开源软件名称（OpenSource Name）：

开源软件地址(OpenSource Url)：

开源编程语言(OpenSource Language)：

开源软件介绍(OpenSource Introduction)：

Carla-Simulink Interface

• Carla-Simulink Interface
  • General Note
  • Requirements
    • Hardware
    • Software
  • Getting Started
    • Using the Python Interface
  • Writing your first MATLAB script for Carla using the Python interface
  • Diving deeper: Running the Examples
    • Adaptive Cruise Control
      • Python Interface: ACC
    • Getting Lidar data
      • Python Interface: Lidar
    • Getting sensors data in Simulink
      • Python Interface: Multiple Sensors
    • Controlling the Actor
      • Python Interface: Keyboard Control
  • Appendix
    • Sensor binding for Python

Carla is an open-source software targeted towards autonomous driving research. It provides a powerful interface to support autonomous driving systems' development, training, and validation. Simulink needs no introduction. It is used widely, especially by engineers, and provides a lot of functionality to assist rapid product development. The primary motivation for the project is to link Simulink to Carla and showcase some examples of how to use the interface.

General Note

This repository shows how to integrate MATLAB/Simulink with Carla using Python or ROS. This README file demonstrates the Python version only. For a complete dump of instructions, refer to the other version. The python version is explored by me while the ROS equivalent is explored by my colleague Ahmethan Başak. The complete version is messy due to the merger of 2 different approaches, but I hope that it is still useful.

For demonstration, different sensors are spawned inside Carla while their data is processed in MATLAB/Simulink in real time.

Finally, due to the paid nature of the MATLAB license, I can no longer update the package. The underlying concepts and techniques can be utilized by the users for their needs. Queries are always welcomed regardless.

Requirements

To run this interface, the following requirements should be met.

Hardware

As Carla uses Unreal Engine, the recommended requirements are basically the same, which are as follows.

• 64-bit Windows/Linux/MacOS
• 8-GB RAM or more
• Quad core Intel or AMD processor, 2.5 GHz or faster
• NVIDIA GeForce 470 GTX or AMD Radeon 6870 HD series card or higher
• 10GB of hard drive space for the simulator setup

Software

To run the interface, the following softwares are required.

• MATLAB
• Simulink
  • Model Predictive Control Toolbox
  • Simulink Control Design
• Carla
• Python
• ROS (only if you want to run the ROS of the interface)
• Unreal Engine (only if you plan to compile everything from scratch)

Getting Started

Using the Python Interface

Start by installing MATLAB and Simulink. The installation process is straightforward and won't be discussed here. Go to the Python website and install the 64-bit version (recommended). Make sure that while installing, you select pip as well. The next step is to install Carla. Go to the website and select the latest release as shown.

You will be redirected to a new page. Click on GET CARLA 0.9.13 or whichever version is the latest. You will be directed to the releases page. One approach is to compile everything yourself. Here, we will only download the pre-compiled version for the sake of simplicity. Under the heading Release, select Windows or Linux depending on whichever operating system we are using. Unfortunately, for MacOS, you will need to compile everything from scratch. The tutorial is available on the Carla website for Linux and Windows.

As the developers of Carla mainly use Linux themselves, Linux builds are the most up to date. If on Windows, select the latest compiled version available.

Since Carla 0.9.12, the wheel file is also available in the same directory. Use pip directly instead

After downloading, extract the files. Go to the extracted folder and to PythonAPI/carla/dist. In that folder, you will find a .egg file. Now open the terminal/cmd in the same folder and type the following command:

pip install setuptools

You may need sudo/admin rights for these operations. Once setuptools are installed, write the following command:

easy_install [name_of_egg_file].egg

This is all that is required for the setup. You can now download the examples and run them without a problem