Reproduce “DRNet: A Decision-Making Method for Autonomous Lane Changingwith Deep Reinforcement Learning” 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
Kunpeng Xu, Lifei Chen, Shengrui Wang
DRNet combines a Q-network with an Action-Subspace safety filter and replay-buffer initialization and prioritization for highway decisions. It represents traffic history with occupancy grids and screens dangerous commands before action selection. In three-lane simulations, its efficiency score was 1.13 times the ablated version, and learning reached rule-based reward about twice as fast as DDQN; these results do not guarantee safety.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.