project
2D Autonomous Driving
A reinforcement-learning comparison of PPO, A2C, and QRDQN agents trained to navigate simulated highway and merge environments.
2D Autonomous Driving compares PPO, A2C, and QRDQN reinforcement-learning agents trained and evaluated in simulated highway and merge environments.
The repository includes configurable training and evaluation scripts, TensorBoard monitoring, Bayesian hyperparameter optimization, saved model artifacts, best-parameter configurations, and exported average reward summaries.
Draft status
Full write-up in progress. This page currently summarizes the repository-backed project record.