In this paper, we present an approach to the online implementation of a gait event detector based on machine learning algorithms. Gait events were detected using a uniaxial gyro that measured the foot instep angular velocity in the sagittal plane to feed a four-state left-right hidden Markov model (HMM). The short-time Viterbi algorithm was used to overcome the limitation of the standard Viterbi algorithm, which does not allow the online decoding of hidden state sequences. Supervised learning of the HMM structure and validation with the leave-one-subject-out validation method were performed using treadmill gait reference data from an optical motion capture system. The four gait events were foot strike, flat foot (FF), heel off (HO), and toe off. The accuracy ranged, on average, from 45 ms (early detection, FF) to 35 ms (late detection, HO); the latency of detection was less than 100 ms for all gait events but the HO, where the probability that it was greater than 100 ms was 25%. Overground walking tests of the HMM-based gait event detector were also successfully performed.
In this paper we develop a system for which applications in the field of personal navigation are planned. In the current version, the system embodies a global positioning system (GPS) receiver and an inertial measurement unit (IMU), composed of two dual-axis accelerometers and one single-axis gyro. The IMU is positioned at a subject's foot instep, and it is intended to produce estimates of some gait parameters, including stride length, stride time, and walking speed. Data from GPS and IMU are managed by a DSP-based control box. The computations performed by the DSP processor allow to detect subsequent foot contacts by a threshold-based method applied to gyro signal, and to reconstruct the trajectory of the foot instep by numerical strapdown integration. Features of human walking dynamics are incorporated in the algorithm to enhance the estimation accuracy against errors due to sensor noise and integration drift. All computations are performed by the DSP processor in real-time conditions. The foot sensor performance is assessed during outdoor level walking trials. The traveled distance estimated by inertial deadreckoning is compared with the estimate produced by GPS in experimental conditions where GPS can be used as a reference source for accurate absolute positioning. Results show the remarkable accuracy achieved by foot inertial sensing.
A sensor fusion method was developed for vertical channel stabilization by fusing inertial measurements from an Inertial Measurement Unit (IMU) and pressure altitude measurements from a barometric altimeter integrated in the same device (baro-IMU). An Extended Kalman Filter (EKF) estimated the quaternion from the sensor frame to the navigation frame; the sensed specific force was rotated into the navigation frame and compensated for gravity, yielding the vertical linear acceleration; finally, a complementary filter driven by the vertical linear acceleration and the measured pressure altitude produced estimates of height and vertical velocity. A method was also developed to condition the measured pressure altitude using a whitening filter, which helped to remove the short-term correlation due to environment-dependent pressure changes from raw pressure altitude. The sensor fusion method was implemented to work on-line using data from a wireless baro-IMU and tested for the capability of tracking low-frequency small-amplitude vertical human-like motions that can be critical for stand-alone inertial sensor measurements. Validation tests were performed in different experimental conditions, namely no motion, free-fall motion, forced circular motion and squatting. Accurate on-line tracking of height and vertical velocity was achieved, giving confidence to the use of the sensor fusion method for tracking typical vertical human motions: velocity Root Mean Square Error (RMSE) was in the range 0.04–0.24 m/s; height RMSE was in the range 5–68 cm, with statistically significant performance gains when the whitening filter was used by the sensor fusion method to track relatively high-frequency vertical motions.
This paper investigates a fall detection system based on the integration of an inertial measurement unit with a barometric altimeter (BIMU). The vertical motion of the body part the BIMU was attached to was monitored on-line using a method that delivered drift-free estimates of the vertical velocity and estimates of the height change from the floor. The experimental study included activities of daily living of seven types and falls of five types, simulated by a cohort of 25 young healthy adults. The downward vertical velocity was thresholded at 1.38 m/s, yielding 80% sensitivity (SE), 100% specificity (SP) and a mean prior-to-impact time of 157 ms (range 40-300 ms). The soft falls, i.e., those with downward vertical velocity above 0.55 m/s and below 1.38 m/s were analyzed post-impact. Six fall detection methods, tuned to achieve 100% SE, were considered to include features of impact, change of posture and height, singularly or in association with one another. No single feature allowed for 100% SP. The detection accuracy marginally improved when the height change was considered in association with either the impact or the change of posture; the post-impact fall detection method that analyzed the impact and the change of posture together achieved 100% SP.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.