Experimental performance assessment of an automated shuttle in a complex, public road environment

Automated, electric shuttles are expected to be key for the future of public transportation, providing a safe, efficient, and robust operation with a minimum carbon footprint. However, in complex, urban environments, their reliable operation is particularly challenging and shows a lack of performance and comfort. This study presents a quantitative benchmark of an automated shuttle compared to a conventional, human-operated bus on the same route. Speed and acceleration across geofenced segments are systematically analyzed based on over 12 million GNSS and IMU data points. The results show that the automated shuttle operates at about half the average speed of the bus. Furthermore, frequent abrupt decelerations are reducing passenger comfort, while the main distributions and mean values of the measured acceleration indicate a smooth operation of the automated shuttle; outliers reveal critical braking events. The presented methodology enables objective performance tracking and supports the iterative improvement of autonomous shuttles through datadriven optimization.

DDC Class

624: Civil Engineering, Environmental Engineering

625: Road and Railroad

Lizenz

https://creativecommons.org/licenses/by/4.0/

Publication version

publishedVersion

Name

futuretransp-05-00165.pdf

Size

10.82 MB

Format

Adobe PDF

Options

Experimental performance assessment of an automated shuttle in a complex, public road environment