Artificial Intelligence deployed to analyze road traffic CO2 emissions

An analysis method developed by a research team at the Swiss Federal Laboratories for Materials Science and Technology (Empa) is able to generate insights regarding changes from one year to the next in the CO2 emissions generated by road traffic in a given country. This new method is based on mathematics and deep learning techniques. According to an Empa press release, it can highlight areas where politicians and consumers could start to reduce CO2 emissions.

Analyzing this has become increasingly difficult in recent years. This is due to the fact that, on account of technical innovations, vehicles can no longer be broken down into traditional categories such as small, medium and luxury classes. In addition, new vehicles are becoming ever bigger and heavier. At the same time, engine displacements are also becoming smaller and engine efficiency is consistently improving.

For this reason, the Automotive Powertrain Technologies department at Empa describes their analysis method as an “important breakthrough” that will make it possible to “evaluate CO2 emissions separately and to perform accurate automatic vehicle classification by analyzing large databases”, explains researcher Nagmeh Niroomand, who adds: “This facilitates the analysis of fleet changes in a country - or in a large company”. The new method is additionally able to eliminate “subjective and expert-based factors”, facilitating comparisons between databases from around the world in the process.

For Switzerland, the team was able to calculate the average CO2 emissions of newly registered cars. The most effective way of promoting decarbonization, according to Niroomand, would be to reduce the numbers of heavy vehicles such as SUVs on Swiss roads. Moreover, buying lower-powered vehicles in the same vehicle class would also be helpful.