Multipath-Optimal UAV Trajectory Planning for Urban UAV Navigation with Cellular Signals

Abstract

Unmanned aerial vehicle (UAV) trajectory planning in urban environments is considered. Equipped with a three- dimensional (3-D) environment map, the UAV navigates by fusing global navigation satellite systems (GNSS) signals with ambient cellular signals of opportunity. A trajectory planning approach is developed to allow the UAV to reach a target location, while constraining its position uncertainty and multipath- induced biases in cellular pseudoranges to be below a desired threshold. Experimental results are presented demonstrating that following the proposed trajectory yields a reduction of 30.69% and 58.86% in the position root-mean squared error and the maximum position error, respectively, compared to following the shortest trajectory between the start and target locations.