Authors:Ander Pijoan Lamas, Iraia Oribe García, Oihane Kamara Esteban, Konstantinos Genikomsakis, Cruz Enrique Borges Hernández, Ainhoa Alonso Vicario
The reduction of carbon emissions in the transportation sector is one of the most important steps against the threat of global warming. Unless strict emissions-reduction and fuel economy policies are in place, the resulting pollution is expected to increase dramatically as more vehicles are on the roads. In this line of action, an accurate quantification of the emissions produced by each type of vehicle is essential in order to evaluate the social and environmental impacts derived. The literature shows a wide range of pollutant emission models, whether empirical, database centric or regression based. In this paper, we propose and analyze 3 regression based models built on data from pollutant emission databases and knowledge models. The first model is based on an exponential regression that improves the results given in the state of the art. In contrast, the other two are based on different Artificial Intelligence techniques, namely Artificial Neural Networks and Support Vector Regression, which further improve the results.