A Novel Data Fusion Algorithm to Improve the Detection and Tracking of “Killer” Drones in Urban Environment

A data fusion algorithm that incorporates joined probability between observed raw data from multiple sensors is described. Remote sensors of various kinds transmit position information to a central station where data are cinematically fused to create a composite measure. It follows that the problem is to find spatial coordinates of that point where the likelihood of finding a target is maximum in that instant. If each raw data has its own associated probability, it’s possible to merge all of the sensor’s likelihood to obtain the joined probability. Because raw sensor measurements can be considered independent, the target spatial coordi- nates with maximum likelihood are the most frequent value of joined probability distribution. This method significantly enhances tracking accuracy, providing superior target position estimates compared to single-sensor approaches. This methodology is particularly relevant for anti-drone systems in urban scenarios aimed at protecting critical infrastructures. A method to estimate the standard deviation of each sensor’s associated probability is also included.