Reproduce “Learning from Shared-Control Overrides: Context-Driven Acceleration Profile Prediction for Personalized Overtaking” 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
Ruizheng Xu, Lounis Adouane, Javier Ibañez-Guzmán, Clément Zinoune
CoP-ACC learns driver-preferred acceleration profiles from accelerator overrides during overtakes using clustered traces, context features, a classifier, and residual decoder. It reconstructs overrides with 0.0306 m/s² mean absolute error and 0.804 Pearson correlation, while predictions on matched forced-ACC contexts shift toward human behavior. This one-driver offline study uses 140 training events; closed-loop validation and reduced interventions remain untested.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.