Skip to content

This provides a sandbox simulator for training a self-driving car. This uses Unity for simulation and Python with Keras and Tensorflow for training. Recently updated to work on Python 3.4+ and Keras 2+

License

Notifications You must be signed in to change notification settings

Ottawa-Autonomous-Vehicle-Group/sdsandbox-pln

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SdSandbox-pln

Self Driving Car Sandbox - Parking Lot Nerds (PLN) Custom version

Summary

This is a custom version of the sdsandbox self driving simulator used by the Parking Lot Nerds team for testing and training.

Original sdsandbox repo can be found here: https://github.com/tawnkramer/sdsandbox

Original README.md file: https://github.com/Ottawa-Autonomous-Vehicle-Group/sdsandbox-pln/blob/sim-racer/README-orig.md

Updates

  1. Pull in changes from Maxime's sdsandbox repo: https://github.com/Maximellerbach/sdsandbox

  2. Update telemetry message to include:

    • linear acceleration
    • angular acceleration
    • wheel encoders

Installing Unity and self-driving simulator

  1. Download and install Unity Hub: https://unity3d.com/get-unity/download. Using the Unity Hub is recommended since it makes it easier to switch between Unity Editor versions.

  2. Run Unity Hub and follow instructions to generate a license file.

  3. Select "Installs" in left hand menu, and then select "Add" button at top right. Choose the version of Unity Editor to install. Latest at time of writing is Unity 2019.3.12.f1.

  4. After the Unity Editor has been install, select "Projects" in the left hand menu. Now select "Add" in top right, and browse to the "sdsim" folder in this repo. This is the Unity source code for the Self Driving Sandbox Simulator. If you are using a newer version of the Unity Editor, you will be asked to convert the project to this new version.

  5. After the conversion is complete, you will be brought into the Unity Editor window.

Happy Developing!!!

Credits (sdsandbox)

Tawn Kramer

About

This provides a sandbox simulator for training a self-driving car. This uses Unity for simulation and Python with Keras and Tensorflow for training. Recently updated to work on Python 3.4+ and Keras 2+

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C# 67.1%
  • HLSL 17.0%
  • ShaderLab 14.0%
  • Python 1.9%