Sensing robustness is critical in an automotive environment where there are many sources of noise, amongst them adverse weather is highly unpredictable and can have strong effects on the perception sensors’ data. RADARs are traditionally seen as weather robust sensors in automotive, and their limitation in resolution has been partially resolved with the introduction of 4D RADARs. Whilst increased information can be integrated to support perception, this technology needs a thorough robustness investigation and characterisation. This talk discusses the effect of adverse weather using collected real world automotive 4D RADAR data. This analysis demonstrates that a quantifiable variation can be observed in data collected in adverse weather when compared against a clear condition baseline. The proposed methodology provides a valuable tool to practitioners to quantify 4D data degradation, and the results show and quantify the weather impact on 4D RADAR data, evaluating also the baseline noisiness of the data.”