Comparative Study of Machine Learning Approaches for Fixed Radar-Based Road User Classification
Bibliographic record
Abstract
This study investigates radar-based road user classification using a fixed infrastructure setup. It evaluates two machine learning (ML) strategies: (1) deploying separate classifiers for single-point and multi-point radar detections and (2) utilizing a unified classifier for both object types. The study uses four supervised learning models: Random Forest (RF), Extreme Gradient Boosting (XGB), Multi-Layer Perceptron (MLP), and Support Vector Machine (SVM), which are trained on statistical features extracted from annotated 3D radar point clouds in the INFRA-3DRC dataset. Feature extraction procedures are adapted to address the structural disparity between single- and multi-point detections, with feature padding applied to ensure compatibility with the unified model. Experimental results show that XGB and RF consistently achieve the highest F1-scores, exceeding 97%, under both modeling approaches. The unified classifier offers deployment simplicity and real-time readiness, while the separate strategy provides lower prediction latency for time-sensitive applications. Overall, the findings confirm the viability of scalable, radar-only traffic monitoring systems and underscore the trade-offs between accuracy, computational cost, and architectural complexity. Future work will explore temporal modeling using recurrent neural networks to capture object dynamics across frames.
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 | 0.005 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".