Researchers flew a DJI M600 UAV over a 500-metre trajectory in 75 seconds while receiving signals from nine Starlink satellites.
Signals transmitted by Starlink satellites, which have sought permission to operate in India, can be used to help an unmanned aerial vehicle determine its position, according to researchers at The Ohio State University, demonstrating a potential additional source of navigation information for drones when conventional satellite navigation is unavailable. The finding comes from an actual UAV flight test conducted in Columbus, Ohio, in August 2025. Researchers flew a DJI M600 UAV over a 500-metre trajectory in 75 seconds while receiving signals from nine Starlink satellites. The aircraft combined information extracted from the Starlink signals with data from its onboard Inertial Measurement Unit (IMU), which measures movement, to estimate its position. The researchers describe the work as the first experimental demonstration of UAV navigation using Starlink Low Earth Orbit (LEO) satellite signals as a “signal of opportunity”.
Starlink was designed as a communications system rather than a navigation system, but the researchers were able to extract information from its signals that could be used for positioning. During the flight, Global Navigation Satellite System (GNSS) signals were available for the first 20 seconds but were virtually cut off during the final 55 seconds. The UAV travelled about 370 metres during that period. The researchers used the Starlink signals together with inertial measurements to continue estimating the aircraft’s position. GNSS is the generic term for satellite-based positioning systems such as the US GPS and India’s NavIC. Their signals are used by aircraft, ships, vehicles, drones and other systems to determine their position. The significance of the experiment is therefore not that Starlink has become a replacement for GPS or India’s NavIC. Rather, it demonstrates that signals from a communications constellation can provide an additional source of positioning information for an autonomous system. The finding could also have implications in areas where access to conventional GNSS signals is sought to be restricted or disrupted by government agencies for security reasons.
If LEO communications signals can independently provide positioning information, a drone equipped to exploit such signals could potentially retain some navigation capability when conventional GNSS signals are unavailable. The Ohio State experiment does not establish that this can currently be done by unauthorised drone operators in India, or that Starlink can be used to circumvent a governmentimposed GNSS restriction. It demonstrates that communications signals from a LEO constellation can, under the tested conditions, contribute to a UAV’s positioning solution. The finding is relevant to the continuing problem of GNSS disruption. A separate 2026 study by researchers from the University of Texas at Austin and Stanford has documented the potential scale of GNSS interference. The study, ‘Chasing Lightning: Detecting, Characterising, and Identifying a Powerful Space-Based GNSS Interference Source’, examined powerful, transient interference events recorded across continental Europe, Greenland and Canada since 2019. Using data from a network of terrestrial GNSS reference stations, the researchers identified the source of the specific events studied as a constellation of Russian early-warning satellites operating in Molniya orbits. They found that the interference could affect receivers across wide areas. The two studies address different aspects of satellitebased positioning.
The Ohio State experiment demonstrates that signals from a communications constellation can contribute to navigation, while the Texas and Stanford research documents a source of interference affecting conventional GNSS signals. For drones, the potential significance is that navigation does not necessarily have to depend on a single source of positioning information. A system could combine satellite signals with inertial measurements and other sensors to estimate its position when one source becomes unavailable or unreliable. The Ohio State experiment, however, does not demonstrate an operational Starlink navigation service. It also does not show that Starlink can remotely control a drone. The researchers used specialised equipment and processing to extract navigation information from the satellite signals and combined that information with onboard sensors.
Nor does the experiment establish that a private individual in India can simply use Starlink to navigate a drone. The research demonstrates a technical method for extracting positioning information from Starlink signals. It does not constitute a Starlink-provided dronenavigation service. Earlier research has examined the use of Starlink and other LEO constellations for positioning, navigation and timing, including situations where GPS signals are weak or unavailable. Ohio State researchers have also reported experiments using Starlink and OneWeb signals for positioning in such environments. The latest UAV flight test takes that research to an airborne platform, demonstrating that signals from a large commercial communications constellation can be incorporated into the navigation solution of a moving unmanned aircraft. For India, the finding comes as the country develops its own satellite-navigation and space capabilities while commercial LEO satellite services expand. The research does not establish any current security threat from Starlink in India. It demonstrates instead that the signals emitted by commercial satellite constellations can have uses beyond the communications function for which they were designed.
