Siân Slawson

Design and Implementation of an Integrated Performance Monitoring Tool for Swimming to Extract Stroke Information at Real Time

Current methods used to monitor performance for swimming do not offer real-time feedback to coaches and significantly increase post session analysis times. The aim of the research outlined in this paper is to design and implement a novel performance analysis tool to measure swimmer performance using wireless technology previously developed at Loughborough University. The Computer Integrated Manufacturing Open System Architecture (CIMOSA) has been coupled with a software architecture based on object-oriented techniques to formalize and structure the development of a computer integrated real-time monitoring system. Filtering and signal-processing algorithms are applied to extract performance indicators in real time, hence allowing faster access of feedback that can be used to enhance the swimming performance during each training session.

Accelerometer Profile Recognition of Swimming Strokes (P17)

The use of technology in sports performance analysis is a rapidly increasing practise. Tools for analysis aim to provide useful information to supplement coach knowledge and improve feedback in the development of athletes. In swimming the use of subjective video analysis is wide-spread, however, unlike some other sports, there are few quantitative measures of performance. Quantitative measures, such as intra cyclic variations of stroke characteristics, have the potential to provide more specific performance metrics from which to make improvements. Such measures are currently not widely available to coaches, support staff and swimmers, due to the infancy or lack of sufficiently developed technologies.

A Multi-sensor System for Monitoring the Performance of Elite Swimmers

A comprehensive system is required to monitor numerous variables of a swimmer’s performance. Current methods of analysis do not offer solutions which record and analyse multiple performance parameters simultaneously. The research presented in this paper provides an overview of an integrated system which has been developed to monitor several components of a swimmer’s start, free swimming and turn concurrently. The integrated system is comprised of a wearable wireless sensor, vision components, force platform, pressure pad, LED markers and audio communication.

Characterizing the swimming tumble turn using acceleration data

Work detailed in this paper describes the development of a swimmer-worn, wireless sensor node capable of capturing real time acceleration data during all phases of swimming, i.e. the start, free swimming, and turns. Analysis of results focuses on the interpretation of these data for the tumble turn. Phases of the turn are characterized in the data enabling the contribution of each phase – approach, rotation, and glide – to be evaluated. A single subject, a triathlete, 400 m swim trial was used to capture acceleration data of multiple turns for analysis. Statistical process control (SPC) charts were used to present results and give an indication of whether correlation occurred between each of the turn phases and total turn time. Informal feedback from swimming professionals suggested that these presentation methods supply results in an easy-to-interpret manner, whereby outliers and variations in performance could be visually identified.

A Novel Instrumented Cycle Ergometer with Automated In-Situ Capabilities

The objective of the research outlined in this paper is to develop a novel cycle ergo meter that incorporates (1) a faster, more accurate automatic bike set-up and adjustment for individual cyclists with greater repeatability and (2) the personalized torque, angular displacement and angular velocity profiles generated via mathematical models of cycling performance from experimental road trial data using SRM or Beru cranks. This ergo meter allows fitting adjustments whilst cycling and aims to reduce the set-up time to about 80 seconds for the first set-up and 17 seconds for the subsequent set-ups as opposed to 30 minutes (required by the end-users current ergo meter). Instrumented cranks have been fitted in to monitor the torque generated by the crank movements and the perpendicular force. Eccentric cycling for muscle strengthening can be readily incorporated into the torque producing capability of the system.

Image processing algorithms to extract swimming tumble turn signatures in real-time

Vision based methods provide a powerful technique for accessing swimmer performance. Currently, manual analysis of recorded images is used to determine performance-relevant metrics, which incurs significant time and effort for the analyst. Real-time automated image processing algorithms to measure a swimmers performance for tumble turns is detailed in this paper. The algorithms yield performance parameters including contact time, foot orientation, distance between feet, lane position and depth of contact with the pool wall. Pressure sensors have been employed to measure pressure exerted on the wall and wearable LED markers are used to track movement of the swimmer during the turn. The system significantly improves the efficiency of the analysis process.