Machine learning surrogate model for athlete aerodynamics with application to cycling

Abstract Aerodynamic drag dominates the resistive forces in many sports at racing speeds, and small athlete posture changes can produce practically meaningful changes in performance. Standard approaches like field testing, wind-tunnel testing and computational fluid dynamics (CFD) can provide accurate results, but are time-consuming. Recent advances in scientific machine learning have enabled surrogate models that predict flow quantities at a fraction of the computational cost of a full CFD simulation. However, most applications currently focus on industrial geometries, such as cars or aircraft, where the surface is morphed to generate large datasets for training. It remains unclear how well these methods transfer to athlete geometries where variability is dominated by articulated pose changes rather than smooth shape morphing. In this work, a dataset for cyclist aerodynamics was generated by combining 12 scanned athlete geometries with 20 postures per athlete. CFD simulations were performed for these 240 geometries, which were then used to train a state-of-the-art surrogate model. The generalization of the model to an unseen geometry was investigated, along with the balance between number of positions and number of unique geometries in the dataset. The surrogate model predicts drag area with a mean absolute percentage error of < 3%, with all predictions falling within ± 10 % of the CFD reference values. The results also indicate that increasing the number of unique athlete geometries improves performance more effectively than increasing the number of postures per athlete at fixed dataset size. These results support the feasibility of surrogate models for flow over athletes, enabling posture screening and comparative assessment at interactive speeds. This may open up new workflows and new application areas of drag calculations in the domain of sports aerodynamics.