Routes Will Feel Familiar
Yandex Maps and Yandex Navigator have introduced AI-powered personalised routing that factors in a driver’s travel history, habits and stated preferences. The update signals a deeper integration of machine learning into mass-market urban mobility services.
The Essence of the Update
Personalised routes are generated using a machine learning model that analyses path options from the routing engine alongside aggregated behavioural data within the app. The system evaluates how often a driver deviates from suggested routes, which roads they choose most frequently, how far they typically travel and at what times of day. Based on this dataset, the algorithm compares several possible routes and prioritises those most likely to match the individual user’s expectations.
Among the newly introduced route types is “Simple”, designed for less experienced drivers. It emphasises predictable trajectories, avoids complex interchanges and minimises left turns. For trips between familiar points – such as commuting from home to work – the system may suggest a “Familiar” route, even if faster alternatives are available. When travelling across the city, the navigator can also present two fundamentally different scenarios: a direct path through the centre for users willing to accept congestion, or a bypass option via less crowded roads.
Yandex Maps and Yandex Navigator currently support more than ten types of personalised routes, with further expansion planned. Ruslan Ledovsky, Head of Navigation at Yandex Maps, said: “We understand that travel time is not always the deciding factor for drivers. For those who have recently started driving, it is important to see a route with the fewest complex turns. And when commuting between home and work, drivers often prefer to follow a familiar path.”
Significance for the Smart Mobility Sector
The rollout of personalised routing marks a notable step in applying machine learning and AI within large-scale consumer services. It enhances user comfort and deepens service personalisation, strengthening Yandex’s competitive position in the domestic market while intensifying competition with global navigation platforms by accounting for local traffic patterns and driving behaviour. The update also aligns with the broader international trend of embedding AI in navigation and recommendation systems.
Within Russia, further development of AI models is expected. Algorithms are likely to offer a wider range of route scenarios with greater predictive accuracy. Personalisation could be integrated with adjacent services, including logistics optimisation, freight routing and public transport planning. There is also potential for API development for commercial partners such as logistics providers and transport management companies.
The technology carries export potential as well. The core personalisation engine may interest international partners, whether through service licensing or API integration. For now, however, developers remain focused on refining Yandex’s own ecosystem and adapting solutions to the needs of Russian users.
What Came Before and What Comes Next
The use of AI in navigation is not new. Modern systems already apply machine learning to predict traffic conditions, analyse congestion patterns and optimise routing strategies. Global mapping services such as Google Maps and Apple Maps offer multiple route options that consider traffic and certain user preferences. Yandex has previously introduced other AI-based features, including voice guidance integrated with Yandex GPT, weather forecasting and real-time traffic signal displays.
The current update represents a deeper layer of integration, embedding behavioural modelling directly into route selection logic. This is not merely a user interface enhancement but a structural shift in how machine learning operates within a widely used mobility platform.
Looking ahead, the range of personalised routes is likely to expand further, with tighter integration across digital services and urban mobility technologies. This could enhance usability and strengthen the market position of domestic digital platforms. At the same time, challenges remain, including the need to ensure robust personal data protection and to adapt regulatory frameworks to rapidly evolving AI capabilities.
