Understanding Human Perception of Vibrotactile Feedback in Walking and Running Tasks

Abstract

Vibrotactile feedback is increasingly becoming an essential feedback component in several non-medical and medical areas. One area that vibrotactile feedback has not been explored as an intervention tool is in sports science. In addition, vibrotactile feedback lacks scientific evidence as a feedback mechanism within the sports world. A portable vibrotactile feedback system was developed to understand how a human perceives vibrotactile feedback while performing walking and running tasks. The system incorporates a model-based simulation framework to check human motion and visualize key performance metrics. We performed a pilot study on twenty-six subjects to understand the accuracy of human perception of vibrotactile feedback using the system we developed. Here, we investigated the type, number, and location of haptic feedbacks that would yield better subject perceived accuracy during walking and running. Findings suggested that staggered vibrotactile feedback would lead to higher accuracy than continuous vibrotactile feedback especially when more than two motors were used. For the subjects perception accuracy for all motor combinations, we observed an average that was 9.3% higher for walking compared to running. For one motor, the perception accuracy decreased by only 3.38% while running compared to walking. The decrease was much more significant for higher combinations of motors. The decrease in the perception accuracy of the motor locations was significantly less for both staggered (p < 0.001) and one motor (p < 0.001) vibration modes.

Forearm muscle activity estimation based on anatomical structure of muscles

“…Estimation of the activity in each muscle is a useful alternative approach to direct measurement. To date, a number of estimation methods have been proposed (OpenSim; Demircan, Recinos, Khoo, Teng, & Wu, 2020; Hirasawa, Ayusawa, & Nakamura, 2013; Nakajima, Keeratihattayakorn, Yoshinari, & Tadano, 2014; Piovanelli et al, 2020; Van den Bogert, Geijtenbeek, Even‐Zohar, Steenbrink, & Hardin, 2013). Estimation based on motion capture data is one of the most common approaches.…”

“…The important feature of this method is its ability to estimate the muscle activity of dynamic motion. Various types of dynamic motion, including the behaviors of top athletes, were analyzed and used for advanced training (OpenSim; Demircan et al, 2020; Hirasawa et al, 2013; Van den Bogert et al, 2013). One of the problems with this approach is the long computation time needed for inverse dynamics.…”