Works in poor visibility conditions: a new system for recognizing UAVs by acoustic signals was created in Saratov

At the Yuri Gagarin State Technical University of Saratov (SSTU named after Yuri Gagarin), a system was created for recognizing types of unmanned aerial vehicles (UAVs) by their acoustic signals.

Image source Generated by the grok neural network

The system, developed by representatives of the university, is capable of overcoming factors that limit the recognition of unmanned aerial vehicles, effectively operating in conditions of urban noise and poor visibility, and adapting to new drone models.

Press service of SSTU named after Yuri Gagarin

During the development of the software, a comprehensive study was conducted, which made it possible to identify the key limitations of existing drone detection systems. As a result, a hybrid approach was proposed, combining methods of frequency-time analysis of sound with deep learning algorithms based on artificial intelligence.

The author of the project, Sergey Kuznetsov, noted that the expansion of the functionality of UAVs is associated with an increase in the risks of their unauthorized use.

Traditional identification methods - visual or radar detection - have limitations: they are ineffective in poor visibility conditions, require significant resources and do not always ensure stealth. Therefore, the development of alternative approaches, including the analysis of acoustic signals generated by UAVs, is becoming an urgent task.

Sergey Kuznetsov, author of the development

The developed system is integrated into the existing infrastructure and operates on standard equipment. This reduces the risks of unauthorized use of UAVs through early detection. Testing confirmed the effectiveness of the system in an urban environment, where it correctly identifies the acoustic patterns of drones against the background of noise.

Earlier www1.ru reported that the production of the Lider unmanned aerial system has been established in St. Petersburg.