Maintaining UAV Stability using Low-Power WANs

Future urban spaces are expected to see Unmanned Aerial Vehicles (UAVs) deployed for wide-ranging applications including product delivery, imaging and search-and-rescue. Yet today's UAVs rely critically on inertial sensors and GPS to remain stable in-flight, meaning they struggle to penetrate urban canyons where GPS is unavailable. This poster presents LoRaTilt, a system that maintains stability of a UAV using a Low-Power Wide-Area Network (LP-WAN) transmitter. We mount a light-weight and ten-year battery-powered LP-WAN transmitter on a UAV and process its signals from base stations up to hundreds of meter away. We demonstrate how our system estimates and corrects for UAV drift in GPS-denied settings with minimal impact on the cost and flight life of a UAV. Our proof-of-concept experiments show a promising median accuracy of 1.2326 mm and 0.0164 radians in estimating the displacement and orientation of a drone with a LoRa LP-WAN radio.