vantuyenatoma / Digital-Race

Self Driving Car for Digital Race contest that is sponsored by FPT Corp.

We are the champions of digital race.

Paper:

Our paper published at ICSCA 2021, you can see details at https://doi.org/10.1145/3457784.3457827.

Demo:

Welcome! This is an open-source self-driving car aimed for rapid prototyping, deep learning, and robotics research. The system currently runs on a jetson tx2 module. Here are our goals:

Goals:

Research and develop a deep learning-driven self-driving car. The vehicle should be able to finish the race.

Role

To know the role, please read documentation.

The modules in this project.

1. Semantic Segmentation
2. Object Detection

For the full documentation of the development process, please visit my website: datvuthanh.github.io

Table of Content

• Introduction
• Try it out
• About ROS
• Semantic Segmentation

Introduction

Digital Race is a contest that is sponsored by FPT Corp. The task of the teams completing the race in the shortest time.

Try it out

To compile the project:

Requirements

1. Make sure that you have ROS installed on your computer. (I am using ROS Melodic)
2. Make sure you have all the dependencies installed.

Clone & Compile

1. Clone the repository. $ git clone https://github.com/datvuthanh/Digital-Race.git
2. $ cd Digital-Race
3. $ cp -r src/. ~/catkin_ws/src/.
4. $ cd ~/catkin_ws/
5. $ catkin_make
6. $ source devel/setup.bash

About ROS

This project uses ROS. For more information on ROS, nodes, topics and others please refer to the ROS README.

Semantic Segmentation

The cart understands its surrounding through semantic segmentation, which is a technique in computer that classifies each pixel in an image into different categories. The vehicle can also make decisions based on the segmentic segmentation results. The cart can change its speed based on the proximity to nearby obstacles.

We deployed the PSPNet architecture for segmentation. PSPNet is design to work well in realtime applications. For more information, please visit the paper. We collect dataset for training and the python code for training and inferencing are located in the segmentation directory.

VIDEO DEMO

Contact / Info

If you are interested in the detailed development process of this project, you can contact me at email address: [e-mail] or [e-mail]

Contributors:

Dat Vu (Leader) | Email | Github | Website

Hai Anh Tran | Email | Github

Tra Dinh | Email

Huy Phan | Email