Fine-Grained Discrimination of Low-Altitude UAV Targets Based on Radar and Remote ID

With the rapid growth of the low-altitude economy, low-altitude Unmanned Aerial Vehicles (UAVs) applications have become widespread. Meanwhile, more and more illegal use of UAVs also poses threats to societal safety. Consequently, the development of robust UAVs surveillance is imperative. While Remote ID (RID) provides identity and position for cooperative UAVs mandatorily mounted with RID modules, it cannot detect non-cooperative targets. In contrast, radar can detect moving targets, but cannot distinguish their cooperative status. Mapping RID broadcasts to radar tracks could enable the fine-grained discrimination of low-altitude UAVs (distinguishing between cooperative and non-cooperative UAVs, and further distinguishing among cooperative UAVs), leading to more efficient surveillance, which however remains seldom studied so far. In this paper, we study how to achieve fine-grained discrimination of low-altitude UAVs based on joint radar and RID. Specifically, we map RID broadcasts and radar tracks into a common coordinate system. Then mapping problem is solved via the global nearest neighbor approach. It can be modeled as a constrained bipartite minimum-weight matching problem. We solve it using the Hungarian algorithm. Real-world experimental results demonstrate that, compared to three baselines, our method effectively suppresses interference from cooperative targets during mapping, with microsecond-scale latency that meets the demands of real-time UAV surveillance.