What question did this study set out to answer?

This research aims to enhance safety in UAV object detection by addressing predictive uncertainties.

April 30, 2026Open Access

Enhancing the Safety of UAV Object Detection: A Novel OOD Algorithm for Filtering Prediction Uncertainties in Low-Altitude Airspace

Key Points

This research aims to enhance safety in UAV object detection by addressing predictive uncertainties.
Developed an anomaly-scoring algorithm for OOD evaluation called PCA-HBOS.
Utilized high-dimensional semantic features from deep networks to score sample distributions.
Evaluated the algorithm against mainstream OOD algorithms in low-altitude scenarios.
PCA-HBOS effectively identifies in-distribution and out-of-distribution samples.
Comparative analysis demonstrated PCA-HBOS's superiority in low-altitude scenarios.

Abstract

With the rapid development of AI technology in the low-altitude field, the predictive uncertainty of ML models, particularly object-detection models, has become a key bottleneck restricting their airworthiness and safe deployment. A typical manifestation of this is that when facing test samples in open-world scenarios, object-detection models exhibit overconfidence in erroneous predictions. To address this issue, this paper proposes an anomaly-scoring algorithm for out-of-distribution (OOD) evaluation based on augmented deep network features, named PCA-HBOS. By integrating the high-dimensional semantic features extracted by deep networks, the algorithm can score sample distributions, thereby enabling the identification of both in-distribution and out-of-distribution samples. Through comparisons with mainstream OOD algorithms, the superiority of the PCA-HBOS in low-altitude scenarios is validated. Experimental results on three multi-sensor

Enhancing the Safety of UAV Object Detection: A Novel OOD Algorithm for Filtering Prediction Uncertainties in Low-Altitude Airspace

Key Points

Abstract

Cite This Study