Everything K-culture — comebacks to K-beauty, straight to your inboxGet it in your inbox›
A Metamorphic Artificial Age Score Decision-Support Prototype for Flight-Log-Based Drone Propeller Health Monitoring
arXiv:2608.180882026-08-20
A prototype scores drone propeller health by tracking how fault signs spread across different flight-log channels, not just one signal
This study builds a decision-support prototype that pulls six health indicators, such as trajectory error and motor-command imbalance, from real drone flight logs. It tests multiple scoring policies with metamorphic testing to see which one behaves consistently, then ranks them using a redundancy-adjusted Artificial Age Score (AAS). Using one healthy and three faulty propeller flight logs from the DronePropA dataset, different fault severities were found to be driven by different indicators.
METAL MEDIA explanatory visual
A prototype scores drone propeller health by tracking how fault signs spread across different flight-log channels, not just one signal
01Six indicators, trajectory tracking error, attitude instability, thrust-command burden, motor-command imbalance, ESC-command instability, and battery-level stress, were computed from real .mat flight-log files in the 2024 DronePropA public dataset
02Values were normalized against a healthy baseline flight into a 0-to-1 scale, and three candidate scoring policies (a simple weighted sum, a capped version, and a threshold-triggered version) were used to generate burden scores
03Six metamorphic relations (checks on whether scores change as expected when inputs are meaningfully altered) were applied to test each policy's consistency, and the results were converted into a redundancy-adjusted Artificial Age Score (AAS), a measure of structural inconsistency rather than chronological age, to select the most adequate policy
04The healthy flight log was assigned to routine monitoring; the Severity 1 fault, driven mainly by ESC-command instability, was assigned to maintenance review; Severity 2 (peak motor- and ESC-command burden) and Severity 3 (peak trajectory tracking error) both triggered mandatory inspection
05The results confirm that propeller fault effects surface through different channels depending on severity, supporting the case for a multi-indicator decision-support layer rather than a single diagnostic signal
An explanatory diagram made by METAL MEDIA, not a figure supplied by the paper's authors.
What they did
Six indicators, trajectory tracking error, attitude instability, thrust-command burden, motor-command imbalance, ESC-command instability, and battery-level stress, were computed from real .mat flight-log files in the 2024 DronePropA public dataset
Values were normalized against a healthy baseline flight into a 0-to-1 scale, and three candidate scoring policies (a simple weighted sum, a capped version, and a threshold-triggered version) were used to generate burden scores
Six metamorphic relations (checks on whether scores change as expected when inputs are meaningfully altered) were applied to test each policy's consistency, and the results were converted into a redundancy-adjusted Artificial Age Score (AAS), a measure of structural inconsistency rather than chronological age, to select the most adequate policy
The healthy flight log was assigned to routine monitoring; the Severity 1 fault, driven mainly by ESC-command instability, was assigned to maintenance review; Severity 2 (peak motor- and ESC-command burden) and Severity 3 (peak trajectory tracking error) both triggered mandatory inspection
The results confirm that propeller fault effects surface through different channels depending on severity, supporting the case for a multi-indicator decision-support layer rather than a single diagnostic signal
Table 1: Signal-to-feature mapping used for DronePropA feature extraction.
Feature
Source matrix
Rows used
Computation
Trajectory tracking error
commander_data
22–24 and 26–28
RMS of Euclidean error between measured and reference position
Attitude instability
QDrone_data
5–7
Mean standard deviation and mean RMS temporal difference of roll, pitch, and yaw-rate signals
Thrust-command burden
commander_data
34
RMS of temporal difference in reference thrust
Motor-command imbalance
QDrone_data
47, 49, 51, 53
RMS of instantaneous standard deviation across four motor-command channels
ESC-command instability
QDrone_data
48, 50, 52, 54
Average of ESC imbalance RMS and ESC temporal-difference RMS
Battery-level stress
QDrone_data
24
Average of voltage drop and battery-level standard deviation
It shows with real flight data that propeller faults don't always show up the same way, giving maintenance teams an explainable basis for why a specific case is flagged. That said, it remains a small-scale proof-of-concept prototype, not a certified diagnostic tool, so broader validation is needed before operational use.
Terms in this paper
ESC (electronic speed controller) · a drone component that controls motor rotation speed
metamorphic testing · a method for checking system consistency by seeing if outputs change as expected when inputs are transformed in defined ways, useful when a definite correct answer is hard to specify
Artificial Age Score (AAS) · a metric that measures structural inconsistency of a system or policy, not actual elapsed time or aging
DronePropA · a 2024 public dataset of real flight logs from drones with healthy and defective propellers
normalization · rescaling different values onto a common range, such as 0 to 1, so they can be compared
Original abstract (English)
Drone propeller faults can create safety and reliability risks when their effects are distributed across multiple flight-log channels rather than appearing as a single diagnostic signal. This paper proposes a Metamorphic Artificial Age Score (AAS) decision-support prototype for flight-log-based drone propeller health monitoring. Using selected historical real flight logs from the 2024 DronePropA public dataset, the framework computes six health-related indicators from raw MATLAB matrices: trajectory tracking error, attitude instability, thrust-command burden, motor-command imbalance, ESC-command instability, and battery-level stress. These indicators are normalized relative to a healthy baseline and evaluated through candidate scoring policies, metamorphic adequacy relations, and a redundancy-adjusted AAS formulation. In this context, AAS is used as a structural policy-adequacy and burden measure rather than as a chronological age measure. A controlled retrospective evaluation was performed using one healthy baseline and three defective propeller cases under the same speed profile and trajectory. The healthy case was assigned to routine monitoring. The Severity 1 case was dominated by ESC-command instability and assigned to maintenance review. The Severity 2 case reached maximum motor-command and ESC-command burden, while the Severity 3 case reached maximum trajectory tracking error; both triggered mandatory inspection. The results show that propeller fault effects may appear through different operational channels, supporting the need for a multi-indicator decision-support layer for post-flight maintenance prioritization and autonomous-system oversight.
