The project called safe.trAIn is also being funded by the German government and will see AI-based methodologies developed for driverless trains to be run across the country’s regional rail network.
The German government has already deployed driverless trains but purely in controlled and closed environments, such as subway tunnels, but Siemens has claimed the safe.trAIn project will ensure these trains can also operate safely in areas with obstructions such as fallen trees or mudslides on the tracks.
Siemens said using conventional automation technology alone for operation will not be sufficient for fully automatic mainline rail operation, but “AI offers major potential in this area”.
A spokesperson for Siemens said: “The challenge that has remained unresolved to date is that of finding a practical way to link AI methodologies to the requirements and approval processes that apply to railways.
“The safe.trAIn project aims to lay a foundation for safe use of AI for driverless operation of rail vehicles and thus address a key technological challenge hindering the adoption of driverless train operation.”
The project, which will run to the end of 2024, will examine AI-based methods for driverless regional trains, validate the safety of the AI components and test processes through collaboration with 16 partners also including BIT Technology Solutions, SETLabs Research, and Otto von Guericke University Magdeburg.
Siemens previously collaborated with Huawei over the introduction of Long Term Evolution (LTE) to the UK railway market after the technology was successfully validated during interoperability testing in France.
