Fast and Efficient Aerial Climbing of Vertical Surfaces Using Fixed-Wing UAVs

Abstract

We present improvements to Sherbrooke's multimodal autonomous drone (S-MAD), a microspine-based perching fixed-wing UAV that enables thrust-assisted climbing along vertical surfaces. Aircraft models are used to predict the performance of various aerial climb regimes and to design a controller for wall distance tracking. It is found that fast, long, and vertical climbs are favorable. Both short and long vertical autonomous climb maneuvers are demonstrated on rough surfaces (e.g., brick, roofing shingles). Results show that the S-MAD compares favorably with existing climbers, reaching a specific resistance of 19 with a much faster vertical speed (i.e., 2 m/s). A reduction in S-MAD's aerodynamic drag and an improved motor efficiency could bring its specific resistance down to 7, at a vertical speed of 5 m/s.