Real-Time Estimation of Muscular Efficiency in Cyclists Using Garmin Devices
Résumé
Muscular efficiency—the ratio of mechanical power output to metabolic energy expenditure—is a key determinant of cycling performance. While traditionally assessed in laboratory settings, such methods are ill-suited for real-world conditions. This study introduces a field-deployable metric for real-time estimation of muscular efficiency using consumer-grade Garmin devices equipped with heart rate (HR) monitors and power meters. Efficiency is computed as the ratio of mechanical energy (power × time) to HR-derived metabolic energy, adjusted for individual characteristics (age, mass, sex) using validated predictive models (Keytel et al., 2005). Calculations are constrained to moderate intensities (between ventilatory threshold 1 and lactate threshold), where HR and power exhibit a linear relationship. Implemented as a Connect IQ data field, the metric was validated using Strava data from professional cyclists and tested over tens of thousands of kilometers in real-world conditions. Results show that intra-session efficiency trends reliably reflect physiological states—declining efficiency signals fatigue, while rising efficiency indicates optimal pacing. Although absolute values are influenced by environmental factors (e.g., temperature, altitude), relative trends remain actionable for in-ride decision-making. Despite limitations in the Garmin SDK (e.g., poor documentation, lack of formal verification tools), the implementation proved robust. The approach offers a practical, integrated alternative to existing tools that merely visualize HR and power without quantifying efficiency.
Domaines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |