AARE-SAC: Dataset for Robust Autonomous Driving Experiments

This dataset contains all experimental data used in the paper “AARE-SAC: Adversarially Adaptive Reward-Enhanced Soft Actor-Critic for Robust Autonomous Driving”.It includes reward curves for lane-change, left-turn, and right-turn tasks, ablation study results, stability metrics, collision statistics, baseline comparisons, and the experimental configuration file (config.yaml).All values correspond exactly to those used for generating the tables and figures in the manuscript.The dataset is released under an open licence to support reproducibility and further research on robust reinforcement learning and autonomous driving.