Propeller Fault Classification for Unmanned Aerial Vehicles and Explainable Artificial Intelligence-Based Feature–Model Matching
Ahmet Çağdaş Seçkin · 2026
Read Propeller Fault Classification for Unmanned Aerial Vehicles and Explainable Artificial Intelligence-Based Feature–Model Matching by Ahmet Çağdaş Seçkin (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13611206.txt.
Description
Ahmet Çağdaş Seçkin built a careful testing framework for spotting faulty drone propellers from sound and motion data in 2026. On a test bench with two motor types, he recorded one healthy and eight faulty propeller states and cut the data into 8,490 one-second windows.
Feature sets and learners were matched without data leakage by splitting at the file level. A stacked ensemble trained on an explainability-ranked subset of 52 features reached a macro-F1 of 0.8027 and accuracy of 0.8816. Time-domain statistics and accelerometer axes mattered most. Models also depended strongly on motor type, a warning for real deployments. The framework doubles as an evaluation protocol for cheap sensor setups.
Catalog record
| CREATOR | Ahmet Çağdaş Seçkin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 89 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |