Efficient Global Indoor Localization for Micro Aerial Vehicles

TL;DR

This paper introduces a real-time, vision-based indoor localization method for micro aerial vehicles that uses textons and a k-NN algorithm, achieving about 0.6m accuracy in a 5m x 5m area.

Contribution

The paper presents a novel onboard computer vision approach combining textons, k-NN, and particle filtering for accurate, real-time indoor MAV localization without error accumulation.

Findings

Localization accuracy of approximately 0.6 meters

Runtime of 32 milliseconds onboard MAV

Scalable computational effort based on sampling

Abstract

Indoor localization for autonomous micro aerial vehicles (MAVs) requires specific localization techniques, since the Global Positioning System (GPS) is usually not available. We present an efficient onboard computer vision approach that estimates 2D positions of an MAV in real-time. This global localization system does not suffer from error accumulation over time and uses a $k$-Nearest Neighbors ($k$-NN) algorithm to predict positions based on textons---small characteristic image patches that capture the texture of an environment. A particle filter aggregates the estimates and resolves positional ambiguities. To predict the performance of the approach in a given setting, we developed an evaluation technique that compares environments and identifies critical areas within them. We conducted flight tests to demonstrate the applicability of our approach. The algorithm has a localization accuracy of approximately 0.6 m on a 5 m$\times$5 m area at a runtime of 32 ms on board of an MAV. Based on random sampling, its computational effort is scalable to different platforms, trading off speed and accuracy.