This paper was produced for the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

Autonomous vehicles are becoming a reality and many vehicle manufacturers are currently investing in this technology. These autonomous vehicles perceive the world through sensors mounted on the exterior of the vehicles. These exterior mounted sensors disrupt the flow field and create noise sources that contribute to the noise inside the cabin. With wind noise being one of the top customer complaints in recent years, OEMs have to pay attention to the shape and placement of these sensors to improve the wind noise performance of their vehicles.



Traditional methods to test wind noise experienced by the vehicle’s occupants are typically used late in product development, when good quality acoustic prototypes are available. But correcting noise problems discovered late in the product’s lifecycle may require extensive design re-work when the noise problems are closely linked to the vehicle’s exterior shape and styling. Late stage redesign work and potential launch delay cost OEMs millions of dollars. To reduce the associated costs as well as development times, there is strong motivation for the use of computational prediction capabilities early in the vehicle design process.



This study presents the use of a well validated computational solution to evaluate and improve wind noise generated by the sensors on autonomous vehicles, early in the design process. This computational solution uses a Lattice Boltzmann method (LBM) based computational fluid dynamics (CFD) solver to predict the transient flow field and the noise sources. A statistical energy analysis (SEA) solver is used to predict the noise transmitted inside the cabin, due to exterior noise sources. The effect of two different shapes of roof mounted sensors on an SUV on wind noise inside the cabin, is studied using this computational solution. These two shapes are ranked based on their contribution to noise inside the cabin. Noise source analysis is done to understand the effect of these two sensor shapes on noise sources and decide on design improvements to reduce the noise inside the cabin.