University of Twente Student Theses


Determining the projected frontal area of cyclists using motion capturing data

Kruistum, G.M. van (2020) Determining the projected frontal area of cyclists using motion capturing data.

[img] PDF
Abstract:Time differences in Grand Tours tend to be a matter of just a few minutes, of which a large portion of can be gained (or lost) in time trial stages. Aerodynamics play a large role, especially in these stages, and decreasing aerodynamic drag seems to be the key to winning valuable seconds. One of the main factors influencing aerodynamic drag in cycling is the projected frontal area. In the current situation, this research is done in lab conditions, i.e. in a wind tunnel or a velodrome. This is not ideal: wind tunnel testing is expensive, riding positions are not tested during longer rides, while velodrome testing is tedious and does not deliver real-time results, all among other disadvantages. These problems can be overcome by a new testing method, that yields real-time results in a real-life setting. From several experiments, it follows that a support vector regression model, as developed in this thesis, is a viable proof of concept for predicting the frontal area of a cyclist from motion data. Experiments also show that such a model still works well when a reduced sensor set is used as input.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science, 76 recreation, leisure
Programme:Business Information Technology MSc (60025)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page