Robust Vision-Based Autonomous Navigation, Mapping and Landing for MAVs at Night

Abstract

This paper is about vision-based autonomous flight of MAVs at night. Despite it being dark almost half of the time, most of the work to date has addressed only daytime operations. Enabling autonomous night-time operation of MAVs with low SWaP on-board sensing capabilities is still an open problem in current robotics research. In this paper, we take a step in this direction and introduce a robust vision-based perception system using thermal-infrared cameras. We present this in the context of safe autonomous landing on rooftop-like structures, and demonstrate the efficacy of our proposed system through extensive real-world flight experiments in outdoor environments at night.