Comparison of different SLAM approaches for a driverless race car
-
Nick Le Large received a Master‘s degree in mechanical engineering from the Karlsruhe Institute of Technology, where he is currently pursuing a Ph. D. degree at the Institute of Measurement and Control Systems. In 2019, he joined KA-RaceIng to participate at the Formula Student Driverless challenge. He was responsible for the localisation and mapping system of KA-RaceIngs’s Formula Student Driverless race car of 2020.
Frank Bieder received a Master‘s degree in electrical engineering and information technology from the Karlsruhe Institute of Technology, where he is currently pursuing a Ph. D. degree in computer vision and machine learning at the Institute of Measurement and Control Systems. Since 2019, he is a research scientist at the Mobile Perception Systems Department, FZI Research Center for Information Technology and a Doctoral Researcher at the Karlsruhe School of Optics & Photonics (KSOP).
and
Martin Lauer received the Diploma degree in computer science from the Karlsruhe Institute of Technology and the Ph. D. degree in computer science from Osnabrück University in 2004. He was a Post-Doctoral Researcher with Osnabrück University in the areas of machine learning and autonomous robots. Since 2008, he has been heading a research group at the Karlsruhe Institute of Technology. His main research interests are in the areas of machine vision, autonomous vehicles, and machine learning.
Abstract
For the application of an automated, driverless race car, we aim to assure high map and localization quality for successful driving on previously unknown, narrow race tracks. To achieve this goal, it is essential to choose an algorithm that fulfills the requirements in terms of accuracy, computational resources and run time. We propose both a filter-based and a smoothing-based Simultaneous Localization and Mapping (SLAM) algorithm and evaluate them using real-world data collected by a Formula Student Driverless race car. The accuracy is measured by comparing the SLAM-generated map to a ground truth map which was acquired using high-precision Differential GPS (DGPS) measurements. The results of the evaluation show that both algorithms meet required time constraints thanks to a parallelized architecture, with GraphSLAM draining the computational resources much faster than Extended Kalman Filter (EKF) SLAM. However, the analysis of the maps generated by the algorithms shows that GraphSLAM outperforms EKF SLAM in terms of accuracy.
Zusammenfassung
Für den Einsatz eines automatisierten, fahrerlosen Rennwagens soll eine hohe Kartierungs- und Lokalisierungsqualität für das erfolgreiche Rennfahren auf bisher unbekannten, engen Strecken gewährleistet werden. Um dieses Ziel zu erreichen, ist es wichtig, einen Algorithmus zu wählen, der die Anforderungen in Bezug auf Genauigkeit, Rechenleistung und Laufzeit erfüllt. Wir schlagen sowohl einen filter- als auch einen glättungsbasierten SLAM-Algorithmus vor und evaluieren diese anhand realer Daten, welche von einem fahrerlosen Formula Student-Rennwagen gemessen wurden. Die Genauigkeit wird durch den Vergleich der SLAM-generierten Karte mit einer durch hochpräzise DGPS-Messungen erfassten Karte ermittelt. Die Ergebnisse der Auswertung zeigen, dass beide Algorithmen die geforderten Zeitvorgaben dank einer parallelisierten Architektur einhalten, wobei GraphSLAM die Rechenressourcen deutlich stärker beansprucht als der EKF SLAM Algorithmus. Die Analyse der von den Algorithmen erzeugten Karten zeigt jedoch, dass GraphSLAM den EKF SLAM Algorithmus im Bezug auf die Genauigkeit übertrifft.
About the authors
Nick Le Large received a Master‘s degree in mechanical engineering from the Karlsruhe Institute of Technology, where he is currently pursuing a Ph. D. degree at the Institute of Measurement and Control Systems. In 2019, he joined KA-RaceIng to participate at the Formula Student Driverless challenge. He was responsible for the localisation and mapping system of KA-RaceIngs’s Formula Student Driverless race car of 2020.
Frank Bieder received a Master‘s degree in electrical engineering and information technology from the Karlsruhe Institute of Technology, where he is currently pursuing a Ph. D. degree in computer vision and machine learning at the Institute of Measurement and Control Systems. Since 2019, he is a research scientist at the Mobile Perception Systems Department, FZI Research Center for Information Technology and a Doctoral Researcher at the Karlsruhe School of Optics & Photonics (KSOP).
Martin Lauer received the Diploma degree in computer science from the Karlsruhe Institute of Technology and the Ph. D. degree in computer science from Osnabrück University in 2004. He was a Post-Doctoral Researcher with Osnabrück University in the areas of machine learning and autonomous robots. Since 2008, he has been heading a research group at the Karlsruhe Institute of Technology. His main research interests are in the areas of machine vision, autonomous vehicles, and machine learning.
