Omnisent

Drones have become one of the biggest threats in modern conflict and security. Most of them are cheap, small and especially hard to stop. The main problem with intrusive drones has spread from battlefields to European airports, military bases, and critical infrastructure.

Detection is currently a weak link. Radar is expensive and struggles with small, low-flying drones. Systems that listen for a drone's radio link miss the growing number that fly autonomously or on fiber. Active sensors can give away the defender’s own position and can be jammed as a result. There's real demand for detection that emits nothing, keeps working when GPS and radio are jammed and catches the small, low flying drones other sensors miss.

Omnisent, a Munich-based startup, founded in late 2024, builds passive acoustic drone detection for defense and homeland security. They make small, low-power nodes that listen for drones, paired with a model that detects, classifies, and locates them in real time. The nodes network into a mesh that covers an area and feeds a continuous airspace picture into existing command-and-control systems. The systems is built to be deployed fast by being placed by hand, drone, or vehicle in minutes and to run for over a year untouched, and to keep working in fog, rain, and electronic interference.

They have developed a, what they call Large Acoustic Model, which is a foundation model trained on non-speech sound. The LAM meant to separate a drone from everything else in a noisy environment such as airports.

One of the biggest risks at play is that sound carries a relatively short distance compared to radar, and acoustic performance falls off with wind, ambient noise, distance, and quieter or higher-flying drones.
The company was founded by Robin Daiber (CEO) and Adrien Jathe (CTO). Daiber has a master's in engineering from Cambridge. Jathe has a master's in mechanical engineering from ETH Zurich and a master's from Imperial. 

They have raised a $3M pre-seed led by Atlantic Labs.