Austrian-Chinese collaboration
TU Graz develops autonomous on-demand shuttle
The research project centres on a battery-electric shuttle for up to 15 passengers that can operate on demand instead of following a fixed timetable.
Seisys
Public transport is becoming more flexible as autonomous vehicles move beyond fixed timetables. TU Graz and Chinese partners are testing a 15-seat battery-electric shuttle that can be summoned on demand and uses both onboard sensors and 5G-linked infrastructure to navigate traffic.
Public transport could become significantly more flexible if autonomous vehicles are able to operate beyond fixed routes and timetables. Researchers at Graz University of Technology (TU Graz), together with partners in China, have developed a battery-electric on-demand shuttle for up to 15 passengers. A test vehicle is already operating in real-world traffic in China, while demonstrator vehicles for Austria are being prepared.
The concept is designed to combine the flexibility of private car travel with the efficiency of public transport. Rather than following a fixed line, the shuttle can be requested when needed and deployed across a broader route network. Similar projects are examining how autonomous on-demand services can complement fixed-route public transport as operators look for more flexible alternatives to conventional bus services.
How does the shuttle work without a fixed timetable?
The operating model is intended to close part of the gap between individual mobility and scheduled public transport. Passengers would request the shuttle as needed, allowing vehicles to respond more directly to demand instead of running only according to a predefined timetable.
“Every car journey that can be avoided – especially one made with a combustion-engine vehicle – helps the environment and the climate,” says Cornelia Lex, Head of the Institute of Automotive Engineering at TU Graz. She says the project should make it easier for people to give up individual car journeys by offering a flexible and more sustainable alternative.
How does infrastructure extend the vehicle’s perception?
For automated operation, the project uses what the researchers call collaborative perception. In addition to its own lidar, radar and camera systems, the shuttle receives information from sensors and cameras installed along the route. The infrastructure data is transmitted over a 5G connection, extending the vehicle’s view beyond what its onboard sensors can see directly.
This can help when buildings, parked vehicles or other obstacles block the shuttle’s line of sight. Infrastructure-based perception can provide information about road users or objects that would otherwise remain hidden from the vehicle’s own sensors.
If the communication link fails, an independent safety function is designed to bring the shuttle to a controlled stop. The development partners include Tongji University, Wenzhou University and Chinese technology company Seisys, which provides intelligent traffic-management solutions and acts as a systems integrator for smart-city applications.
What is needed to move from pilots to regular operation?
The TU Graz project reflects a wider European shift in autonomous mobility. According to a recent analysis by PwC Germany, the underlying technology for autonomous shuttles and public transport services is now largely available. The bigger challenge is economic scale: transport authorities, manufacturers and operators must find ways of moving autonomous shuttle services from pilot projects into regular operations.
PwC sees new operating and procurement models as one possible route to faster deployment. The challenge is increasingly about integrating vehicles, infrastructure, fleet operations and commercial models into a service that can work reliably over time.
Projects such as the TU Graz shuttle illustrate what that next stage could look like: demand-responsive vehicles, digitally connected infrastructure and transport services organised around passenger needs rather than a fixed timetable.
