HEP Statistical Inference for UAV Fault Detection: CLs, LRT, and SBI Applied to Blade Damage

TL;DR

This paper adapts advanced statistical methods from particle physics to UAV fault detection, providing a robust system that detects blade damage with high accuracy, controls false alarms, and quantifies fault severity with uncertainty estimates.

Contribution

It introduces the application of LRT, CLs, and SNPE methods to UAV fault detection, enabling improved detection, false alarm control, and fault severity quantification.

Findings

Achieved AUC of 0.862 on UAV-FD dataset, outperforming baseline methods.

Detected 93% of significant blade damage at 5% false alarm rate.

Provided calibrated posterior estimates of fault severity with high coverage.

Abstract

This paper transfers three statistical methods from particle physics to multirotor propeller fault detection: the likelihood ratio test (LRT) for binary detection, the CLs modified frequentist method for false alarm rate control, and sequential neural posterior estimation (SNPE) for quantitative fault characterization. Operating on spectral features tied to rotor harmonic physics, the system returns three outputs: binary detection, controlled false alarm rates, and calibrated posteriors over fault severity and motor location. On UAV-FD, a hexarotor dataset of 18 real flights with 5% and 10% blade damage, leave-one-flight-out cross-validation gives AUC 0.862 +/- 0.007 (95% CI: 0.849--0.876), outperforming CUSUM (0.708 +/- 0.010), autoencoder (0.753 +/- 0.009), and LSTM autoencoder (0.551). At 5% false alarm rate the system detects 93% of significant and 81% of subtle blade damage. On PADRE, a quadrotor platform, AUC reaches 0.986 after refitting only the generative models. SNPE gives a full posterior over fault severity (90% credible interval coverage 92--100%, MAE 0.012), so the output includes uncertainty rather than just a point estimate or fault flag. Per-flight sequential detection achieves 100% fault detection with 94% overall accuracy.