Estimating 2D Wind Vectors Using Automotive FMCW Radar

This paper investigates the feasibility of obtaining a horizontal wind velocity vector from 77 GHz automotive Frequency Modulated Continuous Wave (FMCW) radar measurements based on reflections off of suspended water droplets. To that end, a dedicated Eiffel-style wind tunnel is constructed with the ability to inject fine water droplets into the air stream. Wind tunnel design is constrained to a low-cost easy-to-build philosophy, avoiding large metal components where possible to reduce unwanted radar reflections. Estimation of the 2D horizontal wind vector is based on the Velocity Azimuth Display (VAD) method utilizing a least-squares approach. The implementation is first verified and tested using a simulated precipitation scenario. Afterwards, the estimation scheme is applied onto real radar captures from wind tunnel experiments. In both cases, VAD provides acceptable estimates for wind direction and speed.