For our group IT-project at the \textit{University of Applied Science Georg Simon OHM}, we built an autonomously driving remote control car. To accomplish this project, we used the framework DonkeyCar. Supervised learning is the primary component, but we provide related work and projects to compare this project with the state-of-the-art Reinforcement Learning. We explain how to connect the car with the microcontrollers and cables, to do a full reverse-engineering on the project. Every single component will be introduced in an appropriate complexity for non-electrical-engineers. The neural network is a VGG16-Network, which will be explained with all details and parts to understand the networks' architecture fully. We further provide information about how the car can be driving with the webserver. The Report gives a detailed explanation of the API and the transferred data.
- Projekt Homepage
- Report/Paper as PDF
- Presentation as PDF
- Pilot/Model
- YouTube Video first drive
Clone the repository, to get our Notebooks, Presentation and Project Report.
git clone https://github.com/Mavengence/Autonomous-RCCar-IT-Project-TH.OHM.git
- Of course you need git to get the source
- If you want to compile the report or the presentation by ur self u need a LaTex Compiler for your OS and maybe an IDE which makes things easier
- If you want to compile, train and play with our Code you need a python working environment. We used Jupyter Notebooks. The requiered packeges you can see in the Notebooks itself.
- For Prerequisites of our Project Hompage check the Website itself. There you willl find a detailed description of how to contribute changes to website.
cd/your_cloned_repo_location jupyter notebook
Just pull the repo, if you wanna change sth you can ask :)
- Tim Löhr - Coding, Report, Presentation - GitHub Mavengence
- Timo Bohnstedt - Coding, Report, Presentation - GitHub Bohniti
Pretty much the BSD license, just don't repackage it and call it your own please! Also if you do make some changes, feel free to make a pull request and help make things more awesome!
The authors would like to thank Prof. Dr. Florian Gallwitz for a really good supervising of our group. Without him we wouldn't be able to have accomplished our success.