The purpose of this study is to provide a method for classifying non-fatigued vs. fatigued states following manual material handling. A method of template matching pattern recognition for feature extraction ($1 Recognizer) along with the support vector machine model for classification were applied on the kinematics of gait cycles segmented by our stepwise search-based segmentation algorithm. A single inertial measurement unit on the ankle was used, providing a minimally intrusive and inexpensive tool for monitoring. The classifier distinguished between states using distance-based scores from the recogniser and the step duration. The results of fatigue detection showed an accuracy of 90% across data from 20 recruited subjects. This method utilises the minimum amount of data and features from only one low-cost sensor to reliably classify the state of fatigue induced by a realistic manufacturing task using a simple machine learning algorithm that can be extended to real-time fatigue monitoring as a future technology to be employed in the manufacturing facilities. Practitioner Summary: We examined the use of a wearable sensor for the detection of fatigue-related changes in gait based on a simulated manual material handling task. Classification based on foot acceleration and position trajectories resulted in 90% accuracy. This method provides a practical framework for predicting realistic levels of fatigue.

The goal of this case study is to answer four research questions related to fatigue through features derived from wearable sensors to measure patterns in steps: (1) How do important gait parameters change over time? (2) How do these sensor-based changes relate to the participant's subjective fatigue ratings over time? (3) Are there consistent patterns in performance across different individuals over time? and (4) Do these patterns vary systematically based on specific demographic characteristics? To answer these questions, we have combined multivariate changepoint methods with hierarchical time-series clustering and exploratory data analysis. The results improve our understanding of fatigue development.

The purpose of this study is to provide a new method of estimating the hip acceleration and trunk posture in the sagittal plane during a walking task using an extended Kalman filter (EKF) and an unscented Kalman filter (UKF). A comparison between these two estimation techniques is also provided. Considering the periodic nature of gait, a modified biomechanical model with Fourier series approximations are utilized as a priori knowledge. Inertial measurement units (IMUs) are placed on the right side of the ankle, hip, and middle of the trunk of twenty recruited participants, as input, a posteriori data, and the ground truth for the model, separately. The results show a better performance of the EKF in estimating the hip acceleration (6.5% error) as well as the trunk posture (3.12% error). Moreover, both the EKF and the UKF provide lower error rates for the trunk posture in comparison to the hip acceleration. This study provides an inexpensive and novel method to estimate and filter the kinematics of motion for different body locations from a single accurate IMU attached to the ankle considering the periodic nature of gait that can be extended to other activities as well as real-time applications.