KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#typescript #ai-agents #cli #claude-code #codex #developer-tools #ai #dsh-plugin #ai-agent #mcp #deepseek-harness #windows

autonomous-driving-simulator

★ 48 Open GitHub ↗

No description available.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

English | [中文](https://github.com/tum-autonomousdriving/.github/blob/main/profile/README_zh.md)
# Autonomous Driving Simulator

## Contents

- [Framework](#framework)
- [Instructions](#instructions)
- [Developers](#developers)

## Framework
![image](https://raw.githubusercontent.com/tum-autonomousdriving/autonomous-driving-simulator/master/figures/framework.png)

## Instructions

### 1. Workflow
![image](https://raw.githubusercontent.com/tum-autonomousdriving/autonomous-driving-simulator/master/figures/structure.png)

```
Step 1: Collect data from the Simulator
Step 2: Design neural network class (torch.nn.Module)
        Write data processing class (torch.utils.data.Dataset)
Step 3: Train the model and save weights (torch.utils.data.DataLoader, optimizer, loss function, …)
Step 4: Drive a car in the Simulator using the trained model

```

### 2. Train

Please open the "Simulator_train" application on your operating system.

Once opened, the Scene will initiate the recording of training data, which will be saved in the "Data/DatasetX/IMG" path and the "Data/DatasetX/VehicleData.txt" file.

In this `VehicleData.txt` the data is: the path of IMG, throttle(range: 0 to 1), brakes(range: 0 to 1), sterring(range: -1 to 1).

**Please note that you should not remove the "Index.txt" file from the "Data/" directory.**

### 3. Drive

Please open the "Simulator_test" application on your operating system.

and then run your: 
```
python drive.py
```
In this scene, 3 images are output to your net, and 3 elements of control are acepted.(throttle, brakes and steering)

### 4. Simulator Functions
<table>
    <tr>
        <th>Function</th>
        <th>Keyboard Defaults</th>
        <th>Gamepad Defaults</th>
    </tr>
    <tr>
        <td>Sterring</td>
        <td>A/D</td>
        <td>Left Stick - Left/Right</td>
    </tr>
    <tr>
        <td>Throttle</td>
        <td>W</td>
        <td>Left Stick - Up, Right Trigger</td>
    </tr>
    <tr>
        <td>Brakes</td>
        <td>S</td>
        <td>Left Stick - Down, Left Trigger</td>
    </tr>
    <tr>
        <td>Reset the Position</td>
        <td>F1</td>
        <td>-</td>
    </tr>
    <tr>
        <td>Switch mouse control</td>
        <td>Mouse right button</td>
        <td>-</td>
    </tr>
    <tr>
        <td>Quit</td>
        <td>ESC</td>
        <td>-</td>
    </tr>
    
</table>

### 5. Simulator Scenes Download Link:

Source code:

https://gitlab.lrz.de/tum-autonomousdriving/autonomous-driving

Build:

https://syncandshare.lrz.de/getlink/fiEQ6uWeJQFmosTX7HsxwG/

<!--
## Functions
* ### High-definition digital twins of real cities and roads
Simulate complex and changeable real road scenarios to improve the ability of autonomous driving algorithms to deal with such scenarios.
<table>
  <tr>
    <td vlign="center">
      <img src="figures/urban.png" alt="Pin popup window">
    </td>
    <td vlign="center">
      <img src="figures/urban2.png" alt="Popup window">
    </td>
  </tr>
</table>

* ### Simulation of light and weather changes
Simulate changes in light and weather to improve the robustness of autonomous driving algorithms.
<table>
  <tr>
    <td vlign="center">
      <img src="figures/light.png" alt="Pin popup window", width ="600">
    </td>
    <td vlign="center">
      <img src="figures/weather.png" alt="Popup window", width ="600">
    </td>
  </tr>
</table>

* ### Automatic Data Labeling
The automatic data annotation function can automatically generate labels for training 2D/3D object detection and semantic/instance segmentation algorithms.
<table>
  <tr>
    <td vlign="center">
      <img src="figures/lable.png" alt="Pin popup window">
    </td>
    <td vlign="center">
      <img src="figures/segmentation.png" alt="Popup window">
    </td>
  </tr>
</table>

* ### Industrial LiDAR Simulation
Integrated Unity industrial-grade lidar simulation for training and testing 3D object detection, distance estimation and SLAM algorithms.
<table>
  <tr>
    <td vlign="center">
      <img src="figures/sim1.png" alt="Pin popup window", width ="600">
    </td>
    <td vlign="center">
      <img src="figures/Sim2.jpg" alt="Popup window", width ="600">
    </td>
  </tr>
</table>
-->

## Useful Vision Algorithms

Pretrained Weights:

https://syncandshare.lrz.de/getlink/fi6mGLJoqSu3GEBxW7aWNT/weights

### 1. Object Detection

1.1 YOLOv5

1.2 YOLOv8

### 2. Semantic Segmentation

2.1 YOLOv5_sg

## Developers

### 1. Affiliation

<a href="https://www.ce.cit.tum.de/air/home/">Chair of Robotics, Artificial Intelligence and Real-time Systems</a> · <a href="https://www.tum.de/">Technical University of Munich</a>

### 2. Supervisor

<a href="https://www.ce.cit.tum.de/air/people/prof-dr-ing-habil-alois-knoll/">Prof. Dr.-Ing. habil. Alois Christian Knoll</a>

### 3. Corresponding Programmer

<a href="https://www.ce.cit.tum.de/air/people/liguo-zhou/">Zhou Liguo</a>

### 4. Programmers

<table>
    <tr>
        <td>Tang Yuhang</td>
        <td>Guo Yujie</td>
        <td>Chen Zehao</td>
        <td>Feng Tinghan</td>
        <td>Tang Shaolong</td>
    </tr>
    <tr>
        <td>Ning Junbing</td>
        <td>Jasmina Stratorska</td>
        <td>Maram Bayadsi</td>
        <td></td>
        <td></td>
    </tr>
</table>

### 5. Former Programmers

<table>
    <tr>
        <td>Gao Yichao</td>
        <td>Yu Zhou</td>
        <td>Michael Sodamin</td>
        <td>Song Yinglei</td>
        <td>Liu Hongshen</td>
    </tr>
    <tr>
        <td>Liu Lian</td>
        <td>Liu Hao</td>
        <td>Xie Tongtong</td>
        <td>Liu Zhuo</td>
        <td>Yang Xiaohan</td>
    </tr>
    
</table>

autonomous-driving-simulator

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README