Development and evaluation of a Deep Reinforcement Learning agent in CARLA simulator
Autonomous driving is a rapidly advancing field that requires robust and efficient training environments for developing and evaluating algorithms. This article introduces CARLA-SB3-RL-Training-Environment, a comprehensive repository that offers an out-of-the-box training and evaluation environment for deep reinforcement learning (DRL) in the CARLA simulator. Built upon the Stable Baselines 3 library, this environment allows researchers and developers to conduct multiple experiments with customizable reward functions, state representations, and reinforcement learning algorithms.