.png){kind=icon}
.png){kind=icon}
Return
From micro to macro: Characterizing heterogeneity of mixed-autonomy traffic flow
.png){kind=icon}
PRE
AI
H
DOI：10.1016/j.trc.2026.105926.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• The paper introduces a continuous-valued macroscopic flow attribute to characterize mixed-autonomy traffic heterogeneity. • Real trajectory data show that within-class differences can exceed average behavioral differences between AVs and HVs. • The flow attribute establishes an interpretable micro-macro link between driving behaviors and the fundamental diagram. • We develop two trajectory-based methods to reconstruct macroscopic flow attributes under abundant or sparse data. • Experiments show up to 20% lower validation error than the Aw-Rascle-Zhang benchmark and good generalization.
Keywords:
Mixed autonomy
Traffic heterogeneity
Autonomous vehicle
Traffic flow theory
Journal
IF:
7.9
Papers:
4.7K
Citations:
3.2W
Organization
No organization information available
Cited Papers