Under complex low-altitude conditions, non-stationary signals and positioning fluctuations cause multi-rotor unmanned aerial vehicles to have fragmented and blurred low-level methods that are difficult to adapt to dynamic scenes and cannot effectively determine the intersection points of geofencing boundaries. This study proposes a monitoring method based on ultra-wideband positioning: tagging unmanned aerial vehicles to collect location data, constructing spatial relationships using an enhanced time difference of arrival algorithm, linearizing equations, and using the Jacobi iterative method for positioning; introducing adaptive Kalman filtering for real-time error correction; and detecting violations using the ray casting method. The experiment shows that this method has a horizontal positioning error of ≤3.2 cm and a vertical positioning error of 2.5 cm in dense cities, with a detection delay of 0.2–0.45 s, and can quickly and accurately monitor violations.
Ning Zhang (Mon,) studied this question.