Comparing the Crash Risk of Vehicle-Pedestrian Interactions using Autonomous Vehicle Data
Bibliographic record
Abstract
There is an increasing interest in autonomous vehicles (AVs) research as they are expected to provide considerable safety and mobility benefits. These vehicles should be able to interact with road users safely, which requires understanding the behavior of actual interactions between human-driven vehicles (HDV) and vulnerable road users (e.g., pedestrians). However, such behavior may vary considerably depending on the driving environment as culture plays an important role in traffic safety. This study uses an Extreme Value Theory Peak Over Threshold framework to estimate the risk of vehicle-pedestrian interactions in four different cities in the US and Asia (i.e., Boston, Las Vegas, Pittsburgh, and Singapore). A Bayesian hierarchical structure is considered to incorporate the effect of different covariates, which enables estimating the risk for each interaction. A large-scale AV dataset is used. As AVs are equipped with several sensors, they can capture information about the environment in real-time, including other road users’ positions and speeds. Results show that the risk varies significantly across different cities. For example, Pittsburgh has a greater risk than Singapore for regular vehicle-pedestrian interactions, which indicates that some cities require additional efforts for the implementation of AVs as the risk of interactions with pedestrians varies. Therefore, modeling frameworks that account for site-specific behavioral parameters should be proposed for the safe coexistence between advanced technologies and vulnerable road users.
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 | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".