Methods for Determining Threat Level and Intent of Unmanned Aerial Systems

Unmanned Aerial Systems (UASs) poses a growing threat to our national security. Commercially available UASs can now carry explosives, and perform electronic surveillance missions with an increasing range against our military, cities, infrastructure, population and information. Swarm attacks are possible by automated drones and dirty bombs can be dropped in populated areas.Due to the ease of small UAS acquisition and modifications, it is inevitable that future attacks will be carried out by small UASs. A low SWAP-C system to search, detect, identity, track and determine intend (SDITI) of this growing threat is now necessary. We propose a low SWAP-C, man portable fusion approach leveraging the benefits of passive radar, image processing, particle filter tracking and deep learning to maximize SDITI performance and provide a comprehensive solution. In addition, multi-spectral analysis techniques will be incorporated to provide radio frequency characteristics of the UAS and controller. The fusion solution can then be used for intent and threat level assessment and to craft an effective countermeasure. We have successfully applied the above techniques individually for detection, tracking and classification on other programs and are confident that a fusion approach will provide the best low SWAP-C solution to performing SDITI against hostile UASs.