Reproduce “Agentic discovery of interpretable car-following laws from multi-source trajectory data” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiamei Xu, Jie Sun, Keshuang Tang, Jian Sun
AERM discovers a car-following law from error-guided structural hypotheses and evolutionary search, selecting candidates by reconstruction and physical criteria. The law combines spacing-dependent equilibrium speed with a response to leader-follower speed difference. In trajectory-specific reconstruction, it has the lowest spacing and speed R-M-S-E on three external datasets, including 0.5231 meters versus 0.6683 for Gipps on Waymo; this does not establish prediction on unseen trajectories.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.