An Autonomous Drone-Based Framework for Real-Time Railway Monitoring Using YOLO-Based Defect Detection
Bibliographic record
Abstract
Railway infrastructure plays a critical role in transportation systems, and its routine inspection is crucial for ensuring operational stability and safety. Traditional railway inspection methods often rely heavily on fixed sensors and human monitoring, which are expensive to set up and time-consuming, respectively. This paper presents an autonomous drone-based railway monitoring system to detect structural defects and obstructions on railway tracks in real time. The flight stack comprises modern robotic frameworks such as PX4-Autopilot and ROS2. The sensor stack consists of an RGB camera for object detection and a depth camera for altitude estimation. Two parallel object detection pipelines, regular and oriented bounding box (OBB) YOLOv11 models, are fine-tuned to enhance detection accuracy under challenging visual conditions. Simulation results demonstrate the system’s effectiveness in detecting anomalies like sleeper misalignments and railway track obstructions. The system performance is tested with varying model sizes. The YOLOv11n model achieved an F1-score of 0.92 and an average latency of 59 ms per frame, providing a strong balance between accuracy and speed. Controller evaluations across speeds up to 1 m/s showed lateral and yaw RMSEs of 0.30 m and 2.01 deg, respectively, confirming stable and precise navigation. These findings highlight the potential of autonomous aerial systems to supplement or replace traditional railway inspection methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".