Técnicas basadas en sensores inerciales para el seguimiento de la actividad física : validación de las técnicas basadas en sensores inerciales para el seguimiento de la actividad física.

Abstract

Currently there are several methods to capture the movement of physical activities, which are expensive and difficult to move. For this reason, the need arises to look for alternatives that offer the same results, but that are not expensive and impractical.This is how this curricular integration project aims to determine the reliability of the use of IMUs inertial sensors for monitoring physical activity. Since the benefits they provide compared to other methods are high. To develop the methodology, experimentation is carried out on a biomechanical model of the elbow using the photogrammetry system and IMUs sensors. Where the importance of calibration is detailed, the data treatment with the creation of scripts in Matlab to later validate both techniques. Additionally, the kinematics obtained with the photogrammetry and OpenSim records are compared, through a simulation. In addition, scripts were created in Matlab detailing the transformation of the results of Euler's experimentation to quaternions since it is the acceptable format in OpenSim. When performing the validation of the techniques, results of the angular positions in the main plane of the arm are obtained with a CMC (Comparative Measurement Close-range) of 0.91 and of the forearm with a CMC of 0.96 and whose curves indicate the same tendency. This confirms the reliability of using IMUs sensors as a motion capture technique where simple models are analyzed.