Can a drone “hear” another drone in flight despite the deafening noise of its own propellers? The innovative technological initiative led by Dr. Maciej Podsędkowski is developing a “flying ear” technology for unmanned aerial vehicles.
At the Institute of Fluid-Flow Machinery, one of the most ambitious initiatives in European aerial robotics has begun. Its goal is to solve one of the most challenging problems in modern airborne robotics: enabling drones to perform effective acoustic detection of other flying objects during flight, despite the intense noise generated by their own propulsion systems. The project is led by Dr. Maciej Podsędkowski.
The key challenge in acoustic detection is the so‑called ego‑noise — the dominant noise produced by the rotors and air turbulence, which masks everything happening in the surrounding environment. The SMART project aims to overcome this barrier through hybrid filtering methods based on engine telemetry data (RPM‑based filtering), combined with advanced beamforming using null‑steering techniques and direction‑of‑arrival estimation algorithms such as MVDR and GEVD‑MUSIC.
The project is carried out across three environments of increasing complexity: an anechoic chamber, an aerodynamic wind tunnel, and an open‑air test range. Additional experiments using a glider in unpowered flight make it possible to separate aerodynamic noise from mechanical noise. The project also employs state‑of‑the‑art signal‑processing techniques such as MVDR with null‑steering (directing the pattern’s “nulls” toward the drone’s own propellers to suppress them), GEVD‑MUSIC (precise direction‑of‑arrival estimation, DoA), and noise models based on empirical data from tunnel tests and flight experiments. This approach enables the creation of a system that not only filters noise but actively “understands” the drone’s acoustic environment.
The project fills an important research gap, as most previous studies have focused on simulations or drones in hover. Here, for the first time, the influence of forward flight on microphone array performance is being analyzed — a factor of great importance for vibroacoustics and the design of quieter, more situationally aware robots. The project has significant potential in the fields of security and defense technologies, as well as in the protection of critical infrastructure. The developed system may become the foundation for mobile monitoring platforms capable of detecting unauthorized drones in challenging terrain.
The SMART project aims to reach a technology readiness level (TRL) that will enable the construction of a system demonstrator and applications for further R&D funding, both in national (NCBR) and international programs. It is a step toward a new generation of autonomous drones that can not only see but also hear and respond to their surroundings. The Institute of Fluid‑Flow Machinery proudly supports the work of Dr. Maciej Podsędkowski and his team, confident that the SMART project will pave the way for further breakthroughs in aerial robotics.