A City on Self-Governance
Russian cities are moving away from the era of the human dispatcher and toward an era of artificial intelligence – systems that see everything and manage everything.
A City Needs Its Own Intelligence
Cities are struggling to breathe. Every year, traffic grows faster than new roads can be built. Moscow, St Petersburg, Yekaterinburg, Novosibirsk – the pattern is the same everywhere: rush hour, gridlock, long queues at stops, people waiting, running late for work. A human dispatcher cannot manually optimise the movement of hundreds of buses, trolleybuses, and trams while simultaneously accounting for weather conditions, congestion levels, and traffic incidents. What is needed is an intelligence that sees the city as a whole, thinks faster than a human, and makes decisions in real time. Digitalisation and artificial intelligence are no longer a luxury – they are a matter of urban survival.
Machines That See, Hear and Think
In September 2025, a fully autonomous tram carrying real passengers took to the streets of Moscow for the first time. The capital’s transport development strategy calls for autonomous control of around two-thirds of Moscow’s tram fleet by 2030, rising to nearly 90% of all trams by 2035.
At the same time, a quiet revolution has taken place in urban traffic management. In Vladimir Region, an intelligent traffic management system has been deployed that does more than respond to congestion – it predicts it. Algorithms analyse traffic density, public transport load, live feeds from road cameras, and weather data in real time to anticipate problems before they occur. The result: average travel speeds have increased from 35 to 40 kilometres per hour, the number of forced stops has fallen by 25%, and drivers are saving up to 15 minutes on every half hour of travel.
In Nizhny Novgorod, a neural network developed by NtechLab has begun analysing public transport flows. The system monitors stops via roadside cameras, counts waiting passengers, identifies overcrowded locations, and automatically alerts dispatchers. Several other regions are preparing to launch similar pilot projects.
Researchers in Perm have developed an alternative to GPS for monitoring public transport. A neural-network system based on YOLO recognises bus licence plates from roadside cameras with 82% accuracy, without requiring GPS trackers, and delivers real-time updates to passengers via a chatbot. The system functions even with weak mobile signals and remains reliable in poor weather conditions.
NtechLab has also rolled out a digital patrol system to tackle potholes. By using existing camera networks on buses, trolleybuses, and trams as a monitoring base, public transport has effectively been turned into mobile observation points. These systems automatically detect potholes, road damage, broken signs, non-functioning streetlights, and other defects in the urban environment.
In 2025, Yandex announced a landmark project: full integration of payment for all forms of urban transport – from buses to suburban trains – into the Yandex Go app. The core idea is simple: users should not need to choose a mode of transport. They need to get from point A to point B, and the system proposes the optimal combination – bus plus tram plus scooter. By 2026, the company planned to introduce unified travel passes and multi-modal subscriptions through a single application.
From the Capital to the Regions
For Russian cities and regions, these innovations amount to a literal reprogramming of daily life. Passengers no longer operate in uncertainty; the system provides them with actionable information for decision-making. For Russia as a whole, these projects strengthen the domestic digital ecosystem, expand national expertise in AI and video analytics, and reduce reliance on foreign technologies.
The export potential is significant. For developing countries, road-monitoring systems are attractive because they enable automated oversight without large-scale investment in new hardware. Neural networks for public transport management can be scaled to cities across Central Asia, the Caucasus, and the Middle East. Autonomous trams and trains being developed in Russia are of interest to both European and Asian markets. Yandex’s MaaS solution could serve as a reference model for the global mobility industry.
From Pilots to the Mainstream
By 2027, autonomous trams are expected to become standard in Moscow, with the first similar systems appearing in St Petersburg and Yekaterinburg. AI-driven traffic management systems will be deployed in all major Russian cities. During rush hour, nearly every driver will receive AI-generated recommendations: bypass the city centre, use a side street, or wait 30 minutes until congestion clears.
By 2030, urban traffic management is expected to be fully automated. Traffic lights will synchronise with one another in real time, creating green waves to maximise throughput. Public transport will operate with minute-level precision thanks to predictive route optimisation.
Mobility will become fully integrated. Yandex Go or a comparable application will propose optimal routes, process payments, provide delay alerts, and recommend the best time to depart to avoid being late. This would position Russia as a global centre for MaaS development.
Cameras on public transport will evolve beyond security tools into instruments of city management. They will monitor road conditions, detect emergencies, and assist in locating missing persons.
Safety on public transport will improve through continuous monitoring and rapid AI-driven response.
Russia is building the city of the future – a city that thinks, sees, and responds. A city where traffic lights do more than switch colours, actively managing flows for the benefit of every resident.
