Despite constant technological innovation, road transport remains a significant source of pollutant emissions, and effective driver-behaviour changes can be considered as solutions that can increase the sustainability of road traffic in a short period. Thus, understanding driver behaviour plays a key role in assessing traffic-related impacts. Since real-world experiments entail some risks and are often not flexible, simulator-based experiments can be relevant to studying vehicle dynamics and driver behaviour. However, the reliability of the simulation results' accuracy must be ensured. The primary objective of this paper is to present an exploratory analysis focused on the study of the reliability of a driving simulator to reproduce driving parameters that can then be used for emission estimation. For that purpose, tests were conducted by two drivers for urban and highway scenarios performed on a driving simulator and in real-world environments. Different road singularities composed events that were microscopically analysed. Second-by-second vehicle dynamic variables were recorded, and the pollutant emissions were estimated using the vehicle specific power (VSP) methodology. The results of this exploratory validation analysis showed that the total average emissions of all events were not significantly different (958.39 g for simulated and 998.06 g for empirical tests). Overall, the driving simulator can replicate vehicle dynamics from a microscopic perspective, especially for the urban scenario. This may be due to the more complex traffic conditions and road specificities that require more restrained driving behaviour. Nevertheless, VSP mode distributions did not follow the same pattern in 4 out of 10 events, meaning that the drivers displayed different behaviours in the simulated and empirical tests for those events. The relative errors range between 4 and 29% for carbon dioxide emissions and between 2 and 33% for nitrogen oxides emissions.
Keywords: driving behaviour; emissions; simulator vs. empirical tests.