Managing Vehicle Test Data

Modern vehicles are increasingly complex – especially as ADAS and autonomous driving technologies continue to mature. And in recent years there has been a direct impact on the breadth and depth of lab and proving ground test data sets.

Koala Technologies serves both lab-side and proving ground -side vehicle development programs with specialized solutions that cover everything from chassis and component testing, tire testing, and more. So, we’ve definitely seen the recent explosion in data collection and data integration needs.

On the lab side, consistency and repeatability dominate the testing and data collection requirements. On the proving ground (full vehicle testing) side, the emphasis is on environmental conditions and, in many cases, scenario diversity. In both cases, the challenge for automotive engineers is to manage increasingly large and complex data sets.

A recent article in ADAS & Autonomous Vehicle International magazine, titled “Data Pipeline,” provides an excellent overview of one possible way to structure data collection and analysis. Logical stages are proposed, consisting of: Define | Collect | Curate | Train & Validate.

This is not a look at individual tools or test scenarios, but instead it offers a framework for making sense of data sets – where “data” is not simply a numerical output from lab and proving ground testing, but a strategic vehicle development asset that can be managed.