Published on: 2026-10-19
Source: Novosibirsk State University –
An important disclaimer is at the bottom of this article.
The Artificial Intelligence Research Center of Novosibirsk State University (NSU AI Center) has completed the development phase of algorithms for atmospheric air monitoring commissioned by CityAir LLC. The solutions improve the accuracy of sensor data processing and reduce the requirements for the computing equipment itself. The acquired experience lays the foundation for broader application of the center’s technologies in urban monitoring networks.
The results of the joint work were reported by the senior research fellow of the AI Center at NSU, Candidate of Technical Sciences Vladislav Alexandrovich Perepyolkin after the presentation of the atmospheric air monitoring station in the Oktyabrsky district of Novosibirsk. The station was installed on the territory of FSBEI HE “NGASU (Sibstrin)” and equipped with automated measuring equipment. The event was organized by the Ministry of Natural Resources and Ecology of the Novosibirsk Region.
— The work was accepted by the client, so this is a completed stage of the project. — noted Vladislav Alexandrovich.
The environmental monitoring system receives information about air temperature, its composition, wind speed, and direction. However, sensor readings require processing before they can be used to assess the state of the atmosphere. Measurement results are influenced by temperature, humidity, and changes in these parameters. Therefore, laboratory calibration does not always ensure the same accuracy of equipment in different seasons and climatic conditions.
An additional problem is the gradual change in sensor characteristics due to physical and chemical processes. Over time, the equipment may provide less accurate readings, which requires recalibration, maintenance, or replacement. The center’s specialists develop adaptive algorithms that take into account external conditions and help compensate for some of these changes.
— Artificial intelligence, by analyzing the history of observations, can partly compensate for these distortions and increase the accuracy of the final calculation result, — explained Vladislav Alexandrovich.
Another part of the task was to reduce the computational complexity of data processing. According to the researcher, classical methods are often narrowly specialized and require significant computational resources. The use of artificial intelligence should make the analysis more flexible and open up possibilities for creating compact hardware-software complexes that do not require powerful computers.
In the future, such devices can be installed on small sites, including on urban infrastructure supports. This will allow expanding the observation network without constructing a separate pavilion for each observation point.
For the NSU Artificial Intelligence Center, this project is important as a transition from scientific research to fulfilling the applied tasks of the client. The accumulated groundwork will allow adapting solutions for other monitoring networks, although each new project will require consideration of operating conditions and equipment characteristics.
— The Artificial Intelligence Center has gained relevant experience. Based on this foundation, similar solutions can be developed for new tasks, — Vladislav Alexandrovich emphasized.
Please note; This information is raw content obtained directly from the source. It represents an accurate account of what the source claims and does not necessarily reflect the position of MIL-OSI or its clients.