Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.21311 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910241174585344 |
|---|---|
| author | Poudel, Bibek Zhu, Lei Heaslip, Kevin Swaminathan, Sai Li, Weizi |
| author_facet | Poudel, Bibek Zhu, Lei Heaslip, Kevin Swaminathan, Sai Li, Weizi |
| contents | Modern vision systems can detect, track, and forecast urban actors at scale, yet translating perception outputs to urban design remains limited. We introduce DeCoR, a two-stage reinforcement learning framework that leverages flow observations to co-optimize crosswalk layout and network-level signal control. The design stage encodes the pedestrian network as a graph and learns a generative policy that parameterizes a Gaussian mixture model over crosswalk location and width, from which new crosswalks are sampled. For each layout, a shared control policy learns adaptive signal timings to minimize joint pedestrian and vehicle delay. On a 750 m real-world urban corridor with demand sensed from video and Wi-Fi logs, DeCoR learns a layout that reduces pedestrian arrival time to their nearest crosswalk by 23% while using fewer crosswalks than existing configurations. On the control side, DeCoR reduces pedestrian and vehicle wait time by 79% and 65%, respectively, relative to fixed-time signalization. Further, the control policy generalizes to demands outside of training and is robust to layout changes without retraining. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_21311 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DeCoR: Design and Control Co-Optimization for Urban Streets Using Reinforcement Learning Poudel, Bibek Zhu, Lei Heaslip, Kevin Swaminathan, Sai Li, Weizi Machine Learning Artificial Intelligence Modern vision systems can detect, track, and forecast urban actors at scale, yet translating perception outputs to urban design remains limited. We introduce DeCoR, a two-stage reinforcement learning framework that leverages flow observations to co-optimize crosswalk layout and network-level signal control. The design stage encodes the pedestrian network as a graph and learns a generative policy that parameterizes a Gaussian mixture model over crosswalk location and width, from which new crosswalks are sampled. For each layout, a shared control policy learns adaptive signal timings to minimize joint pedestrian and vehicle delay. On a 750 m real-world urban corridor with demand sensed from video and Wi-Fi logs, DeCoR learns a layout that reduces pedestrian arrival time to their nearest crosswalk by 23% while using fewer crosswalks than existing configurations. On the control side, DeCoR reduces pedestrian and vehicle wait time by 79% and 65%, respectively, relative to fixed-time signalization. Further, the control policy generalizes to demands outside of training and is robust to layout changes without retraining. |
| title | DeCoR: Design and Control Co-Optimization for Urban Streets Using Reinforcement Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.21311 |