Comparison of Different Controller Architectures for Autonomous Driving and Recommendations for Robust and Safe Implementations
Bibliographic record
Abstract
This comprehensive review examines various controller architectures for autonomous driving systems, from rule‐based approaches to advanced deep learning methods. Research trends reveal a significant shift toward deep learning approaches (65.6%) compared to rule‐based methods (34.4%), reflecting the growing dominance of data‐driven techniques in autonomous vehicle research. Performance analysis of transformer‐based models demonstrates exceptional accuracy, with ViT‐SAC achieving 100% success rate in low‐density traffic scenarios and DRLNDT reaching 99.9% success rate in navigation tasks. Temporal reasoning capabilities assessment shows BEVWorld excelling in context maintenance and historical data integration (both 95/100), while Holistic Transformer demonstrates superior noise robustness (95/100). Computational efficiency varies significantly, with VCNN (38.50 FPS) and DSCNN Transformer (34.07 FPS) exceeding real‐time thresholds, while complex BEV architectures like BEVSegformer (3.97 FPS) require further optimization. Simulation platform comparison identifies CARLA as the most comprehensive environment, supporting five of seven key testing features, though no single platform provides complete coverage of all requirements. Technical challenges assessment quantifies real‐time processing requirements as the most critical challenge (90/100), followed by generalization limitations (85/100). These suggest that while rule‐based approaches offer computational efficiency and interpretability, deep learning methods demonstrate superior perception and decision‐making capabilities. A balanced combination of learning‐based, rule‐based, and simulation‐based validation approaches, with particular emphasis on addressing real‐time performance and generalization capabilities, will likely be necessary to achieve reliable autonomous driving systems capable of navigating complex and dynamic 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 | 0.000 | 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 | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".