This repository contains scripts and models necessary to reproduce our results for the qualification round of the RoboCup Autonomous Robot Manipulation Challenge 2023.
Before proceeding, ensure that you have the following dependencies installed:
- MATLAB R2023a
- Aerospace Toolbox
- Computer Vision Toolbox
- Deep Learning Toolbox
- Image Processing Toolbox
- ROS Toolbox
- Statistics and Machine Learning Toolbox
- Clone the Repository
First, clone this repository to your local machine using the following command:
git clone https://github.com/fabioscap/arm_2023
- Set Up the Simulated Environment
After cloning the repository, run the simulated environment. Ensure that the environment is set up correctly and all the necessary components are properly initialized. In our examinations, we've consistently utilized the gpu mode (command: ./run_bash nvidia), thus maintaining a "real time factor" of 0.9 or higher.
- Connect to the Environment
Open the main.m file and replace the placeholder IP with your own. This will connect your local setup to the simulated environment.
- Execute the Main File
Finally, execute the main.m file to solve the challenge.
-
YOLO Dataset (The link provided is private. If you wish to view it, kindly request access.)
If you have any questions, please feel free to contact us.
- Dennis Rotondi rotondi.1834864@studenti.uniroma1.it
- Emanuele Giacomini giacomini@diag.uniroma1.it
- Fabio Scaparro scaparro.1834913@studenti.uniroma1.it
- Leonardo Brizi brizi@diag.uniroma1.it
- Marco Montagna montagna.1882418@studenti.uniroma1.it