Enhancing Drone Localization with Joint UWB and Radar Measurements Using CIR Processing

Very often, applications such as drone inspection use Ultra-Wideband (UWB) as a positioning method in areas where Global Navigation Satellite System (GNSS) is unavailable or degraded. Additionally, they sometimes employ a radar system to detect obstacles or monitor the distance to the object being inspected. This study proposes an approach where radar measurement is conducted by utilizing the Channel Impulse Response (CIR) from cost-effective, off-the-shelf UWB devices to eliminate the need for a dedicated radar system. Unlike similar approaches, the radar measurement here is performed from a mobile object, significantly complicating the processing of background clutter. In this method, certain radar echoes are identified and associated with obstacles of known positions, then fused with UWB distances measurements. Field tests have shown that our approach provides a radar distance measurement accuracy of 6.5 cm and reduces horizontal and vertical positioning errors from 138 cm and 287 cm, respectively, when using UWB alone, to 96 cm and 68 cm by combining it with radar measurements.