Big Data Solutions for Porsche Driverless Technology
Software to make it safer for autonomous vehicles to drive on wet and slippery road surfaces.
The race to driverless cars is pitching tech giants against traditional car manufacturers in a global effort to build businesses selling vehicles that don't need human drivers. It has been the stuff of science fiction for decades, but we may now be approaching the time when motorists find themselves behind the wheel of a car they don't have to drive. The technologies involved are complex, the obstacles are still many, and the overall challenge is immense.
Big Data Challenge
Luxury car maker Porsche is working on a system that would give their customers the opportunity to use their cars how they want - a driverless option or a human-centric one. To help it develop one aspect of driverless technology Porsche AG hosted their first Big Data Challenge at the beginning of 2018. It was open to mathematicians, computer scientists and anyone else with an interest in data science.
The contest was the brainchild of Bettina Frana (Chassis Innovation & Strategy, Porsche AG) and Tobias Grosse-Puppendahl (Innovation Management Electrical/Electronic Engineering, Porsche AG). They were encouraged by seeing how well hackathons have served other companies find solutions to their challenges.
Participants were given access to a huge battery of data from 4,500 Porsche test days and were tasked with trying to make precise predictions about the condition of the road surface during the test drives, without knowing actual weather conditions. The benefit of this in the real world would be that if a self-driving vehicle can accurately predict wet surfaces it will be able to adjust its behavior in good time.
In an interview with medium.com Frana and Grosse-Puppendahl talked about their aspirations for the hackathon: “We hope that the hackathon participants will be able to provide valuable food for thought for our challenges; ideally, we want to learn from them and use them for our developments at Porsche.
“It is particularly important for us that we also maintain good contacts with the external participants, so that we can identify talent and inspire them to apply for a position at Porsche, for example.”
Porsche received 90 applications and selected seven teams to take part in their three-day hackathon. Among them were Porsche employees, master’s students and teams from partner companies.
The contest jury awarded first prize to a four-member team from GIGATRONIK a company that is a development partner for the automotive industry for electronics and information technology. The winning team was able to predict road conditions with close to 94 percent accuracy thanks to a machine learning model that had been developed by some of the company’s experts.
Porsche treated their first big data hackathon as a pilot project and will analyze the results, feedback and interest to help them develop future hackathons in an automotive environment.
Next Story »