teamspatzenhirn / Simulator

Carolo-Cup Autonomous Driving Simulator

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SpatzSim

Screenshot of the simulator: parking section

This is the Carolo-Cup simulator of Team Spatzenhirn.

The following aspects are simulated:

  • Vehicle motion using Pacejka's tire model
  • Primary high-resolution color camera, bayer encoded
  • Depth camera: Currently configured for the resolution of two side-by-side Intel D435
  • Laser distance sensors on the right side of the vehicle
  • Static obstacles
  • Dynamic obstacles: Motion triggered by arriving vehicle
  • Many static objects, such as road signs, floor markings, giraffe, parking sections, startbox, etc
  • Lane markings: Straight section, curved sections and crossings with arbitrary number of entries and arbitrary angles, placed in interactive editor

IMU, wheel encoder and steering angle sensors are currently not simulated. The entire vehicle state (with perfect odometry and position data) is available as an output.

A rule-checking module is included, which we use for automated testing.

ROS nodes adapting the custom communication interface to ROS can be found in the teamspatzenhirn_simulator_interface package.

We know that documentation on how to use the simulator is currently rather sparse. If you are trying to use the simulator and need assistance, feel free to open an issue. If you have the simulator up and running, open the keyboard shortcut help page from the menu under Show->Help!

Getting the source

This repo uses git LFS for storage of the 3D model. Run git lfs pull after cloning to download all models.

Build

SpatzSim is build using CMake (version > 3.0) and a compiler supporting C++17. Most dependencies are included directly in the git repository. Additionally, SpatzSim requires GLFW and GLEW to be installed as system libraries. On Ubuntu 18.04, 20.04 or 22.04 these can be installed with:

sudo apt install libglfw3-dev libglew-dev

To build the simulator execute the following in the project directory:

mkdir build
cd build
cmake ..
make

To build and run the simulator the "run" target is used:

make run

Necessary shared libraries:

The following dependencies must be installed as shared libraries on the system:

GLEW

Used to load OpenGL extensions.

http://glew.sourceforge.net/

GLFW

Provides an OpenGL context, a window and simple input handling.

https://www.glfw.org/ https://github.com/glfw/glfw

Libraries included in repository:

The following additional libraries are used internally and are included in the repository.

GLM

GLSL like matrix and vector math library for C++.

https://github.com/g-truc/glm https://glm.g-truc.net/0.9.9/index.html

ImGui

Immediate mode OpenGL gui library.

https://github.com/ocornut/imgui

OBJ-Loader

Loads 3D model in the .obj (and corresponding .mtl) file format.

https://github.com/Bly7/OBJ-Loader

Json11

A json library for modern C++. Used to save and load configurations.

https://github.com/nlohmann/json

Contributing

Contributions are welcomed and encouraged. We would love to hear from you if this simulator is helpful to you and your Carolo-Cup team. If you would like to contribute and don't know where to start, take a look at the issues labeled good first issue!

About

Carolo-Cup Autonomous Driving Simulator

License:MIT License


Languages

Language:C++ 92.7%Language:GLSL 2.3%Language:CMake 1.9%Language:C 1.7%Language:Python 1.4%Language:Shell 0.0%