MPCFormer explicitly models multi-vehicle social interaction dynamics via physics-informed discrete state-space and Transformer-learned coefficients, yielding 0.86m ADE over 5s and 94.67% planning success with near-zero collisions in closed-loop tests.
General Lane-Changing Model MOBIL for Car -Following Models
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
AACC combines online IOC for driving style identification with a Stackelberg game planner to proactively protect right-of-way against cut-ins, reporting up to 79.8% safety gains in simulation.
NOVA applies symbolic regression to 4.7 million NGSIM observations to identify a two-term car-following model (RMSE 1.376 m/s²) and a lane-change model (67.4% balanced accuracy) that outperform recent baselines and transfer zero-shot between sites.
citing papers explorer
-
MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
MPCFormer explicitly models multi-vehicle social interaction dynamics via physics-informed discrete state-space and Transformer-learned coefficients, yielding 0.86m ADE over 5s and 94.67% planning success with near-zero collisions in closed-loop tests.