22.07.2019

Researchers of the FZI Appeared in ATZ extra with the Article "Absicherung der Reichweitenschätzung basierend auf aufgezeichneten Realdaten"

The energy consumption of an electric vehicle strongly depends on driver-, vehicle-, road- as well as environment-related factors. A precise forecast of the remaining range is thus essential for a high acceptance and the comprehensive introduction of electric vehicles. The FZI Research Center for Information Technology develops methods for a data-based development support of vehicle functions.

Image source: Pixabay

The energy consumption of an electric vehicle strongly depends on driver-, vehicle-, road- as well as environment-related factors. A precise forecast of the remaining range is thus essential for a high acceptance and the comprehensive introduction of electric vehicles. The FZI Research Center for Information Technology develops methods for a data-based development support of vehicle functions. In this context, researchers of the FZI were invited to write an article entitled "Elektromobilität" (electric mobility) on the topic of "Absicherung der Reichweitenschätzung basierend auf aufgezeichneten Realdaten" (safeguarding of the range evaluation based on recorded real data) in the special issue ATZ extra. Using the example of the range prediction, they demonstrate in the article how recorded real data from vehicles contribute to the simulation-based safeguarding as well as development and optimisation of a reliable and precise range prediction.

Functions in the field of electric mobility as well as relating to autonomous driving in particular, which are for the most part based on procedures of artificial intelligence (AI), require a new approach as well as the utilisation of recorded real data in the development procedure at an early stage. Together with partners, the FZI intensively researches on data-driven methods and tools for the development and safeguarding of automotive functions and systems. The aim is to analyse real data from different vehicles  – from the first prototype to serial fleets. In doing so, critical and relevant situations shall be identified on the one hand and the highest possible realism shall be ensured for simulations at the desk on the other hand. New software versions of algorithms can thus be scalably evaluated and tested realistically and on a wider database at early stages of the development. The constantly increasing data pool of recorded rides allows comprehensive data analyses and statistically significant assessments, without referring on new test drives or extensive simulations.

The (paid) article can be found via the following link. Contact person at the FZI is Stefan Otten, vice division manager of Embedded Systems and Sensors Engineering (ESS).