human upper-body pose. The obtained results reveal the potential of a simple RGB camera sensor for human pose estimation applications.

Improved Multi-Cell Joint Channel Estimation for the TD-SCDMA Downlink

XUE, Peng,CAO, Ning,KANG, Dong Kwan,KIM, Duk Kyung The Institute of Electronics, Information and Comm 2009 IEICE TRANSACTIONS ON COMMUNICATIONS - Vol.92 No.4

<P>In this paper, a multi-cell joint channel estimation (JCE) method is proposed for the TD-SCDMA downlink. In the proposed multi-cell JCE approach, the received midambles from adjacent cells are jointly processed, rather than being treated as interference as in single cell channel estimation. By jointly processing all the received midambles, the user can simultaneously estimate the channel impulse responses (CIRs) for both its home cell and adjacent cells. If the received signal from adjacent cells has a delay, multi-cell JCE is still operable with slight adjustment in the midamble matrix, and the performance loss is also minor. The performance of multi-cell JCE is analyzed and evaluated by simulations. The results demonstrate that the proposed multi-cell JCE method can significantly improve the channel estimation accuracy. When the signal from each cell has similar power level, the mean square error (MSE) of the estimated CIRs for all cells is lower than 0.01. With more accurate CIRs from multi-cell JCE, multi-cell JD also yields better performance compared with the single cell channel estimation methods.</P>

Dynamic Property Identification of Structures with Hinge Joint using Maximum Likelihood Estimation

원준호,Chekyu Lim(임체규),Dooho Lee(이두호),Jooho Choi(최주호) 대한기계학회 2011 대한기계학회 춘추학술대회 Vol.2011 No.10

Identification of dynamic properties for structural joints is important in the prediction of dynamic behavior of assembled systems. However, the real values of these properties have large uncertainty and identification of the properties using analytical or experimental approach is extremely difficult or sometimes impossible. Several studies proposed hybrid or synthesis methods which simultaneously use analytical and experimental approaches to identify the dynamic properties of joint. However, among the many types of joints, the bolt joint was only treated as a practical example in most of the studies. In this study, for verification model composed of simple plate with a hinge joint, dynamic properties are identified by using maximum likelihood estimation (MLE). Finally, the proposed method is applied to a real glove box that includes the hinge joints in a vehicle.

A Kalman Filter for Inverse Dynamics of IMU-Based Real-Time Joint Torque Estimation

최지석(Ji Seok Choi),이창준(Chang June Lee),이정근(Jung Keun Lee) Korean Society for Precision Engineering 2022 한국정밀공학회지 Vol.39 No.1

One of the problems in inverse dynamics calculation for the inertial measurement unit (IMU)-based joint force and torque estimation is the amplified signal noises of segment kinematic data mainly due to the differentiation procedure and segmental soft tissue artifacts. In order to deal with this problem, appropriate filtering methods are often recommended for signal enhancement. Conventionally, a low-pass filter (LPF) is widely used for the kinematic data. However, the zero-phase LPF requires post-processing, while the real-time LPF causes an unignorable time lag. For this reason, it is inappropriate to use the LPF for real-time joint torque estimation. This paper proposes a Kalman filter (KF) for inverse dynamics of IMUbased joint torque estimation in real time without any time lag, while utilizing the smoothing capability of the KF. Experimental results showed that the proposed KF outperformed a real-time LPF in the estimation accuracy of hip joint force and torque during jogging on the spot by 100 and 29%, respectively. Although the proposed KF requires the process of adjusting covariance according to the dynamic conditions, it can be expected to improve the estimation performance in the field where joint force and torque need to be estimated in real time.

Kalman Randomized Joint UKF Algorithm for Dual Estimation of States and Parameters in a Nonlinear System

Safarinejadian, Behrouz,Vafamand, Navid The Korean Institute of Electrical Engineers 2015 Journal of Electrical Engineering & Technology Vol.10 No.3

This article presents a new nonlinear joint (state and parameter) estimation algorithm based on fusion of Kalman filter and randomized unscented Kalman filter (UKF), called Kalman randomized joint UKF (KR-JUKF). It is assumed that the measurement equation is linear. The KRJUKF is suitable for time varying and severe nonlinear dynamics and does not have any systematic error. Finally, joint-EKF, dual-EKF, joint-UKF and KR-JUKF are applied to a CSTR with cooling jacket, in which production of propylene glycol happens and performance of KR-JUKF is evaluated.