Modelling Weather Precipitation Intensity on Surfaces in Motion with Application to Autonomous Vehicles

Area of Science:

• Autonomous Systems
• Environmental Science
• Mechanical Engineering

Background:

• Autonomous vehicle (AV) performance degrades in adverse weather, impacting sensor functionality due to visibility reduction and precipitation accumulation.
• Understanding precipitation effects is crucial for reliable AV operation in diverse environmental conditions.

Purpose of the Study:

• To develop and validate a model for quantifying precipitation (rain and snow) as perceived by moving autonomous vehicles.
• To analyze the impact of weather parameters like wind and particle size on precipitation intensity.
• To identify optimal sensor orientations for minimizing precipitation flux on AV surfaces.

Main Methods:

• Developed a mathematical model incorporating wind direction, particle size distribution, and surface inclination.
• Calculated precipitation intensity on various inclined surfaces and along a circular driving path.
• Partially validated model outputs against direct measurements from a test vehicle.

Main Results:

• The model demonstrated a strong correlation between its outputs and experimental data for both rain and snow.
• Results highlight the critical influence of sensing surface angle on perceived precipitation levels.
• Identified optimal surface orientations for minimizing precipitation accumulation.

Conclusions:

• The proposed analytical model accurately quantifies precipitation effects on AVs.
• Findings are vital for developing mitigation strategies against heavy precipitation and designing robust sensor systems.
• The model aids in determining optimal sensor positioning to enhance AV performance in adverse weather.