Ambulatory Assessment of Instantaneous Velocity during Walking Using Inertial Sensor Measurements

Abstract

A novel approach for estimating the instantaneous velocity of the pelvis during walking was developed based on Inertial Measurement Units (IMUs). The instantaneous velocity was modeled by the sum of a cyclical component, decomposed in the Medio-Lateral (ML), VerTical (VT) and Antero-Posterior (AP) directions, and the Average Progression Velocity (APV) over each gait cycle. The proposed method required the availability of two IMUs, attached to the pelvis and one shank. Gait cycles were identified from the shank angular velocity; for each cycle, the Fourier series coefficients of the pelvis and shank acceleration signals were computed. The cyclical component was estimated by Fourier-based time-integration of the pelvis acceleration. A Bayesian Linear Regression (BLR) with Automatic Relevance Determination (ARD) predicted the APV from the stride time, the stance duration, and the Fourier series coefficients of the shank acceleration. Healthy subjects performed tasks of Treadmill Walking (TW) and Overground Walking (OW), and an optical motion capture system (OMCS) was used as reference for algorithm performance assessment. The widths of the limits of agreements (±1.96 standard deviation) were computed between the proposed method and the reference OMCS, yielding, for the cyclical component in the different directions: ML: ±0.07 m/s (±0.10 m/s); VT: ±0.03 m/s (±0.05 m/s); AP: ±0.06 m/s (±0.10 m/s), in TW (OW) conditions. The ARD-BLR achieved an APV root mean square error of 0.06 m/s (0.07 m/s) in the same conditions.

A Method to Estimate Horse Speed per Stride from One IMU with a Machine Learning Method

“…In human research, a wide literature exists on computing human walking speed from data collected by one IMU placed on the foot, as for example [29], who compared two methods of walking speed estimation, whose ARMSE range was 0.2-0.3 km/h (0.06-0.08 m/s) depending on the walking speed and the method used. The work in [13] estimated an instantaneous velocity decomposed in the three space directions from two IMUs' data placed on the pelvis and on the shank of the subject, whose accuracy was in the same range as the previous study. The work in [31] developed a model for walking speed estimation based on a regression model, which used data from one wrist worn inertial sensor.…”

“…Thirdly, new methods based on statistical approaches are developed to estimate human speed [13,14] from IMU data. Those approaches provide accurate estimation of walking speed, but the regression models' accuracy seems to be dependent on the range of motion.…”

An Open-Source 7-DOF Wireless Human Arm Motion-Tracking System for Use in Robotics Research

“…Furthermore, the system was also tested with motions that ensure adequate arm-tracking when executing flexion/extension, abduction/adduction and full elbow flexion movements. The accuracy evaluation was done using the root mean squared error (RMSE) metric commonly applied for comparative analysis with ground truth references [2,10,32]. During the tests, a total of 60 (x, y, z) elbow and wrist position samples, estimated using both inertial systems, were collected.…”

“…Four types of human body motion-tracking systems are currently available [4]: vision-based systems with markers such as CODA (www.codamotion.com), markerless vision-based systems (see, e.g., [5]), robot-guided tracking systems (see, e.g., [6]), and non-vision-based inertial systems (see, e.g., [3,7,8]). The systems in the first three categories are not completely wearable or portable, and they normally require the presence of grounded devices, ambient light conditions, and/or structured environments [9,10].…”

A Multi-Sensor Matched Filter Approach to Robust Segmentation of Assisted Gait

“…Wearable sensor systems, enabled by smartphones, foot switches, pressure insoles, accelerometers, and gyroscopes present a more portable, low cost, and easy to use alternative to traditional motion capture systems. Consequently, wearable sensor systems have gained substantial interest in the research community for monitoring gait activity [ 16 , 17 , 26 , 27 , 28 , 29 , 30 , 31 ]. Given their ubiquity in today’s society, the accelerometers within mobile phones have been widely explored for recognition of gait and walking terrain [ 32 , 33 , 34 ].…”