Safe Reinforcement Learning for Autonomous Vehicles through Parallel Constrained Policy Optimization | IEEE Conference Publication | IEEE Xplore