Russia: These solutions will be deployed on a section of the Baikal-Amur Mainline (BAM) this year. Andrei Romanchikov, Managing Director for the Development of Intelligent Control Systems at TMH, and Pavel Mylnikov, Deputy CEO and Director of the NIIAS’ St Petersburg branch of RZD, announced the plans at the PRO//Motion. Transport Control Systems conference and exhibition.
“This year, for the first time, we are installing a virtual coupling system on 3T28 diesel locomotives. The programme has been agreed with RZD … We have equipped two locomotives on the BAM,” Romanchikov said. “Once the automation work is complete and communications are in place, we will move to the next stage: autonomous operation of mainline trains in virtual coupling mode on the BAM,” Mylnikov added.
The Driverless BAM project is designed to boost the line’s throughput and carrying capacity. It draws on a broad range of technologies, including virtual coupling, automatic train operation, machine vision, radio communication and predictive diagnostics.
Virtual coupling has been rolled out across the network since 2019. It allows two or more freight trains to run one behind the other – RZD has already operated packages of up to five trains – at a minimum safe distance. The locomotives leading each train communicate automatically by radio and adjust their operation accordingly, boosting infrastructure capacity and letting RZD move more freight.
Every new 3ES5K Yermak electric locomotive built by TMH has come fitted with virtual coupling since 2021. Romanchikov said almost 2,500 electric locomotives on the network now carry the technology, which has helped lift freight volumes by 7.6%. TMH is now extending it to the 3TE28 diesel locomotives mentioned above, as well as to the 3TE25K2M and 2(3)ES6 fleets. ROLLINGSTOCK covered virtual coupling in detail in an earlier article.
Interface of the experimental machine vision system on the 3TE38 mainline diesel locomotive. Source: TMH IS
Separately, AVP Tekhnology (part of TMH IS) has been trialling an onboard obstacle detection system on the 3TE28 at the Tynda-Severnaya service locomotive depot, part of the LocoTech Group. “We have already confirmed that the technology can detect objects up to 600 m ahead of the train,” said Anton Suslov, Technical Director of AVP Tekhnology, summing up the trial results at the conference. “We have also established that it can locate objects and measure the distance to them, and confirmed that it can classify and recognise objects in the train’s path.”
Mylnikov said RZD is working with Bureau 1440 on satellite communications for sections beyond the reach of ground-based networks. “As their satellite constellations come online, we’ll be able to connect these locomotives fully to the infrastructure on remote sections and move towards a fully autonomous operating area,” he said.
TMH has said it expects to have GoA3 solutions ready for mainline rolling stock by 2026. Its onboard computer vision system has already been trialled on shunting locomotives and electric multiple units, and is now being tested by Movista Group on a tram in Verkhnyaya Pyshma.
UPD 01/09/2026: RZD has published an official statement confirming the fleet. The trains will be prepared for operation on the Moscow Central Circle (MCC) at automation level GoA3, the last stage before full driverless running.
VSM-Service, which services Lastochka and Sapsan trains in RZD’s fleet, will carry out the installation work. The company confirmed this in response to a request from ROLLINGSTOCK.
RZD also said testing of onboard systems and computer vision for the highest automation level, GoA4 (fully driverless operation), is now in its final stage. The work is being carried out by NIIAS, part of the RZD group. Longer-term plans call for 60 electric trains on the MCC to receive this level of automation; combined with the necessary infrastructure upgrades, this is expected to enable fully driverless operation on the line within a few years.
