Skip to content
/ EDDA Public

The EDDA - Evaluative Data-driven Design Application is a project for a simulation based controller verification and parameter estimation for autonomous driving vehicles.

License

Notifications You must be signed in to change notification settings

Iarbas/EDDA

Repository files navigation

EDDA

Introduction

The EDDA - Evaluative Data-driven Design Application is a project for a simulation based controller verification and parameter estimation for autonomous driving vehicles.

Overview

The Platform is subdivided into independent programs for:

- EDDAsim | Graphical output for the simulated environment (for the time being, this will be a 2D visualization). This includes:
  + A guide to design own test streets.
  + A graphical output of the simulated area.
  + Plots of the validation criteria.
  + A mechanism to save data for post-processing tools.
  + A configuration tool to manipulate the simulation.
- EDDApath | Planning a path, to give the controller the needed/optimal trajectory of the vehicle.
- EDDAmodel | The vehicle model as an ordinary differential equation, to calculate the movement of the vehicle in the simulation.
- EDDAcontrol | A set of controller to validate their quality. The set will consist of the actual controller as PID, H∞, LQR, MPC, non linear etc. and observer, filter and predictions.
- EDDAestimate | Offline and online parameter estimation algorithm. 
- EDDAvalidate | Validation of the results. 
- EDDAoptimal | An optimization tool to enhance the used controller with respect to the result of a simulated drive.

To keep it simple, the project tries to use only the tools coming with a fresh installation of Ubuntu 20.04 LTS. Every additional software will be mentioned in this README and will aslo be explained, how to install it.

Additional Software

REQUIRED

The project uses CMake for managing the build process:

$ sudo apt install cmake

For the ODE solver the project uses BOOST:

$ sudo apt install libboost-dev
$ sudo apt install libboost-all-dev

OPTIONAL

If you want to see the 2D simulation running, you will need to install SFML:

$ sudo apt-get install libsfml-dev

You can also build and run EDDA without visualization (just a few changes to the CMakeLists.txt and the configuration file), but it is recommended to have a look on what your program is doing. For more information about the library, you can read the Wikipedia aricle:

https://de.wikipedia.org/wiki/Simple_and_Fast_Multimedia_Library

Credits

Author: Andre Alexander Pieper.

Version 1.0

Date: 29.08.2019

License: LGPLv3

About

The EDDA - Evaluative Data-driven Design Application is a project for a simulation based controller verification and parameter estimation for autonomous driving vehicles.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published