Reproduce “Prescribed Performance Guided Safe Reinforcement Learning for Roundabout Navigation of Autonomous Vehicles” 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
Wenhan Li, Yangrui Zhang, Hang Zhou, Peiran Liu, Yiding Ji
The system combines RL with MPC for roundabout driving by having the learner generate a reference trajectory that the controller tracks within a tightening error envelope, while a safety filter corrects controls when needed. In MetaDrive simulation, the full method reaches 96.25% success versus 90.00% for its ablation, while comfort effects differ by measure and results do not establish real-world performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.