Automated classification and quantification of spatio-temporal parameters in cross-country skiing skating technique by analysis of inertial sensor data and sensor insole data

Keywords Cross-country skiing, skating technique

Abstract The aim of this thesis was (1) to develop a method for dividing cross-country skiingcycles into phases, and (2) extract and select features characterizing the change in center of pressure (CoP) during these phases. Both tasks were solved by the help of Moticon OpenGo insole data (pressure sensors + accelerometer) from athletes on different skill levels. The accelerometer data produced extremal values which could be detected by an algorithm. The different phases were identified through combining search areas (based on total foot pressure) with the extremal points from accelerometer data. Selected features were compared between subjects in order to reveal whether or not they could differentiate between subjects on different skill levels. Analysis of the foot movement in cross-country skiing could facilitate a more accurate understanding of the athletes’ techniques, which might facilitate a better performance.
This study showed that it was possible to divide cycles into phases by the use of Moticon’s OpenGo insole data. The calculated phase-features indicated that features of athletes on the same skill level had more in common than features of athletes on different skill levels. One of the main results was that elite athletes spent a larger percentage of their cycle time on the gliding phase, than athletes on lower skill levels. Another result was that features describing the change of CoP during the gliding phase could classify five out of six athletes correctly according to skill level.


This website uses cookies to improve your experience. If you continue using our website, we assume your consent with this. Further information is provided in our data privacy statement.