15.11.2018

Audi Autonomous Driving Cup: FZI Team AlpaKa Wins the Title

At the final of the Audi Autonomous Driving Cup 2018, the team AlpaKa won against seven other university teams and relegated the competitors to their places. The prize money of EUR 10,000 therefore goes to Karlsruhe.

Am 14.11.2018 haben die Studenten Simon Roesler, Mark Timon Hüneberg, Yimeng Zhu, Maximilian Zipfl und Shuxiao Ding des Team AlpaKa mit Unterstützung durch das FZI den Audi Autonomous Driving Cup 2018 gewonnen.
On 14.11.2018 the students Simon Roesler, Mark Timon Hüneberg, Yimeng Zhu, Maximilian Zipfl and Shuxiao Ding from the team AlpaKa won the Audi Autonomous Driving Cup 2018 with the support of the FZI.

The three-day final event of the Audi Autonomous Driving Cup 2018 included the overcoming of prescribed, partially unknown driving tasks on a street parcours. The solution approaches used for this purpose were presented in an academic lecture and included in the evaluation. Moreover, there was an “open challenge” in which the teams presented innovative and creative approaches on the topics of Car2x, artificial intelligence and closed loop in autonomous driving.

Open challenge: optimal networking and use of artificial intelligence is required

The overall winner AlpaKa also showed the best performance in this challenge in which the team fascinated with its contribution and won the special prize. The team AlpaKa consisting of the students Simon Roesler, Mark Timon Hüneberg, Yimeng Zhu, Maximilian Zipfl and Shuxiao Ding therefore developed a cloud-based shuttle bus system that automatically uses information of other traffic participants – for example about current track closures or for unclear situations like breakdowns – for the planning and the real time adjustment of routes. In addition, a visualisation of the sensor-based vehicle environment model was presented, which enables a tele operator to remotely control a shuttle bus from a central office. This was demonstrated within the framework of a concrete application scenario in which the autonomous vehicle independently determines and gives the information that the current situation cannot be tackled in a safe and reliable way.

Solving driving tasks independently with prepared high-tech model cars

The team could also convince in other parts of the competition. The robust lane detection as well as the detection of obstacles, police vehicles and people on the roadside helped the vehicles to drive smoothly and safely through the model environment. Challenges, which were introduced this year, could also be successfully solved by the team AlpaKa: For example, crossing a road over a bridge and the subsequent driving on inclined roads caused difficulties for some teams. The students were also able to master the challenge through a specially designed regulatory approach.

“I am proud of the performance of our team. As a transfer institute, the FZI always aims to transmit knowledge and know-how and to qualify young academics for the challenges of future mobility. I am very pleased to see how the students developed, together with our employees with a lot of enthusiasm and with total commitment, solutions for the tasks of the competition and are now rewarded for their efforts with victory!”, congratulates the member of the FZI’s Board of Executive Directors and Director of the FZI, Prof. Dr.-Ing. J. Marius Zöllner, AlpaKa on the success.

The team AlpaKa was supported by FZI employees in the framework of the project “Karlsruhe Applied Machine Learning Living Lab” (shortly: KarAMeL³), which is funded by the German Federal Ministry of Education and Research under the funding code 01IS17062.

Team AlpaKa succeeded

The Audi Autonomous Driving Cup 2018 attracted an international line-up. Ten nominated university teams – including teams from the United Kingdom, Italy and Austria – had been supplied with the basic software and two model cars at a launch event in June 2018. At the test event in October 2018, five teams with identical model cars supplied by their departments also took part: high-tech model cars with front camera, reversing camera, laser scanner, and ultrasound and acceleration sensors.

“I was really impressed by the students’ creativity and commitment. The young software developers have exactly what we will need for Audi in the future,” said Thomas Kamla, Head of Strategy and Innovation Management at AUDI AG, who presented the winner’s certificate and EUR10,000  prize money at the Audi Forum Ingolstadt. The team frAISers from the University of Freiburg was ranked second and received a prize money of EUR 5,000. Team OptILmDriveX from the Technical University of Ilmenau was ranked third, earning EUR 1,000 prize money.

For more information and results, visit: www.audi-autonomous-driving-cup.com