Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
This repo contains the code for paper "Dense reinforcement learning for safety validation of autonomous vehicles"
| Date | Stars |
|---|---|
| 2026-07-31 | 380 |
| 2026-08-01 | 380 |
| 2026-08-02 | 380 |
| 2026-08-05 | 381 |
| 2026-08-06 | 381 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Dense reinforcement learning for safety validation of autonomous vehicles
<!-- ABOUT THE PROJECT -->
# Introduction of the Project
## About
This project contains the source code and data for the paper titled "Dense reinforcement learning for safety validation of autonomous vehicles".
## Code Structure
```
Dense-Deep-Reinforcement-Learning/
|__ conf: experiment configurations
|__ maps: maps for SUMO simulator
|__ checkpoints: model checkpoints for D2RL
|__ source_data: source data for constructing NDE and D2RL-based testing
|__ mtlsp: simulation platform
|__ envs: NDE and D2RL-based testing environments
|__ controller: vehicle controllers (e.g. IDM)
|__ data_analysis: refer to "Usage" section for detailed information
|__ main.py: main function for running NDE and D2RL-based testing
|__ utils.py: utility functions
|__ nadeinfoextractor.py: information extractor for logging experiment information
|__ requirements.txt: required packages
```
# Citation
Feng, S., Sun, H., Yan, X., Zhu, H., Zou, Z., Shen, S., and Liu H.X. (2023). Dense reinforcement learning for safety validation of autonomous vehicles. Nature 615, 620–627. https://doi.org/10.1038/s41586-023-05732-2
```
@article{Feng2023,
title = {Dense reinforcement learning for safety validation of autonomous vehicles},
volume = {615},
url = {https://www.nature.com/articles/s41586-023-05732-2},
doi = {https://doi.org/10.1038/s41586-023-05732-2},
number = {7953},
journal = {Nature},
author = {Shuo Feng, Haowei Sun, Xintao Yan, Haojie Zhu, Zhengxia Zou, Shengyin Shen, Henry X. Liu},
year = {2023},
note = {Publisher: Nature Publishing Group},
pages = {620-627},
}
```
# Installation
## Pre-requirements
- Python installation
- This repository is developed and tested under python 3.10.4 on Ubuntu 20.04 system.
- Download all required datasets
- The user should download the `data_analysis` folder from [here](https://dense-deep-reinforcement-learning.s3.us-east-2.amazonaws.com/data_analysis.zip). Then, the user should merge the downloaded `data_analysis` folder with the original `data_analysis` folder in the repo.
## Installation and configuration
### Clone this repository
```bash
git clone https://github.com/michigan-traffic-lab/Dense-Deep-Reinforcement-Learning.git
```
### Create a new virtual environment (Optional)
To ensure high flexibility, it is recommended to use a virtual environment when running this repository. To set up the virtual environment, please follow the commands provided below:
```bash
virtualenv venv
source venv/bin/activate
```
### Install all required packages
Due to the compatibility issue betwen gym and python setuptools, user should run follow commands to install a specific version of setuptools first:
```bash
pip install setuptools==65.5.0
```
To install the Python packages required for this repository, execute the command provided below:
```bash
pip install -r requirements.txt
```
### Install ipykernel (Jupyter Notebook) for data analysis
In order to use Jupyter notebooks for data analysis, it is necessary to have the ipykernel installed. To install it, users can execute the command provided below:
```bash
pip install ipykernel
```
<!-- USAGE EXAMPLES -->
### D2RL Training Environment
Please refer to README_D2RL_Train.md for detailed information about D2RL training environment installation and detailed usage.
> Please note that D2RL training process requries a different python environment compared to the python environment in README.md, with different python verison requirements and different python packages.
# Usage
The project includes a data_analysis directory, which stores data (raw & processed), code (data processing and data analysis), and outcomes (figures) related to data analysis. The project analyzes various performance metrics such as time-to-collision, post-encroachment-time, bumper-to-bumper distance, crash rate, crash type, and crash severity. This section will focus on the post-encroachment-time (PET) anaExcerpt of 10,537 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:720ce7354c6c8f15, name:reinforcement learning, desc:reinforcement learning