|
Guest >> Sign In |
|
|
DIRECTED
ENERGY
PROFESSIONAL
SOCIETY
Abstract: 26-Symp-009
|
|
|
|
UNCLASSIFIED, PUBLIC RELEASE
Vision-based detection and tracking of sUAS with Event Vision Systems
US Army Space and Missile Defense Command (SMDC) and Radiance Technologies have developed a system which integrates the recent commercial-off-the-shelf (COTS) Sony IMX636 Event Vision Sensor (EVS) with a specialized algorithm to detect and track small unmanned aerial systems (sUAS) drones. By exploiting the high sensing rate of EVS compared to traditional visual sensors, the consistent high-frequency motion of individual sUAS propellers can be detected and identified as a distinguishing signature that does not present itself in natural environments. The team developed an algorithm which uses radar-like frequency analysis and decomposition to isolate the frequency content of target propellers from motion of the sensor’s own platform (in the demonstration case the sensor was mounted to a 10-inch quadcopter). The focus of this presentation will be the work performed in sUAS target detection and identification using novel EVS that can exploit prior knowledge of physical phenomena to build robust classification models with smaller datasets than common artificial intelligence and machine learning (AI/ML) models.
UNCLASSIFIED, PUBLIC RELEASE
|