Acknowledgment
The authors want to thank the KA-RaceIng Team of 2020 and all its supporters for the time, effort and resources that were spent to enable this project and lead to KA-RaceIng’s victory in the Formula Student Online (FSO) competition in 2020.
References
1. Formula E. URL: https://www.fiaformulae.com/ (accessed 26.02.2021).Search in Google Scholar
2. Roborace. URL: https://roborace.com/ (accessed 26.02.2021).Search in Google Scholar
3. I. of Mechanical Engineers. History of Formula Student. URL: https://www.imeche.org/events/formula-student/about-formula-student/history-of-formula-student/ (accessed 22.05.2020).Search in Google Scholar
4. S. Thrun, W. Burgard and D. Fox. Probabilistic Robotics. Cambridge, Mass.: MIT Press, 2005. ISBN: 0262201623, 9780262201629.Search in Google Scholar
5. M. Zeilinger et al. “Design of an autonomous race car for the Formula Student Driverless (FSD)” (May 2017).Search in Google Scholar
6. M. Valls et al. “Design of an autonomous racecar: Perception, state estimation and system integration” (Apr. 2018).10.1109/ICRA.2018.8462829Search in Google Scholar
7. S. Nekkah et al. “The autonomous racing software stack of the KIT19d”. arXiv e-prints, arXiv:2010.02828 [cs.RO] (Oct. 2020).Search in Google Scholar
8. J. Kabzan et al. “AMZ driverless: The full autonomous racing system” (May 2019).10.1002/rob.21977Search in Google Scholar
9. M. Anderson and M. Baerveldt. “Simultaneous localization and mapping for vehicles using ORB-SLAM2”. M. Sc. Thesis. Gothenburg, Sweden: Chalmers University of Technology, May 2018.Search in Google Scholar
10. K. Wahlqvist. “A comparison of motion priors for EKF-SLAM in autonomous race cars”. M. Sc. Thesis. Stockholm, Sweden: KTH, School of Electrical Engineering and Computer Science (EECS), July 2019.Search in Google Scholar
11. A. Brunnbauer and M. Bader. “Traffic cone based selflocalization on a 1:10 race car” (2019).Search in Google Scholar
12. F. Nobis et al.“Autonomous racing: A comparison of SLAM algorithms for large scale outdoor environments”. In: ICVARS ’19, Feb. 2019, pp. 82–89. DOI: 10.1145/3332305.3332319.Search in Google Scholar
13. L. B. Cremean et al.“Alice: An information-rich autonomous vehicle for high-speed desert navigation”. In: The 2005 DARPA Grand Challenge: The Great Robot Race. M. Buehler, K. Iagnemma and S. Singh (eds.). Berlin, Heidelberg: Springer Berlin Heidelberg, 2007, pp. 437–482. ISBN: 978-3-540-73429-1. DOI: 10.1007/978-3-540-73429-1_14.Search in Google Scholar
14. M. Ester et al.“A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise”. AAAI Press, 1996, pp. 226–231.Search in Google Scholar
15. J. Neira and J. D. Tardos. “Data association in stochastic mapping using the joint compatibility test”. IEEE Transactions on Robotics and Automation 17(6) (2001), pp. 890–897.10.1109/70.976019Search in Google Scholar
16. J. L. Blanco. “Development of scientific applications with the Mobile Robot Programming Toolkit” (Jan. 2008).Search in Google Scholar
17. G. Grisetti et al.“A tutorial on graph-based SLAM”. IEEE Intelligent Transportation Systems Magazine 2(4) (2010), pp. 31–43. DOI: 10.1109/MITS.2010.939925.Search in Google Scholar
18. R. Kummerle et al.“G2o: A general framework for graph optimization”. In: 2011 IEEE International Conference on Robotics and Automation (2011), pp. 3607–3613.10.1109/ICRA.2011.5979949Search in Google Scholar
19. D. W. Marquardt. “An algorithm for least-squares estimation of nonlinear parameters”. Journal of the Society for Industrial and Applied Mathematics 11(2) (1963), pp. 431–441. ISSN: 0368-4245.10.1137/0111030Search in Google Scholar
© 2021 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Beiträge
- Terahertz based non-destructive testing (NDT)
- X-ray imaging and computed tomography for engineering applications
- Comparison of different SLAM approaches for a driverless race car
- Combination of optical metrology and non-destructive testing technology for the regeneration of aero engine components
- Capacitive micromachined ultrasonic transducers: Transmission evaluation with different membrane materials and dimensions
Articles in the same Issue
- Frontmatter
- Beiträge
- Terahertz based non-destructive testing (NDT)
- X-ray imaging and computed tomography for engineering applications
- Comparison of different SLAM approaches for a driverless race car
- Combination of optical metrology and non-destructive testing technology for the regeneration of aero engine components
- Capacitive micromachined ultrasonic transducers: Transmission evaluation with different membrane materials and dimensions
