The paper presents a new method for detecting and assessing potential landing sites for safe autonomous quadrotor landing. The solution integrates 2D and 3D environmental information, eliminating the need for GPS and computationally intensive elevation maps. The proposed pipeline combines semantic data from a Neural Network to extract environmental features, with geometric data from a disparity map, to extract key geometric attributes such as slope, flatness, and roughness. The method can effectively assess and identify suitable landing areas, enabling the safe and autonomous landing of a quadrotor.
Publication date: 20 Nov 2023
Project Page: https://www.youtube.com/watch?v=3tH621vF8LM
Paper: https://arxiv.org/pdf/2311.10065
