Automated detection of train malfunctions with Train Scanner
Currently, a large part of rolling stock maintenance is manual inspection performed by human eyes.
The objective has therefore been to proceed towards automated and systematic processes. In June, the rolling stock maintenance company VR FleetCare will deploy Train Scanner, a device designed to automatically inspect the external condition of trains and use the device for the maintenance of Helsinki Region Transport’s rolling stock. The objective is to be able to reliably identify any malfunctions and stains on the exterior of trains.
The device has been developed in cooperation with VR FleetCare and Vire Labs, a Finnish company providing IoT solutions based on machine intelligence.
Machine doing the work
Train Scanner is based on machine vision, shape recognition and artificial intelligence. 6.5-metre-high poles with scanners are installed on both sides of the track and used to scan the train that drives by.
The sides and roof of the train are scanned with a line camera with extremely high imaging speed and analysis accuracy. The resolution is approximately a few millimetres. The data from the scanner is instantly processed by means of edge computing that analyses the data and reports any deviations. The information is available for the train maintenance operator in almost no time. The data is also stored in a cloud service for further analysis.
“The shutter speed and resolution of the camera could be compared to photo-finish cameras used in sports competitions, for instance,” says Samuli Suuriniemi, Project Manager at VR FleetCare who oversaw the development work of Train Scanner. “We taught the device how a train should look like so that it can automatically detect any deviations. Artificial intelligence becomes increasingly more accurate based on the feedback the device receives from humans. The final objective is to be able to detect any rolling stock malfunctions in a reliable manner,” Suuriniemi explains.
First scanner to be deployed in June
Train Scanner has now been piloted for one year with excellent results. In June, the device will be deployed by the first customer and owner of rolling stock. Flirt commuter trains of the capital region will be inspected using Train Scanner by driving the trains past the device before arriving to the depot.
“There are various benefits related to implementing the system. The condition of rolling stock is constantly assessed, and safety is improved as the system can detect the malfunctioning components of a bogie, for example. By utilising the available data, it will also be possible to accurately define maintenance intervals in the future. Inspection intervals can also be easily increased,” Suuriniemi adds.
Up until now, Train Scanner has been only tested with passenger trains but due to its scalability, the device can be used to inspect any rolling stock. In border control, for instance, Train Scanner could be used at border stations to automatically detect whether the size of a wagon adheres to the local instructions or used to verify the safety of wagons.
Affordable and easy to implement
Train Scanner is not a completely unique invention. A few devices with similar user purposes already exist on the market.
“The prices of the available devices are quite high, and their installation requires extensive construction work. Our intention has been to develop a cost-effective solution that is scalable and possible to implement quickly,” Vice President of Digital Services Mihail Lipiäinen from VR FleetCare explains.
“Automated inspection of rolling stock will be one of the most important changes made possible by the current technology, and the benefits for the owner of the rolling stock include improved quality of maintenance and better usability of the rolling stock,” Lipiäinen says.