Toward Real-Time Intrusion Detection for Autonomous Vehicles: A Vision for Deep Learning-Based Security Frameworks
Bibliographic record
Abstract
Background: AI-driven autonomous vehicles (AVs) combine machine learning, control systems, and embedded technologies, creating significant software engineering challenges, especially in securing cyber-physical systems. Intrusion Detection Systems (IDS) are essential for detecting anomalies and cyberattacks in real time, thereby safeguarding AV operations. Aims: This vision paper aims to design and implement deep learning-based IDSs specifically tailored for real-time anomaly detection in autonomous drones, cars, and robots. Method: The approach begins with the development of platform-specific taxonomies of AV software vulnerabilities and a formal threat model. Controlled experiments will then be conducted using both physical aerial and ground vehicles, as well as industrial-grade simulators, in network-connected and isolated environments to generate labeled datasets of AV anomalies. These datasets will be used to design, train, and customize IDSs for each AV platform, which will subsequently be deployed on physical devices for evaluation under realistic conditions. Results: The work is expected to produce comprehensive taxonomies of AV software vulnerabilities, multiple labeled datasets capturing both normal and compromised AV behaviors, and a set of validated, deployable IDS frameworks tailored to various AV platforms. Conclusions: This study addresses core empirical software engineering challenges, such as the sim-to-real transfer gap in machine learning, risks of overfitting in data-driven IDS models, and hardware-software integration complexities. The anticipated outcome is a robust family of IDS solutions that enhance AV security in dynamic operational environments.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".