Artificial intelligence analyzes CO2 emissions from traffic
A team from Empa uses artificial intelligence to analyze the CO2 emissions of the Swiss car fleet. Thanks to the new method, databases around the world will be comparable.
An analysis method developed at the Eidgenössische Materialprüfungs- und Forschungsanstalt ( Empa ) can make statements about how the consumption of a country’s vehicle fleet changes from year to year. This new method is based on math and deep learning techniques. According to a communication , it is able to show where politicians and car buyers could start to reduce CO2 emissions.
Analyzing this has become increasingly difficult in recent years. Because vehicles can no longer be divided into classic segments such as small, medium and luxury classes due to technical innovations. In addition, new vehicles are getting bigger and heavier. In addition, the cubic capacities would decrease, while the efficiency of the engines would get better and better at the same time.
That is why the Empa Vehicle Drive Systems department describes its analysis technology as an “important breakthrough”: It enables “CO2 emissions to be assessed separately and an accurate automatic vehicle classification to be carried out by analyzing large databases,” explains researcher Naghmeh Niroomand. “This makes it easier to analyze changes in fleets in a country or a large company.” Thanks to this new method, “subjective and expert-based factors” would be eliminated and databases from all over the world would be comparable.
For Switzerland, the team was able to calculate the average CO2 emissions of newly registered cars. If less heavy vehicles such as SUVs were on Swiss roads, this would be the most effective way of promoting decarbonization, says Niroomand. It would also be helpful to buy vehicles with lower performance in the same vehicle class.
