Reproduce “Simulation Priors for Rapid Adaptation in Model-Based RL for Autonomous Racing” 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
Adrian Remonda, Jiajun Xi, Nicklas Hansen, Marcus Greiff, Thomas Lew, John Talbot, John Subosits, Xiaolong Wang
PIA trains a racing world model on heterogeneous simulation data, then adapts it with limited real-car driving while penalizing plans with high estimated uncertainty. In simulation, PIA-ODE reports average best lap time of 97.39 seconds versus 103.11 for FOWM across five task conditions. On one oval hardware study, success reaches 8/8 episodes after 44 minutes accumulated data.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.