#machine learning
Jun 2026
A Geometry-Aware Triplane Field Network for Vehicle Aerodynamic Prediction
Alation results show that AFNO mixing, local CNN refinement, and query-side geometric encoding each contribute to accuracy, supporting the proposed mechanism of combining structured triplane representation with explicit aerodynamic geometry cues.
Kang-Kang Qi, Hui-Yu Yang, Ke-Qi Ding et al.
· Theoretical and Applied Mech... · 1 citation