Aerodynamic Methods of Maintaining Advanced Driver-Assistance System Sensor Operability in Adverse Conditions

The impacts of surface contamination, or soiling, have been a long-standing issue in the automotive industry, with research going back as far as the 1960s. Until now, the impact of contaminant build-up has been largely cosmetic. However, with the advent of autonomy and other advanced driver-assistance systems (ADAS), surface contamination may have more serious repercussions. ADAS rely on sensors to monitor the external environment in which they operate and surface contamination risks obstructing this ability, limiting functionality and increasing the likelihood of road traffic accidents.



Previous studies at Loughborough University have shown that very good levels of agreement can be found between experimental studies and computational simulations using a high-fidelity (Detached Eddy Simulation with Lagrangian particle tracking) numerical method. These studies have largely concentrated on the soiling of the rear of SUV type geometries. This project will consider the spray on, and suspended in front of, ADAS sensor locations around the vehicle. This will include consideration of the contaminated, high turbulent, wake of other vehicles. The first aim of the project will be to use experimental and computational studies to find the level of contamination that can be expected at sensor conditions for a baseline configuration. Using this data, the intent is to quantify the effects of contaminant build-up and identify possible mitigatory aerodynamic designs.

The main body of the work will be to investigate the use of aerodynamic control methods to reduce the impact of road spray on ADAS systems. Data analysis methods such as proper orthogonal decomposition (POD) techniques will be used to identify the main flow structures responsible for contamination and highlight mechanisms to reduce particle entrainment into vehicle wakes.
These will fall into two categories; the first will be to investigate methods of reducing the amount of spray that is entrained into the vehicle wake from the wheels, the second will consider active devices, such as air barriers, that generate streams of local, high-energy air, to help shield sensors from impinging contaminants. Methods will be developed in simulation before tested in experiment. As well as considering the effectiveness in terms of contamination reduction, the 'cost' in terms of energy use or equivalently vehicle drag increase will be determined.