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.
Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator
| Date | Stars |
|---|---|
| 2026-07-31 | 265 |
| 2026-08-01 | 265 |
| 2026-08-02 | 265 |
| 2026-08-06 | 265 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
rl-CARLA
===============
The basic idea is using Raw Image as state spaces to train DDPG Agent. The network architecture is quite simple, if you want to know more, you can check [here](https://github.com/zhangfuyang/rl_CARLA/blob/master/src/network/ddpg_network.py). In order to evaluate the performance of the RL method, we first used supervised learning to train a network as baseline. Then we investigate the performance of RL methods (DDPG), both with and without pretraining.
## Demo
### imitation
<img src="./assets/imitation1.gif" width="200"/> <img src="./assets/imitation2.gif" width="200"/>
### DDPG
  
## Result
<img src="./assets/result1.png" width="300"/> <img src="./assets/result2.png" width="300"/>
              Town1(Train)                              Town2(Test)
## Installation
### Clone the git repository
$ git clone https://github.com/zhangfuyang/DDPG-CARLA.git
$ cd DDPG-CARLA & pip -r requirements.txt
(Use $DDPG_DIR as the root directory of the source)
### Install Carla Simulator
1. Download the Carla
you can just download the compiled version from [here](https://github.com/carla-simulator/carla/releases/tag/0.8.2) . Make sure the version of the simulator is 0.8.2(stable), I'm not sure if other development or old stable versions are compatible.
2. extract to the directory you want
$ tar -xvf CARLA_0.8.2.tar.gz $CARLA_DIR
## Training
1. First start the Carla server
cd $CARLA_DIR
./CarlaUE4.sh -carla-server -benchmark -fps=10
2. Run ddpg_main.py
cd $DDPG_DIR
python ddpg_main.py
## Test the model
1. Start the Carla server
2. python test_ddpg.py -model_path='models/'
## Some Detail of implementation
1. Use pretrain Network https://github.com/carla-simulator/imitation-learning to extract feature. Then connected with actor and critic net.
Excerpt of 2,160 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a7651406e8ddd199, llm:description: 'Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator'
matched fp:a7651406e8ddd199, llm:description: 'Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator'
matched fp:a7651406e8ddd199, llm:description: 'Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator'