Porsche AG organized the first "Big Data Challenge" in Ludwigsburg from 31 January to 2 February. Using a mountain of data from 4,500 Porsche testdays, the participants were to make precise predictions about the condition of the road surface during the test drives, without knowing the real weather conditions. The benefit in practice: an autonomous vehicle that reliably predicts wet surfaces can adjust its driving behaviour in good time. This is an important safety aspect that can prevent an accident, especially when starting or braking the vehicle on slippery roads.
Out of 90 applicants, seven teams were invited to the Hackathon, including masters students, Porsche employees and teams from partner companies. GIGATRONIK was represented by a four-member analyst team from the Analytics & Diagnostics department. During the three-day Hackathon, GIGATRONIK's data specialists were able to predict the road condition with almost 94 percent accuracy. This result was achieved through a machine-learning model developed by the GIGATRONIK experts in Python.
The jurors unanimously voted the GIGATRONIK team the winner of the Hackathon. The jury consisted of Thilo Koslowski (Managing Director Porsche Digital), Anja Hendel (Director of the Porsche Digital Lab) and Alexander Thamm (Managing Director Alexander Thamm GmbH) as well as other Porsche divisional and departmental managers.
Peter Mehrle, Managing Director of GIGATRONIK, is pleased about the success: "The performance of the Hackathon team once again shows the outstanding expertise of our developers in the field of analytics. High qualification, mutual support and the bundled experience from past analysis projects in the automotive sector are the decisive success factors."