Informative Path Planning for Active Learning in Aerial Semantic Mapping

TL;DR

This paper introduces an autonomous UAV-based system that actively plans data collection routes to efficiently improve semantic segmentation models for aerial mapping, reducing labeling effort and enhancing model accuracy.

Contribution

It presents a novel active learning framework that links model uncertainty to UAV path planning for improved aerial semantic mapping.

Findings

Active learning improves segmentation accuracy with less labeled data.

UAV planning based on uncertainty outperforms static coverage paths.

The approach reduces labeling effort while maintaining high model performance.

Abstract

Semantic segmentation of aerial imagery is an important tool for mapping and earth observation. However, supervised deep learning models for segmentation rely on large amounts of high-quality labelled data, which is labour-intensive and time-consuming to generate. To address this, we propose a new approach for using unmanned aerial vehicles (UAVs) to autonomously collect useful data for model training. We exploit a Bayesian approach to estimate model uncertainty in semantic segmentation. During a mission, the semantic predictions and model uncertainty are used as input for terrain mapping. A key aspect of our pipeline is to link the mapped model uncertainty to a robotic planning objective based on active learning. This enables us to adaptively guide a UAV to gather the most informative terrain images to be labelled by a human for model training. Our experimental evaluation on real-world data shows the benefit of using our informative planning approach in comparison to static coverage paths in terms of maximising model performance and reducing labelling efforts.