Deep Reinforcment Learning Model for Self Driving in Amazon DeepRacer

Overview

This repository contains the code and resources for developing a deep learning model aimed at enabling autonomous driving capabilities. The primary goal is to create a model capable of driving based on image input. The trained model is then deployed onto an Amazon DeepRacer vehicle for real-world testing. The training environment utilized for this project is a simulation built in Unity. This project is developed under FIT DataLab at CTU

simulation.mov

Installation

1. Clone the repository to your local machine:

   git clone https://github.com/ItsOKayCZ/self-driving

2. Navigate to the project directory:

   cd self-driving

3. Create conda environment:

   conda env create --name yourenv --file=environment.yml

Demonstration at the Open Days at FIT CTU

Open Days FIT CTU	Demo
	fit-dod.mp4

Contributors

Tadeáš Báča

Oldřich Krupauer