The growth of electric propulsion systems motivates theautomotive industry to transfer the focus from exhaust to non-exhaustemissions, with special attention to brake-related emissions. The literaturelacks well-established approaches that describe the particulate emissionsthrough reliable analytical correlations. Moreover, the mechanisms of brakeparticulate formation entail highly stochastic phenomena, which cannot becaptured by means of traditional deterministic modelling tools. Machinelearning algorithms have been recently used as an alternative method to seekfor a branched correlation between tribological properties (i.e. frictioncoefficient and wear rate), pad composition, environmental and operatingconditions. In this regard, the presented work focuses on the study andidentification of sophisticated meta-models for the prediction of the number ofemitted brake particulate. Specifically, artificial neural networks aredeveloped and validated against brake emission data collected in real drivingconditions at Technische Universität Ilmenau. The developed algorithms areintended for multiple use: (i) in the course of real driving emissions (RDE)testing, to support the experimental data; (ii) while driving, to inform thedriver about the brake-related emission levels; (iii) as an on-boardoptimisation tool that identifies the brake actuation rules to minimise therelease of particulate emissions.