Model based Automated Cycling Ergometer

Abstract

Listen

Translate article

Laboratory testing of cyclists is currently undertaken using turbo trainers or cycle ergometers. The benefits of laboratory testing are the ability to measure performance: (i) more accurately and repeatedly and (ii) under controlled conditions enabling, for example, video analysis to determine joint-specific power production or enable novel instrumentation to be applied to the bicycle, for example, to measure seat interface pressure. Influence of the bicycle fit on torque production have been presented by Irriberri et al(2008) and Peveler et al(2007). Market leading bicycle ergometer manufacturers, such as Lode and Monark, provide feedback on performance metrics including cadence and force measurement. However, neither ergometer provides real time adaptation of bicycle fit to the resolution (i.e. < mm precision) required by elite athletes or allows adjustments to position whilst cycling under simulated road / track conditions. The objective of the research presented in this paper is to demonstrate and provide initial validation results for a novel, fully automated cycle ergometer that incorporates faster, repeatable and more accurate adjustments to bicycle geometry. The ergometer also allows the cyclist to use their preferred handlebars and saddle to accommodate the different cycling disciplines, e.g. track, road, mountain and BMX. The ergometer enables fitting adjustments to be controlled whilst cycling and aims to reduce initial set-up times for different athletes to about 30 seconds as opposed to 30 minutes (required by the end-users current ergometer instantiations). Instrumented cranks have been fitted to monitor the torque and force generated by the crank movements in 2-axes through 360 degrees of crank motion. The ergometer can be coupled (via a user selectable clutch mechanism) to an AC servo motor within the drive chain which supports the application of models of bicycle performance to the ergometer to enable torque versus position versus speed profiles as derived from road and / or track trials to be readily mapped into the laboratory environment