In the master thesis we present a newly developed system for assessing the physical abilities of athletes.
The aim of this thesis is to design, implement and technically evaluate a system that evaluates the athlete's readiness before competition or training progress in terms of both the result (time) and the quality of performance by measuring time and movement in given tests.
The added value of our system is in more accurate and precise time measurements compared to manual measurements and in the acquisition of many additional spatiotemporal and kinematic parameters of the athlete during the test.
In the theoretical part of the work we focus on the field of sports: we introduce the meaning of the word agility, the current evaluation methods and some standardized tests.
We also introduce well-known devices for measuring and recording the other kinematic parameters in the assessment of physical abilities.
After an overview of the field, we introduce the architecture of the system, describing the individual elements (server, measuring device, kinematic sensor, optical door).
Their functionalities and operation are presented in the chapter System Operation.
We studied delays and measurement errors of the system and concluded that the accuracy of a single measurement device meets the pre-set demands, since its measurement error is less than 2ms.
The accuracy of the overall system is the sum of the error of a single device (< 2 ms), the error of the clock of a single device (1.5 us / 10 ms) and system synchronisation timing errors (15 ms +- 10 ms).
In cooperation with the Faculty of Sports of the University of Ljubljana, we participated in the measurements for agility assessment of the Slovenian national men's volleyball cadet team.
%The results of the measurements are presented from the{\tiny } acquisition and basic processing of time-space parameters and kinematic signals point of view.
The results of these measurements are presented in the view of the acquisition and basic processing of spatiotemporal parameters and kinematic signals.
Further interpretation will be done later in close cooperation with experts in the field of sport and physical education.
We can conclude that the accuracy of the proposed system meets the demands.
Nevertheless, we can propose some improvements.
Our system could be improved by using a separate radio channel for better synchronization.
It is also possible to improve the evaluation with models based on machine learning algorithms.
These could also take into account the acquired motion signals.
|
