Multiple-input multiple-output radar for drone detection and localisation
Author
Term
3. semester
Education
Publication year
2025
Submitted on
2025-01-06
Pages
61
Abstract
The growing use of drones, particularly small and cost-effective suicide drones, in mod- ern warfare poses substantial threats to military assets. The purpose of this project is to contribute to further developing the design of the Archangel system: a drone de- tection and localisation system. This project applies a radar and a camera to acquire spherical measurements of a drone. The multiple-input multiple-output radar principle is applied to obtain azimuth angle measurements in addition to range and velocity mea- surements. Four different frameworks: angle FFT, Bartlett and Capon beamformers, and the MUSIC algorithm are investigated for direction-of-arrival estimation. Multiple data collections have been conducted in environmental conditions to evaluate the performance of the radar target detection. A data fusion scheme has been developed using a Kalman filter and spherical-to-Cartesian transformed radar and camera measurements to improve the drone location estimations. The Kalman filter effectively combines the multi-sensor measurements from the radar and camera, reducing measurement noise, and successfully models the linear movement of the drone. The method produces reliable estimations of the location and velocity of the drone. The fundamental advantages and limitations of the methods have been identified. This study further lays the groundwork for designing an effective drone detection and localisation system.
Keywords
Rdar ; MIMO radar ; Target detection ; CA-CFAR ; Data fusion ; Kalman filter ; Tracking
Documents