The popularity of wearable technology in sport has increased, due to its ability to provide unobtrusive monitoring of athletes. This technology has been used to objectively measure kinetic and kinematic variables, with the aim of preventing injury, maximising athletic performance and classifying the skill level of athletes, all of which can influence training and coaching practices. Wearable technologies overcome the limitations of motion capture systems which are limited in their capture volume, enabling the collection of data in-field, during training and competition. Inertial sensors are a common form of technology used in these environments however, their high-cost and complex calibration due to multiple sensor integration can make them prohibitive for widespread use. This thesis focuses on the development of a strain sensor that can be used to measure knee range of motion in sports, specifically rowing and cycling, as a potential low-cost, lightweight alternative to inertial sensors which can also be integrated into clothing, making them more discreet. A systematic review highlighted the lack of alternate technologies to inertial sensors such as strain sensors, as well as the limited use of wearable technologies in both rowing and cycling. Strain sensors were fabricated from a carbon nanotube-natural rubber composite using solvent exchange techniques and employed a piezoresistive sensing mechanism. These were then characterised using mechanical testing, to determine their electrical properties under cyclical strain. The strain sensors displayed hysteretic behaviour, but were durable, withstanding over 4500 strain cycles. Statistical analysis indicated that over 60% of the tests conducted had good intra-test variability with regards to the resistance response range in each strain cycle and sensor response deviating by less than 10% at strain rates below 100 mm/min and less than 20% at a strain rate of 350 mm/min. These sensors were integrated into a wearable sensor system and tested on rowing and cycling cohorts consisting of ten athletes each, to assess the translational use of the strain sensor. This preliminary testing indicated that strain sensors were able to track the motion of the knee during the rowing stroke and cycling pedalling motion, when compared to the output of a motion capture system. Perspectives of participants on the wearable system were collected, which indicated their desire for a system that they could use in their sport, and they considered the translation of this system for real-life use with further development to improve comfort of the system and consistency of the sensor response. The strain sensors developed in this project, when integrated into a wearable sensor system, have the potential to provide an unobtrusive method of measuring knee kinematics, helping athletes, coaches and other support staff make technical changes that can reduce injury risk and improve performance.
© Copyright 2023 Veröffentlicht von Imperial College London. Alle Rechte vorbehalten.