Robustness Evaluation of Localization Techniques for Autonomous Racing

• Published in: 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE) pp. 1 - 2
• Main Authors: Lim, Tian Yi, Ghignone, Edoardo, Baumann, Nicolas, Magno, Michele
• Format: Conference Proceeding
• Language: English
• Published: EDAA 25.03.2024
• Subjects: Autonomous Racing; Benchmark testing; Computational modeling; Computers; Filtering algorithms; Graphics processing units; Localization; Location awareness; Monte Carlo Localization; Particle Filter; Robustness; Simultaneous localization and mapping; SLAM
• ISSN: 1558-1101
• DOI: 10.23919/DATE58400.2024.10546662

This work introduces SynPF, an MCL-based algorithm tailored for high-speed racing environments. Benchmarked against Cartographer, a state-of-the-art pose-graph SLAM algorithm, SynPF leverages synergies from previous particle-filtering methods and synthesizes them for the high-performance racing do-main. Our extensive in-field evaluations reveal that while Cartogra-pher excels under nominal conditions, it struggles when subjected to wheel-slip-a common phenomenon in a racing scenario due to varying grip levels and aggressive driving behaviour. Conversely, SynPF demonstrates robustness in these challenging conditions and a low-latency computation time of 1.25 ms on on-board computers without a GPU. Using the FITENTH platform, a 1:10 scaled autonomous racing vehicle, this work not only highlights the vulnerabilities of existing algorithms in high-speed scenarios, tested up until 7.6 \text{ms}^{-1} , but also emphasizes the potential of SynP F as a viable alternative, especially in deteriorating odometry conditions.