Performance Evaluation of Vision-Based Algorithms for MAVs

TL;DR

This paper compares various real-time vision-based localization and mapping algorithms for Micro Aerial Vehicles operating indoors without GPS, highlighting their strengths, weaknesses, and integration into navigation systems.

Contribution

It provides a comprehensive evaluation of existing vision-based algorithms and discusses their application in MAV localization, mapping, and navigation in GPS-denied environments.

Findings

Identification of the most reliable algorithms for indoor MAV localization.

Analysis of the computational efficiency of different methods.

Insights into the integration of vision-based algorithms with control systems.

Abstract

An important focus of current research in the field of Micro Aerial Vehicles (MAVs) is to increase the safety of their operation in general unstructured environments. Especially indoors, where GPS cannot be used for localization, reliable algorithms for localization and mapping of the environment are necessary in order to keep an MAV airborne safely. In this paper, we compare vision-based real-time capable methods for localization and mapping and point out their strengths and weaknesses. Additionally, we describe algorithms for state estimation, control and navigation, which use the localization and mapping results of our vision-based algorithms as input.