A novel monitoring system for the training of elite swimmers

Swimming performance is primarily judged on the overall time taken for a swimmer to complete a specified distance performing a stroke that complies with current regulations defined by the Fédération Internationale de Natation (FINA), the International governing body of swimming. There are three contributing factors to this overall time; the start, free swimming and turns. The contribution of each of these factors is event dependent; for example, in a 50m event there are no turns, however, the start can be a significant contributor. To improve overall performance each of these components should be optimised in terms of skill and execution. This thesis details the research undertaken towards improving performance-related feedback in swimming. The research included collaboration with British Swimming, the national governing body for swimming in the U.K., to drive the requirements and direction of research. An evaluation of current methods of swimming analysis identified a capability gap in real-time, quantitative feedback. A number of components were developed to produce an integrated system for comprehensive swim performance analysis in all phases of the swim, i.e. starts, free swimming and turns. These components were developed to satisfy two types of stakeholder requirements. Firstly, the measurement requirements, i.e. what does the end user want to measure? Secondly, the process requirements, i.e. how would these measurements be achieved? The components developed in this research worked towards new technologies to facilitate a wider range of measurement parameters using automated methods as well as the application of technologies to facilitate the automation of current techniques. The development of the system is presented in detail and the application of these technologies is presented in case studies for starts, free swimming and turns. It was found that developed components were able to provide useful data indicating levels of performance in all aspects of swimming, i.e. starts, free swimming and turns. For the starts, an integrated solution of vision, force plate technology and a wireless iii node enabled greater insight into overall performance and quantitative measurements of performance to be captured. Force profiles could easily identify differences in swimmer ability or changes in technique. The analysis of free swimming was predominantly supported by the wireless sensor technology, whereby signal analysis was capable of automatically determining factors such as lap times variations within strokes. The turning phase was also characterised in acceleration space, allowing the phases of the turn to be individually assessed and their contribution to total turn time established. Each of the component technologies were not used in isolation but were supported by other synchronous data capture. In all cases a vision component was used to increase understanding of data outputs and provide a medium that coaches and athletes were comfortable with interpreting. The integrated, component based system has been developed and tested to prove its ability to produce useful, quantitative feedback information for swimmers. The individual components were found to be capable of providing greater insight into swimming performance, that has not been previously possible using the current state of the art techniques. Future work should look towards the fine-tuning of the prototype system into a useable solution for end users. This relies on the refinement of components and the development of an appropriate user interface to enable ease of data collection, analysis, presentation and interpretation.