Recently, the estimation of joint kinetics such as joint force and moment using wearable inertial sensors has received great attention in biomechanics. Generally, the joint force and moment are calculated though inverse dynamics using segment kinematic data, ground reaction force, and moment. However, this approach has problems such as estimation error of kinematic data and soft tissue artifacts, which can lead to inaccuracy of joint forces and moments in inverse dynamics. This study aimed to apply a recurrent neural network (RNN) instead of inverse dynamics to joint force and moment estimation. The proposed RNN could receive signals from inertial sensors and force plate as input vector and output lower extremity joints forces and moments. As the proposed method does not depend on inverse dynamics, it is independent of the inaccuracy problem of the conventional method. Experimental results showed that the estimation performance of hip joint moment of the proposed RNN was improved by 66.4% compared to that of the inverse dynamics-based method.

In this study, the Inertial Measurement Unit (IMU) signals and clinical evaluation scales for Parkinson"s disease were correlated. The study included 16 patients diagnosed with Parkinson"s disease. Each subject was evaluated based on Korean Mini-Mental State Examination (KMMSE), Unified Parkinson"s Disease Rating Scale (UPDRS) part 3, New Freezing of Gait Questionnaire (NFOGQ) parts 2 & 3, and Hoehn & Yahr Scale (H&Y). All subjects performed the Time Up and Go test by attaching IMU sensors to both ankles and torso. Based on the tilting angle of torso and the time of first step, the freezing and non-freezing windows were determined. Seven IMU features involving the ankle signals were calculated in the specific window. Spearman’s correlation analysis of clinical evaluation scales was performed. As a result, the freezing index and power of locomotion band (0.3-3 Hz) were recommended to determine UPDRS part 3. Also, the intensity of the locomotion band facilitated evaluation of NFOGQ part 3 regardless of freezing of gait.

This paper proposes a practical method, for evaluating positioning of outdoor mobile robots using Unscented Kalman Filter (UKF). Since the UKF method does not require the linearization process unlike EKF localization, it can minimize effects of errors caused by linearization of non-linear models for position estimation. This method enables relatively high performance position estimation, using only non-inertial sensors such as low-precision GPS and a digital compass. Effectiveness of the UKF localization method was verified through actual experiments and performance of position estimation was compared with that of the existing EKF method. Experimental results revealed the proposed method has better performance than the EKF method, and it is stable regardless of initial error size, and observation period.