This repository is dedicated to the advancement of autonomous vehicle safety through the application of dense reinforcement learning. Leveraging the power of the highway-env environment, our goal is to create a robust validation framework that ensures the safety of autonomous vehicles under a wide range of driving scenarios.
This repository focuses on improving the safety of autonomous vehicles through the application of dense deep reinforcement learning. By harnessing the capabilities of the highway-env
environment, our project aims to establish a robust safety validation framework to assess the performance of autonomous vehicles in diverse driving scenarios.
- Clone the repository:
git clone https://github.com/nagasriramnani/Safety-autonomous-driving-Dense-Deep-reinforcement-learning-Highway-env.git
- 2.Install dependencies:
-
conda create --name highwayenv python==3.9
pip install tensorflow
pip install gymnasium
pip install gym
pip install pygame
pip install pytorch
pip install stable-baselines3[extra]
pip install stable-baselines3
- activate conda environmet using conda activate your-env-name 4.To check the environment how it works .py files are given in project directory
- After completing the basic setup to test the files, which are A2c_test.ipynb ,epslion_greeady.ipynb , PPo_test.ipynb.
- Dense Reinforcement Learning Integration: Implement state-of-the-art reinforcement learning algorithms tailored for autonomous vehicle safety validation.
- Highway-env Integration: Use the
highway-env
environment for simulating a plethora of traffic situations. - Visualization Tools: Comprehensive tools to visualize and understand the agent's decision-making processes.
- Evaluation Metrics: Metrics to assess safety, efficiency, and overall performance of the autonomous agents.
We welcome contributions from the community! If you'd like to contribute, please follow these steps:
- Fork the repository.
- Make your changes.
- Submit a pull request with a detailed description of your changes.
Special thanks to Leurent, Edouard for creating the highway-env
environment which serves as the backbone for our simulations.