Experience from Three European Cities Provided Lessons on the Practical Aspects of Operation for Setting Up Fully Automated Shuttle Demonstration Sites.
Lessons learned from setting up a demonstration site with autonomous shuttle operation – based on experience from three cities in Europe
Summary Information
Transportation agencies often have limited knowledge about what to expect when starting a fully automated shuttle project. This study described lessons learned from setting up automated shuttle operations in three different cities in Europe: Brussels (Belgium), Linköping (Sweden) and Turin (Italy), focusing on the practical aspects of operation. The deployments operated different brands and number of shuttles, different types of infrastructure and varying local conditions. In total, five SAE Level 4 automated shuttles were utilized across five different sites. The Belgian shuttle service operated at three sites: Woluwe Park, the Solvay business campus and the Brugmann Hospital. Each shuttle operated for four-month periods between 2019 and 2020. The shuttle service set up at the Sweden site ran within the Campus Valla area at Linköping University between March 2020 and June 2021, while the demonstration period for the site in Italy took place at the International Training Centre of the International Labour Organization (ITC-ILO) campus between February and July 2020. It should be noted that passenger operation was limited during the operational period(s) due to the Covid-19 pandemic.
- Use sound-based Radio Detection and Ranging (radar) technology to complement the Light Detection and Ranging (LiDAR) technology to mitigate weather-dependent braking and stopping issues. The shuttle should then be able to recognize various weather conditions and decide to utilize the radar instead of the LiDAR in case of fog, for instance.
- Use Global Navigation Satellite Systems with Real Time Kinematics to determine vehicle location more accurately. Real-time kinematics can help eliminate issues caused by LiDAR-based localization systems.
- Take injury mitigation actions in terms of seat belt usage for the protection of shuttle riders. The combination of no seat belt and wooden seats increases the risk of falling off the seat in hard braking events. It is recommended to provide seat belts on seats facing forward (and if possible, upholstered seats), until hard braking events can be avoided. Additionally, in one site, the shuttles were equipped with a sign on the back asking trailing vehicles to keep a safe distance to the shuttle.
- Consider fitting into the existing traffic environment rather than adapting the road infrastructure to enable the fully automated shuttles. For wider use of shuttles, it is vital that shuttle development prioritizes fitting into the existing traffic environment and ecosystem, instead of adapting the road infrastructure to enable the fully automated shuttles. Further development of the fully automated shuttles is required to ensure that they operate smoothly in complex traffic situations considering lane and road width, shared spaces, snow, dust, rain, leaves, birds, etc.
- Gather a knowledge base for properly charging, cleaning, and maintaining fully automated shuttles. Shuttle-specific requirements and guidelines should be incorporated. For example, for countries with a cold climate, bus depots must be heated during wintertime to avoid temperatures below 0 °C (32°F) to prevent damage to the batteries.
- Do not underestimate the complexity of the fully automated shuttle program. A shuttle operation needs an ecosystem of stakeholders (e.g. local and regional partners, landowners, safety operators, insurance companies and users). A fully automated shuttle is a new component, and this study indicated the number of adaptations regarding road infrastructure before the realization of its operation. It was recommended that the effort necessary to have fully automated shuttles up running should not be underestimated.
