Image Segmentation to Identify Safe Landing Zones for Unmanned Aerial Vehicles

TL;DR

This paper explores using lightweight semantic image segmentation models on single RGB images to efficiently identify safe landing zones for UAVs, reducing energy use and sensor requirements.

Contribution

It introduces a lightweight semantic segmentation approach for UAV landing zone detection using only RGB images, simplifying sensor setup and energy consumption.

Findings

Lightweight models achieve high accuracy in identifying safe landing zones.

Single RGB camera images are sufficient for effective segmentation.

Energy consumption is reduced by working with images instead of video.

Abstract

There is a marked increase in delivery services in urban areas, and with Jeff Bezos claiming that 86% of the orders that Amazon ships weigh less than 5 lbs, the time is ripe for investigation into economical methods of automating the final stage of the delivery process. With the advent of semi-autonomous drone delivery services, such as Irish startup `Manna', and Malta's `Skymax', the final step of the delivery journey remains the most difficult to automate. This paper investigates the use of simple images captured by a single RGB camera on a UAV to distinguish between safe and unsafe landing zones. We investigate semantic image segmentation frameworks as a way to identify safe landing zones and demonstrate the accuracy of lightweight models that minimise the number of sensors needed. By working with images rather than video we reduce the amount of energy needed to identify safe landing zones for a drone, without the need for human intervention.