Scuderia Segfault secured a top placement yet again at the 15th F1TENTH Autonomous Grand Prix!

For those of you who haven’t been following Scuderia Segfault’s past successes: The F1TENTH Grand Prix is an international racing competition in which 1:10 scale race cars are fitted with sensors and a computer so they can drive autonomously on a racetrack. The goal? Don’t crash and minimize laptime! This year’s race was held from May 13th to 17th as part of IEEE International Conference on Robotics and Automation (ICRA 2024) in Yokohama, Japan, and was fought out between over 20 teams from universities and research institutes worldwide. In the final head-to-head race competition, the team scored 4th place, making this their forth top-five placement in a row.

Countless hours were invested to prepare for the Grand Prix by the Scuderia Segfault team, which is supervised by Andreas Brandstätter and Rado Grosu, Head of the Research Unit Cyber-Physical Sytstems. First, an optimized hardware setup was needed, with carefully adjusted suspension, the right tires for optimal grip on the racetrack, and visual chassis indicators. The computing platform for this year’s competition was upgraded with a new powerful CPU and much more memory to run complex algorithms more efficiently. Most importantly, the team needed to implement software for fast localization on the racetrack, find the optimal race line, and precisely control the car. The team also spent weekends and evening hours to test the car and prepare for the big race.

Arrival, Trials and Tribulations

After a long journey to Japan, the team arrived early in the morning at the venue to start setting up. Having travelled halfway around the world, the model car needed to be re-assembled and the floor conditions in the competition hall carefully evaluated for proper tire selection. The team created a map of the racetrack so that the car could localize itself and determine its position. To figure out the best way to get around the racetrack, Scuderia Segfault developed a custom-made optimization tool to define the best speed and curve parameters.

The first challenge at the racetrack was to demonstrate obstacle avoidance capabilities. Unfortunately, there were severe problems with one of the software components, which could have led to a complete disqualification from the race, and so urgent modifications were needed to identify the cause of this problem and address it before the start of the time-trial races. The team mobilized all its problem-solving capabilities and worked in separate teams to modify the software, and this ultimately paid off: The check was passed successfully. Following the testing of object avoidence capabilities were the time-trials, in which the model cars drive laps as fast as possible while also demonstrating their reliability. What counts in this competetion is not only the best lap-time, but also the most number of consecutive laps without crashing. Team Scuderia Segfault attained good results (5th place overall) in speed as well as demonstrating object avoidence and reliability, but compared to past competitions, it became clear that the other teams had also greatly improved as the fastest five teams were only seperated by less than a second.