Utvärdering av Radarbrustolerans i Autonoma Drönaravskärningssystem
Erik Werkström · Diva portal (Dalarna University Library) · 2026
The thesis addresses how radar noise affects the ability of an autonomous unmannedaerial vehicle interception system to reliably complete intercepts. This issue is relevant due to the recent advent of drone warfare in the Russo-Ukrainian war and thelacking technology that exists to reliably defend against drone attacks without human intervention. Using a digital experimental environment, interception scenarioswere simulated and the percent of successful interception, average interception timeand average miss distance of failed interceptions were calculated. The maximum andminimum values, as well as standard deviations were also calculated for the interception time and miss distance metrics. The environment utilized radar noise levelsincreasing in severity, and a baseline with no noise, to evaluate how the metrics wereaffected by introducing more noise into the simulation. The extended Kalman filterwas also used to mitigate the effects of the noise. The resulting metrics show thatwhile low amounts of noise do not cause the interception to fail, higher levels do, andthat even low noise increases the average interception time. They also show that asactual noise strays further from the noise model, tracking performance decreases.