Case 1 HC2RL
Video unavailable. Open the MP4 file.
HOCBF-Informed Constrained Reinforcement Learning with Adversarial Training for Robust Autonomous Driving
Six navigation tasks. Two cases per task. A comparison of longitudinal driving behavior
with and without the HOCBF-based safety filter.
Left turn to target lane 1
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Left turn to target lane 2
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Go straight to target lane 1
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Go straight to target lane 2
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Right turn to target lane 1
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Right turn to target lane 2
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Two left-turn navigation tasks. Two cases per task.
Evaluated with HC2RL-MC under truncated Gaussian noise perturbations.
Left turn to target lane 1
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Left turn to target lane 2
Video unavailable. Open the MP4 file.
Video unavailable. Open the MP4 file.
Main symbols and their definitions