Enhancing perceived risk prediction of human-vehicle collisions in urban and construction environments by incorporating motion dynamics and behavior-based features
Bibliographic record
Abstract
Human-vehicle collisions pose risks in urban and construction environments, necessitating proactive assessment. Most prior studies rely on proximity and time-to-collision (TTC), overlooking behavior-driven factors such as sudden motion changes and instability. Unlike human factors that emphasize key cognitive states, behavior-driven factors cannot directly capture them but infer them through external movement patterns associated with perceived risk. This study proposes a machine learning framework integrating motion-based and behavior-driven features to predict human-perceived collision risk. Perceived risk from 21 participants were collected in virtual urban, factory, and construction settings. Features including proximity, TTC, velocity, velocity change, and entropy were extracted across multiple time windows (0.5–5.0s prior to perceived risk) and analyzed to identify optimal prediction frames. LightGBM achieved 98.00% accuracy for detectingcollision risk presence, while random forest yielded 80.13% for binary risk levels. Performance improved when incorporating features 2–3s before risk, underscoring the predictive value of early behavioral signals. SHAP analysis confirmed proximity and TTC as dominant predictors, but behavior-driven features amplified risk signals in borderline cases. This framework bridges objective risk indicators with human perception, supporting AI-driven safety systems through warnings and intervention timing in autonomous driving, construction safety, and pedestrian risk management, contributing to proactive risk prevention.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".