Looking to the future with camera gantries

In 2022, camera gantry technology for freight wagon diagnostics was successfully tested and made the transition to a series concept at DB Cargo as part of the Shift2Rail European funding programme. The Germany-wide rollout of the camera gantries was successfully completed on schedule at the end of 2022 when the last camera gantry went into operation in Kornwestheim. The camera gantries create images of passing freight wagons from different perspectives. These images are then sent to employees, who use them to order maintenance services at DB Cargo maintenance depots.

5 July 2023: Nationwide rollout of AI in rail freight transport

The first camera gantry was tested at the Nuremberg marshalling yard in 2017, and now another milestone has been reached. Dr Sigrid Nikutta, Member of the Management Board for Freight Transport at Deutsche Bahn, and Judith Gerlach, Bavarian Minister for Digital Affairs, launched the nationwide use of AI applications in practice in rail freight transport today. These applications are used to analyse the images taken by the cameras and provide information on the condition of the freight wagons and load.

"Rail freight transport offers tried and tested logistics and already saves us seven million tonnes of CO2 every year," says Dr Sigrid Nikutta, Member of the Group Management Board for Freight Transport. "Artificial intelligence will now become an integral part of our operations, starting today. It will help us get more freight wagons and thus more freight onto climate-friendly rail."

The operation of the camera gantry in practice

Click to skip the following slider

  of

End of the slider

• 
• 
• 
• 

  of

There are now a total of 13 camera in eight DB Cargo locations, which are gradually being equipped with the new AI technology. They will record up to 300,000 high-resolution images from about 10,000 daily freight wagon movements. After the AI software was successfully piloted for photo analysis in rail freight transport, it is now being used at DB Cargo locations throughout Germany. AI provides information on the condition of brake blocks, for example. The camera technology also creates an additional perspective from above. Automatic scrap type detection is currently being tested with the first customer from the steel industry.
 
The camera gantry at the Nuremberg marshalling yard allows careful analysis of the freight wagon on screen. Artificial intelligence methods also increase efficiency for maintenance commissioning employees. This has created a new, attractive job profile, the digital diagnostics expert. AI will also reduce the future workload of wagon inspectors, who are responsible for trackside diagnostics of freight wagons. Without the technology, it currently takes more than an hour to inspect a freight train, depending on the train's length.

AI model supports freight wagon diagnostics

As part of another research and development project of the Shift2Rail European funding programme, the images from the camera gantries were used for automatic freight wagon analysis, including analysis of brake blocks. Specifically, the thickness of the brake blocks is recorded. The artificial intelligence (AI) model developed was tested, and comparative measurements were also carried out to assess the quality of the values calculated by the AI model.

Camera gantries and artificial intelligence will equip us for the rail freight transport of the future.

The aim of the tests within the project was to evaluate the user-friendliness of the system, test the functions and specifically check how helpful the AI model's suggestions are in order to preventively replace brake blocks. The ultimate aim is to increase the availability of rolling stock. In addition, business processes were identified that could be optimised in the future using data from continuously measured brake blocks.

Close cooperation with users

In the future, digital freight wagon diagnosticians will have AI to help them. AI will provide support during the technical analysis of the vehicle's condition, allowing employees to order exactly the high-quality maintenance services that are needed. This solution was developed in close cooperation with specialists from the industry. 

Feasibility successfully demonstrated

The project supported by Shift2Rail demonstrated the feasibility of AI in the freight wagon inspection process. During the tests, it was shown that the system can be used in the process and that it simplifies diagnostics with additional information. One of the next steps is to optimise the model further in order to integrate it into the standard rail freight transport process. This success based on digitalisation and automation is part of the path towards establishing a smart rail freight transport system in order to bring more goods onto environmentally friendly rail.

^{_{This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No:101004051. The content of this document reflects only the author’s view.}}