Geneva Demonstrator Sites

The Geneva Canton, composed of 45 municipalities and with a population of 500,000 inhabitants, is one of the most densely populated areas of Switzerland. In the last 40 years, the region’s population increased by almost 50%, the biggest part of this rise having been absorbed by the suburban communities. As a result, the population of many rural areas has been increasing drastically during the last 40 years, with an explosion of private car commutes. The city is also a major regional economic centre, and each day more than 100,000 people enter the canton to work. The Geneva Public Transport Network, named UNIRESO, is an interconnected system of trams, buses, trolleys, mouettes (transport boats) and trains that is fully integrated and uses the same tickets as well as coordinated timetables. The public transport system in the Geneva Canton represents the most common model of public transport, based on a backbone of surface tramways (33km) and a network of buses (360km) and trolleys (30km).

The pilot sites

Transport publics Genevois company (TPG) has hot-started the work of the AVENUE project by directly integrating an operation of autonomous minibus (AM) service in Meyrin, a suburban industrial and residential commune of Geneva with more than 25’000 inhabitants and one of the highest population growth rate in Switzerland. The service, in the form of a conventional line called the Xa line, was perfectly integrated within the TPG networks, operated on a high frequentation public roads itinerary and has been serving some regular passengers during peak hours. Due to major works on its route, the service ended in February 2020.

The project then settled on the Belle-Idée site which is situated within a medical/psychiatric medical facility in Thônex, a nearly 15,000 inhabitants peri-urban commune in the Geneva area. This hospital is part of the Geneva University Hospital (HUG), which is committed to the development of green mobility initiatives. This new demonstration was set up in spring 2020, and remained in operation beyond the end of the AVENUE project. It has been AVENUE’s most advanced automated minibus public transportation deployment, operating a regular on-demand service in open, mixed traffic roads. Within this demonstration site, three Navya^® autonomous minibuses are deployed and operate under mission management mode which is receiving directly the destination via the fleet orchestration service (by Bestmile and later by IOKI), without any operator intervention. Passengers use the provided mobile app (developed by MobileThinking) to order a ride, which is transmitted to the Fleet Orchestration Platform that chooses the optimal path and vehicle to fulfill the request. The solution proposed here is end-to-end operator-free, from depot door opening at the start of the service to the end. The service operates day and night, and under diverse weather conditions, serving 70 virtual stops covering the entire 38hectares domain from Monday to Friday from 8:00 to 22:00.

During this experiment, the vehicles faced regular traffic on a daily basis for weeks without generating a single incident. The behavior of the vehicle was rough and unsteady at first, but day after day it improved greatly and the journeys are now smooth, only poorly parked cars and vegetation still cause some issues. We reached almost 100% automated operation during the initial testing and validation and during regular operation.

Belle-idée site achievements

• Operating seamlessly in urban environment
• Operating on busy public roads with many obstacles and heavy traffic
• Implementing an on-demand service
• Implementing a dynamic routing system
• Integrating AMs within the existing public transport system
• First Swiss AM vehicle homologation
• Contributing to the Swiss autonomous driving legislation evolution
• Transporting regular customers’ base
• Helping many mobility impaired persons within the Belle-idée domain
• Testing long term operation with a very high level of safety and reliability
• Participating in the improvement of obstacle and vegetation detection