have turned to neural networks and machine learning to continually improve drone identification using the acoustic signature of different types of drones. Both of these solutions also represent a step change in costs from earlier counter drone systems.
Detection and identification are important, but to really be effective, a system also has to be able to take out a malicious drone. Drone mitigations can be kinetic, like nets or projectiles, or non-kinetic, like jamming or radio frequency signal substitution. Kinetic solutions are important, particularly in military settings, but in environments like airports, having drone pieces scattered about actually creates new problems.
Despite the engineering challenges inherent in drone detection and mitigation, the biggest obstacle is not technical at all. It’s legal. Right now, public safety agencies face a confusing landscape concerning what they are permitted to do to identify and stop drones.
A few drones are helpful. Thousands are a nuisance and dangerous.