Numerical and experimental investigations of the aerodynamic drag characteristics and reduction of an autonomous vehicle

Externally mounted sensors on autonomous vehicles can introduce interference effects, altering the aerodynamic characteristics compared with those of conventional vehicles and significantly increasing aerodynamic drag. In this study, the structure of the external flow field around an autonomous vehicle and the constitutive law of its aerodynamic drag were revealed through computational fluid dynamics simulations. A wind tunnel experiment was conducted to validate the reliability of the numerical method. To minimize the aerodynamic drag of the autonomous vehicle, the shapes of the externally mounted sensors were parameterized. The design of experiments, the Kriging surrogate model, and the optimization algorithm were then used to optimize these shapes. Finally, the improvements in the steady and unsteady aerodynamic performance of the autonomous vehicle were analyzed, and the mechanisms and effects of the drag reduction were explored in detail. The results indicate that, compared with the baseline model, the optimized model significantly reduces the proportion of interference drag within the total aerodynamic drag, along with a marked decrease in the interference effects of the externally mounted sensors. Furthermore, compared with the baseline model, the optimized model reduces the aerodynamic drag coefficient by 5.99% under steady-state simulation and significantly improves aerodynamic performance under unsteady simulation. These findings demonstrate the high effectiveness of the optimization design for autonomous vehicles.