Visual-Assisted UAV Auto-Landing System

Abstract

Autolanding system in the unmanned aircraft system (UAS) is an important function to assist operators with more precise flight manipulation. Accompanying with GPS waypoint navigation, an ILS-like (Instrument Landing System) approach to landing is proposed based on visual-aid runway recognition and autopilot flight control. Simple mature and reliable detection algorithm is formulated to implement into Paparazzi autopilot system. In addition to the conventional RGB detection, the Hue-Saturation-Value (HSV) color space is introduced to give more precise result. The Line Segment Detector (LSD) is added to search for runway edge during approach to touch down at Waypoint 0. An approach procedure close to runway is established for UAV homing guidance and merged to a holding point before heading to runway. Similar to ILS glide slope, markers are created for UAV with a specific glide slope process to touchdown.