Autonomous Vehicle–Pedestrian Interaction Modeling Platform: A Case Study in Four Major Cities
Bibliographic record
Abstract
Accurately evaluating the safety effects of autonomous vehicles (AVs) has become more pressing with the increased adoption rate of AVs. This study utilizes a multiagent adversarial inverse reinforcement learning (MAAIRL) framework for modeling the interactions between AVs and pedestrians in four different cities: Boston, Las Vegas, Pittsburgh, and Singapore. Multiagent actor-critic with Kronecker factors deep reinforcement learning (MACK DRL), a paradigm that extends deep reinforcement learning (DRL), was used to model the behavior of both AVs and pedestrians and to determine their policies and collision avoidance strategies. Simulated trajectories are compared to actual trajectories and the results are evaluated to analyze the behavior of both AVs and pedestrians in terms of their evasive actions such as swerving, accelerating, or decelerating. The multiagent model provides a more comprehensive insight into how road users act in situations of conflict and accounts for changes in the environment. The study also shows that the level of competition between AVs and pedestrians varies significantly across different cities. Las Vegas has the most competitive relationship between AVs and pedestrians, while Singapore has the least competitive environment. The study also highlights the importance of cooperative behavior, particularly in yielding to pedestrians, in reducing the level of competition between AVs and pedestrians. In summary, this research provides valuable insights into the behavior of AVs and pedestrians and can be used to inform the development of more efficient and safe autonomous mobility systems.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| 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".