Simulation-based reinforcement learning for autonomous driving

Qualitative comparison of autonomy

Below graphs show the scale between blue (full autonomy in all experiments) and red (human assistance in all experiments).